{
    "0010/astro-ph0010242_arXiv.txt": {
        "abstract": "The Alpha Magnetic Spectrometer in a precursor version (AMS-01), was flown in June 1998 on a 51.6\\mydeg~orbit and at altitudes ranging between 320 and 390 km, on board of the space shuttle Discovery (flight STS-91). AMS-01 included an Aerogel Threshold \\cher counter (ATC) to separate $\\bar{p}$ from $e^{-}$ and $e^{+}$ from $p$, for momenta below 3.5 \\GeVc. This paper presents a description of the ATC counter and reports on its performances during the flight STS-91. ",
        "introduction": "The first phase of the AMS experiment (AMS-01) was achieved on board of the space shuttle Discovery, during 10 days in June 1998. The main objective was to test the spectrometer's instrumentation in orbit, in preparation for the second phase that will take place on board of the International Space Station (ISS) for 3 to 5 years. During the shuttle flight, 100 million events were recorded, allowing the fluxes of several particle species ($e^{\\pm}$, $p^{\\pm}$, He) to be measured~\\cite{amsres}.\\\\ The AMS-01 detector included a permanent magnet, a Time-of-Flight scintillation counter (TOF), a silicon tracker (TRK), anti-coincidence scintillation counters (ACC) and an Aerogel Threshold \\cher counter (ATC). A detailed description of the AMS spectrometer may be found in \\cite{amsrep}.\\\\ This paper describes the ATC counter and its performances during the flight on board of Discovery. \\subsection{Role of ATC in AMS-01.} One of the main purposes of the AMS Shuttle flight was to measure cosmic antiproton spectrum for momenta below 3.5~\\GeVc~(the ATC momentum threshold). Antiproton spectrum measurement, as well as positron sprectrum, can be achieved by using the ATC counter :\\\\ $\\bullet$ {\\bf antiprotons :} The major background component to the $\\bar{p}$ sample is expected to come from misidentified electrons. Using the measured electron flux~\\cite{amsres} and the previously measured $\\bar{p}$ flux~\\cite{besspbar}, the signal to background ratio is estimated to be : ${\\bar{p}}/e^{-} \\sim \\!10^{-3}\\!-\\!10^{-2}$ for the considered P range.\\\\ While TOF counters~\\cite{kn:choumilo} (${\\Delta \\beta }/\\beta \\simeq 3.3 \\, \\%$) allow the separation of $\\bar{p}$ and $e^{-}$ below 1-1.5 \\GeVc~(fig.~\\ref{tofatc}), ATC extends this discrimination range to 3.5 \\GeVc. Therefore, ${\\bar{p}}/{e^{-}}$ separation can profit from ATC redundancy up to 1-1.5 \\GeVc~and can rely on the ATC up to 3.5 \\GeVc.\\\\ $\\bullet$ {\\bf positrons : }Positrons were also an important issue for AMS-01. They had to be discriminated from a much larger proton flux, with a typical ratio : ${p}/e^{+} \\sim 10^{3}$. Although the ATC design was not optimized for this selection, ${e^{+}/p}$ discrimination could be achieved by using appropriate ATC cuts~\\cite{amsres}, as shown in sec.~\\ref{posi}. \\subsection{Principle of the ATC.} The principle of the ATC counter, used in AMS-01, is based on the \\cher effect to separate $\\bar{p}$ from $e^{-}$ at low energy. Basic relations are recalled here for the reader's convenience. The number of photons created by the \\cher effect, in a material of refractive index $n$, is proportional to : \\begin{equation} N_{\\gamma} \\propto L\\times Z^{2} \\times \\sin^{2} \\theta = L\\times Z^{2} \\times (1-\\frac{1}{n^{2} \\beta^{2}}) \\label{cherequa} \\end{equation} \\noindent where $L$ is the path length in the material, $\\theta$ is the \\cher angle, $Z$ the charge of the incoming particle and $\\beta$ the particle velocity.\\\\ This leads to the following threshold values (in beta or momentum) : \\begin{equation} \\beta_{thres}={1}/n~;~\\mathrm{P}_{thres}=\\frac{mc}{\\sqrt{n^{2}-1}} \\end{equation} \\noindent where $m$ is the mass of the particle at rest.\\\\ The aerogel refractive index ($n=1.035\\pm 0.001$) was chosen \\cite{gougas} to provide a high threshold and a sufficient number of photo-electrons (p.e). The corresponding thresholds, for several particle species, are given in the following table. \\begin{table}[htb] \\begin{center} \\begin{tabular}{|c|c|c|c|c|c|c|} \\hline Particle   & $e^{\\pm}$ & $\\pi^{\\pm}$ & p ($\\bar{p}$) & He ($\\bar{He}$) \\\\ \\hline $\\mathrm{P}_{thres}$& 1.91 \\MeVc & 0.52 \\GeVc & 3.51 \\GeVc & 14.0 \\GeVc \\\\ \\hline \\end{tabular} \\end{center} \\end{table} \\hspace{-10mm} \\par \\noindent Electrons and positrons in the 0.5-3.5 \\GeVc~range are far above their threshold, thus giving a full amplitude signal in ATC. In principle, $p$ and $\\bar{p}$ of momentum less than 3.5 \\GeVc, are not expected to give any \\cher signal, thus leading to ${\\bar{p}}/e^{-}$ separation. In the following, this value of 3.5 \\GeVc~will be referred to as the \\cher threshold, i.e. the ATC momentum threshold for antiproton selection. ",
        "conclusions": "The general behaviour of the AMS apparatus was satisfactory during its first test flight on board the space shuttle Discovery \\cite{amsrep}. The ATC counter allowed ${\\bar{p}}/e^{-}$ separation with a rejection of 330 and an efficiency up to $\\sim 48 \\%$, extending the ${\\bar{p}}/e^{-}$ separation range up to 3.5 \\GeVc. As a secondary result, it has been used, with appropriate cuts, to separate $e^{+}$ from protons with a rejection up to 260 and an efficiency of $\\epsilon_{e^{+}} \\simeq$ 41 \\%. \\newpage \\listoffigures \\newpage"
    },
    "0010/astro-ph0010597_arXiv.txt": {
        "abstract": "We have performed high-resolution synthesis observations of the $^{12}$CO (J=1-0) line emission from the radio lobe edge-on spiral NGC 3079 using a 7-element mm-wave interferometer at Nobeyama, which consisted of the 45-m telescope and 6-element Array. The molecular nuclear disk (NMD) of 750 pc radius is found to be inclined by 20\\deg\\ from the optical disk, and the NMD has spiral arms. An ultra-high-density molecular core (UHC) was found at the nucleus. The gaseous mass of the UHC within 125 pc radius is as large as $\\sim 3\\times 10^8\\Msun$, an order of magnitude more massive than that in the same area of the Galactic Center, and the mean density is as high as $\\sim 3\\times10^3 {\\rm ~H_2~cm^{-3}}$. A position-velocity diagram along the major axis indicates that the rotation curve starts at a finite velocity exceeding 300 \\kms\\ already from the nucleus. The surface mass density in the central region is estimated to be as high as $\\sim 10^5 \\Msun~{\\rm pc}^{-2}$, producing a very deep gravitational potential. We argue that the very large differential rotation in such a deep potential will keep the UHC gravitationally stable from current star formation.  Subject headings: ISM: molecules --- galaxies: individual (NGC 3079) --- galaxies: ISM --- galaxies: kinematics and dynamics --- galaxies: nuclei  --- galaxies: structure --- radio lines: ISM ",
        "introduction": "NGC3079 is an edge-on galaxy known for its pronounced radio, \\ha\\ and X-ray lobes emerging from the nucleus (Hummel et al 1983; Duric et al 1983; Lord et al 1986; Veilleux et al 1994; Pietsch et al 1999). The nuclear outflow may originate with accretion of dense gas onto a compact core (Irwin and Seaquist 1988), and the nucleus exhibits LINER and Seyfert 2 nuclear activity (Lord et al 1986). NGC 3079 has been studied in various radio wavebands in continuum, HI and CO lines as well as maser lines (Irwin and Seaquist 1991;   Ford et al. 1986; Duric et al 1983; Duric and Seaquist 1988; Irwin et al 1987; Irwin et al 1988; Young et al. 1988; Sofue and Irwin 1992; Irwin and Sofue 1992; Sawada-Sato et al. 2000). The galaxy's distance is 15.6 Mpc ($H_0=75$ \\kms/Mpc), and the major axis position angle of the optical disk is $167\\Deg$ and the inclination angle 85\\deg.  Earlier CO observations with the Nobeyama mm Array (NMA) inferred the existence of a dense nuclear molecular disk, rotating at a high velocity within the central 10$''$ (750 pc) (Sofue and Irwin 1992; Irwin and Sofue 1992). NGC3079 is a galaxy whose unusual nuclear activity can be probed by observing the affected interstellar medium, and can be compared with the Milky Way for its similar edge-on orientation. Knowledge of the kinematics, morphology and density of the interstellar medium in the immediate vicinity of nucleus is particularly important, since the ISM can both fuel the nucleus and trigger activity. The nuclear disk would also play a role in collimating outflows. For this, higher resolution CO observations have been required to determine how the morphology and kinematics of the nuclear molecular disk are related to the nuclear activity.  In this paper we report the result of high-resolution CO-line observations of NGC 3079 using a 7-element millimeter wave interferometer at Nobeyama by combining the NMA (Nobeyama mm Array) and 45-m telescope, whose code name was \"RAINBOW\". ",
        "conclusions": ""
    },
    "0010/astro-ph0010418_arXiv.txt": {
        "abstract": "We discuss the quantification of the local galaxy population and the impact of the ``New Era of Wide-Field Astronomy'' on this field, and, in particular, systematic errors in the measurement of the Luminosity Function. New results from the 2dFGRS are shown in which some of these selection effects have been removed. We introduce an INT-WFS project which will further reduce the selection biases. We show that there is a correlation between the surface brightness and the luminosity of galaxies and that new technologies are having a big impact on this field. Finally selection criteria from different surveys are modelled and it is shown that some of the major selection effects are surface brightness selection effects. ",
        "introduction": "Galaxy populations were first studied by Hubble (1926), who developed the familiar Tuning Fork diagram of Ellipticals, Spirals and Barred Spirals. Most bright galaxies can be morphologically classified by their Hubble type. However, many types of galaxy have been found that don't fit the Tuning Fork. These occur both at low redshift and at high redshift where the galaxies can be intrinsically different due to evolution. Some of these galaxies are shown in Fig. 1. The Tuning Fork can be extended to include these new types of galaxy as shown in Fig. 2.  \\begin{figure}[h] \\psfig{file=highz1.ps,height=25.0mm,width=25.0mm} \\vspace{-25.0mm} \\hspace{30.0mm} \\psfig{file=highz3.ps,height=25.0mm,width=25.0mm} \\vspace{-25.0mm} \\hspace{5.0mm} \\psfig{file=lowz1.ps,height=25.0mm,width=25.0mm} \\vspace{-25.0mm} \\hspace{5.0mm} \\psfig{file=lsbg2.ps,height=25.0mm,width=25.0mm} \\vspace{50.0mm} \\caption{Some peculiar galaxies. The two on the left hand side are high redshift galaxies in the Hubble Deep Field North. The two on the right are low redshift galaxies from the Millennium Galaxy Catalogue. The farthest right galaxy is a Low Surface Brightness Galaxy.} \\end{figure}  \\begin{figure} \\centerline{\\psfig{file=htf.ps,height=55.0mm,width=130.0mm}} \\caption{The various different classes of galaxies that astronomers have named over the years. The classes are positioned roughly in order of magnitude from bright on the left to faint on the right and in surface brightness from high surface brightness at the top to low surface brightness at the bottom. The Hubble Tuning Fork types make up most of the bright galaxies.} \\vspace{-5mm} \\end{figure}  However, morphological classification only tells us what type of galaxies there are. It does not tell us what proportion are of each type or whether this varies over time. Galaxy formation and evolution is a relatively new field and so far very little is known about what produced the galaxy population we see today. To understand this process it is first necessary to have some quantitative information about the local galaxy population. The luminosity is an easy to measure quantity which can be used to classify galaxies. The Luminosity Function (LF) (Peebles \\& Hauser 1974) measures the space density of galaxies as a function of luminosity. This can be convolved with different evolutionary models and compared to number - magnitude counts (Driver et al. 1994). However, recent surveys have produced a large range in the measured LF, Fig. 3, with the variation between the surveys much greater than random errors. The systematic errors causing this wide variation must be understood before any progress is going to be made.  \\begin{figure} \\centerline{\\psfig{file=lfs2.ps,height=55.0mm,width=130.0mm}} \\caption{This is a plot of eight recent Galaxy Luminosity Functions. The variation at the M$^*$ ($M\\approx-19.5$) point is a factor of 2 and at $M=-14.5$ the variation is a factor of 10. The $M^*$ point is where the statistics are best. The variation would be expected to be of the order 10\\% if it where due to random errors. For more details see Cross et al. (2000).} \\vspace{-5mm} \\end{figure}  Disney (1976) pointed out that surface brightness selection effects are important to take into account when measuring the luminosity function. Phillipps \\& Disney (1986) and Phillipps, Davies \\& Disney (1990) went on to calculate the effects of surface brightness. These calculations take into account light lost below the isophote, Malmquist bias and surface brightness dimming due to cosmological expansion.  Here, we will discuss how we have used the ``Two Degree Field Galaxy Redshift Survey'' (2dFGRS) to produce a Bivariate Brightness Distribution (BBD), which corrects for these effects. Then we will describe some of the results and some of the selection effects that we have not corrected for. We will then briefly describe the Millennium Galaxy Catalogue (MGC), one of the projects in the Isaac Newton Telescope Wide Field Survey (INT-WFS). The MGC is a deep, imaging survey designed to remove some of the remaining selection effects. The selection effects are then modelled by convolving the BBD with different functions based on the visibility theory of Phillipps, Davies \\& Disney 1990) adopting parameters appropriate for the ESO Slice Project (ESP) (Zucca et al. 1997), the Las Campanas Redshift Survey (LCRS) (Lin et al. 1996) and the Second Southern Sky Redshift Survey (SSRS2) (Marzke et al. 1998). ",
        "conclusions": "To understand the effects of evolution, it is essential to have a good quantitative knowledge of the local galaxy population.  It is apparent from Fig. 4 that our knowledge of the local galaxy population is limited by systematic errors rather than random errors.  A major systematic error is caused by surface brightness selection effects. These can be removed using a Bivariate Brightness Distribution.  The new technologies in this ``New Era of Wide Field Astronomy'' allow us to detect larger numbers of galaxies with higher precision. These new technologies are the multi-fibre spectrometers such as 2dF and the large format CCDs such as those in the INT Wide Field Camera. These new surveys allow us to properly tackle the selection effects that have dogged our surveys and therefore will allow us to finally pin down the space and luminosity densities of galaxies. \\vspace{-5mm}"
    },
    "0010/astro-ph0010132_arXiv.txt": {
        "abstract": "This paper is the first one in a series devoted to the study of the dynamics of the atmospheres of long-period variable (LPV) stars. Results from a two-month-long monitoring of the Mira variables RT\\,Cyg and X\\,Oph around maximum light with the ELODIE spectrograph at the Haute-Provence Observatory are presented. The monitoring covers phases 0.80 to 1.16 for RT\\,Cyg and phases 0.83 to 1.04 for X\\,Oph. The cross-correlation profile of the spectrum of RT\\,Cyg with a K0\\,III mask confirms that the absorption lines of RT\\,Cyg in the optical domain appear double around maximum light. No line doubling was found in the optical spectrum of X\\,Oph around maximum light, indicating that this feature is not common to all LPVs.  This paper also presents the application to RT\\,Cyg of a new tomographic\\footnote{The word {\\it tomography} is used here in its etymological sense ({\\it `display cuts'}), which differs somewhat from the broader sense in use  within the astronomical community (reconstruction of a structure using projections taken under different angles).} technique deriving the velocity field across the atmosphere by cross-correlating the optical spectrum with numerical masks constructed from synthetic spectra and probing layers of increasing depths. This technique reveals that both the temporal evolution of the line doubling, and its variation with depth in the atmosphere of RT\\,Cyg, are consistent with the `\\sch\\ scenario'. This scenario relates the temporal evolution of the red and blue peaks of the double absorption lines to the progression of a shock wave in the atmosphere.  The temporal evolution of the Balmer \\ha, \\hb, \\hg\\ and \\hd\\ emission lines around maximum light is also presented for RT\\,Cyg and X\\,Oph. The velocity variations of \\ha\\ and of the absorption lines are discussed in the framework of two competing models for the formation of Balmer emission lines in LPV stars. ",
        "introduction": "Long-period variable stars (LPVs) are cool giant stars showing more or less periodic light variations with amplitudes of several magnitudes in the visual and with periods of several hundred days. Depending on their visual amplitudes and on the regularity of their variability cycles, they appear in several flavours, namely Mira Ceti-type variables (Mira stars or Miras), Semi-Regular variables (of the SRa or SRb subtypes) or Irregular variables (of the Lb subtype). Mira and SR variables represent one of the latest stages in the evolution of stars with initial masses in the approximate range 1 to 9~\\Msun.  It is known since long that the brightness variations of LPVs go along with spectral changes: (i) hydrogen and some metallic lines turn from absorption to emission after minimum light (Merrill 1921), (ii) the velocity of the emission and absorption lines correlates with phase and excitation potential (Merrill 1923a; Adams 1941), (iii) several absorption lines appear double around maximum light (Adams 1941; Merrill \\& Greenstein 1958; Maehara 1968).  In a review on red stars, Merrill (1955) was the first to suggest that the bright emission lines sometimes appearing in LPVs may be explained by some kind of ``hot front'' moving outward. He further suggested that this running hot front may have the structure of a shock wave.  Although the shock wave scenario is nowadays quite widely accepted (see de la Reza 1986 and references therein), the lack of self-consistent pulsation models for LPV stars (due to the important role played by convection and the difficulty of modelling it) prevents from grasping exactly how, why and where emission and double absorption lines form. As a consequence, several (sometimes conflicting) theories aiming at explaining the spectral peculiarities of LPVs have appeared in the literature. Although the most popular models locate the formation of emission lines in the hot wake of the shock (e.g., Gillet 1988a), an alternative model proposed by Magnan \\& de Laverny (1997) associates emission lines with purely radiative non-LTE processes independently of any shock wave. In this model, the double-peak nature of the emission lines is not connected in any way with the velocity field, contrary to the shock-wave model.  Concerning the absorption lines, their doubling around maximum light is most easily seen in the rather clean near-infrared spectral domain (e.g., Gillet et al.\\ 1985; Hinkle et al.\\ 1997 and references therein). Long-term radial velocity measurements of the infrared rotation-vibration lines of CO have revealed well-defined variations with phase, regularly repeating from one cycle to the next and following a typical S-shaped curve (Hinkle et al.\\ 1997). These radial-velocity variations are generally believed to reflect the differential bulk motions occurring in the large and tenuous atmosphere of LPVs and associated with its pulsation. Alternatively, the complex absorption profiles have also been explained without resorting to differential velocity fields: Gillet et al.\\ (1985) have argued that the doubling of the metallic absorption lines results in fact from the development of an emission core in the line. This emission component is caused by the radiative release of the thermal energy dumped into the post-shock layer by the shock front. This model has the advantage of being able to account for the P~Cygni line profiles sometimes observed in LPV spectra (and appearing when the emission core becomes brighter than the local continuum). A similar explanation in terms of the shock-induced temperature inversion has been proposed by Karp (1975) for the line doubling observed in Cepheid variables. Detailed radiative-transfer calculations have shown that the formation of double absorption lines does not necessarily require a velocity gradient, since a temperature inversion is in principle sufficient.  Recent studies have clearly validated theories accounting for the complex absorption line profiles observed in several kinds of pulsating stars (RR\\,Lyrae: Chadid \\& Gillet 1996, Fokin \\& Gillet 1997; $\\beta$\\,Cephei stars: Mathias et al.\\ 1998; BL\\,Herculis: Gillet et al.\\ 1994, Fokin \\& Gillet 1994; RV\\,Tauri: Gillet et al.\\ 1989a) in terms of velocity fields. The validity of the \\sch\\ scenario accounting for double absorption lines in terms of a velocity gradient in the photosphere remains however to be demonstrated for LPV stars. The present paper presents the first clear observational evidence thereof. ",
        "conclusions": "The origin of double-peaked emission and absorption lines in the complex LPV atmospheres is a long-standing problem. If the shock wave and the associated velocity gradient are generally believed to be at the origin of the complex line profiles of LPV stars (de la Reza 1986; Gillet 1988b; Hinkle et al.\\ 1997), alternative models not resorting to differential atmospheric motions are sometimes invoked. They involve either purely non-LTE effects to account for the double-peaked Balmer emission (Magnan \\& de Laverny 1997), or a temperature inversion to account for the splitting of the absorption lines (Karp 1975; Gillet et al.\\ 1985).  Thanks to the cross-correlation technique and a daily survey of the Mira star RT\\,Cyg, the role of velocity fields in the formation of genuinely double absorption lines has been undoubtedly confirmed in the framework of the \\sch\\ mechanism. Evidence for the existence of strong shock waves in the outer layers of LPVs becomes compelling, although not all Mira stars exhibit the \\sch\\ scenario, as demonstrated by the observations of X\\,Oph collected in the present paper. The origin of this distinct behaviour will be investigated in the next paper of this series.  A tomographic technique that opens new perspectives for the study of the dynamics of LPVs has been used to visualize the outward motion of the shock wave in the atmosphere of RT\\,Cyg.  The evolution of the \\ha, \\hb, \\hg\\ and \\hd\\ emission-line profiles during 35\\% (20\\%) of the cycle of RT\\,Cyg (X\\,Oph) around maximum light has also been discussed. The comparison of this data set with the temporal evolution of the profile of absorption lines offers interesting perspectives to identify the origin of the complex profile of the \\ha\\ emission line.  The present paper had no other ambition than providing a description of the data collected with the ELODIE spectrograph around maximum light for RT\\,Cyg and X\\,Oph. These data may be used as constraints for future detailed models. Firm conclusions on the line-formation mechanisms in Mira variables should thus await the advent of such detailed dynamic models coupling hydrodynamics to radiative transfer."
    },
    "0010/astro-ph0010654_arXiv.txt": {
        "abstract": "In a previous paper on the Magellanic Clouds we demonstrated that coeval clusters provide a powerful tool for probing the progenitor masses of Wolf-Rayet stars (W-Rs) and Luminous Blue Variables (LBVs).  Here we extend this work to the higher metallicity regions of the Milky Way, studying 12 Galactic clusters. We present new spectral types for the unevolved stars and use these, plus data from the literature, to construct H-R diagrams.  We find that all but two of the clusters are highly coeval, with the highest mass stars having formed over a period of less than 1~Myr.  The turn-off masses show that at Milky Way metallicities some W-Rs (of early WN type) come from stars with masses as low as 20--25$\\cal M_\\odot$. Other early-type WNs appears to have evolved from high masses, suggesting a large range of masses evolve through an early-WN stage. On the other hand, WN7 stars are found only in clusters with very high turn-off masses, $>120\\cal M_\\odot$. Similarly the LBVs are only found in clusters with the highest turn-off masses, as we found in the Magellanic Clouds, providing very strong evidence that LBVs are a normal stage in the evolution of the most massive stars. Although clusters containing WN7s and LBVs can be as young as 1~Myr, we argue that these objects are evolved, and that the young age simply reflects the very high masses that characterize the progenitors of such stars. In particular we show that the LBV $\\eta$~Car appears to be coeval with the rest of the Trumpler~14/16 complex. Although the WCs in the Magellanic Clouds were found in clusers with turn-off masses as low as $45\\cal M_\\odot$, the three Galactic WCs in our sample are all found in clusters with high turn-off masses ($>70\\cal M_\\odot$); whether this difference is significant or due to small-number statistics remains to be seen. The BCs of Galactic W-Rs are hard to establish using the cluster turn-off method, but are consistent with the ``standard model\" of Hillier. ",
        "introduction": "\\subsection{Background}  Massive, luminous stars begin their H-burning lives as hot, O-type stars. During their main-sequence evolution (2.5--8~Myr for stars with initial masses of 120--20$\\cal M_\\odot$) they may lose a significant amount of their mass due to strong stellar winds.  The observed mass-loss rates suggest that the highest-mass stars will lose as much as half their mass during the H-burning stage. Since these winds are driven by radiation pressure through highly-ionized metal lines, the mass-loss rates will increase with stellar luminosity, metallicity, and effective temperature. It was Conti (1976) who first suggested that mass-loss provided a simple explanation for how Wolf-Rayet (W-R) stars form. In the modern version of the ``Conti scenario\" (Maeder \\& Conti 1994), this mass loss results in a stripping off of the H-rich outer layers of the star, resulting in a WN-type W-R star, in which the H-burning products (He and N) are enriched at the surface, with strong, broad emission lines indicative of enhanced stellar winds forming an extended atmosphere.  Most WNs are presumed to be He-burning objects, although there is evidence that a few H-rich late-type WNs are still near the end of the core-H burning phase (Conti et al.\\ 1995). Further mass-loss and evolution reveals the products of He-burning (C and O) at the surface, and the star is spectroscopically identified as a WC-type W-R.  Possibly the highest mass stars also go through a Luminous Blue Variable (LBV) stage on their way to becoming W-Rs, with the large, episodic mass loss that characterizes LBVs aiding the process. Stars of slightly lower luminosity (mass) will not have an LBV phase, but recent observations (Massey 1998a) suggest that they do go through an intermediate red supergiant (RSG) phase, even at high metallicities, albeit for a short time. At some lower luminosity one expects that the mass-loss rates are sufficient to produce a WN-type W-R, but not a WC.  And, at even lower luminosities, mass-loss rates are so low that the He-burning stage is spent entirely as a red supergiant (RSG) and not as a W-R star.  In this version of the ``Conti scenario\" we thus might expect the following ``paths\" to be followed, in order of decreasing luminosities:  \\begin{centering}  O$\\longrightarrow$LBV$\\longrightarrow$WN$\\longrightarrow$WC (1)  O$\\longrightarrow$RSG$\\longrightarrow$WN$\\longrightarrow$WC (2)  O$\\longrightarrow$RSG$\\longrightarrow$WN (3)  OB$\\longrightarrow$RSG (4)  \\end{centering}   The masses corresponding to these paths are unknown (and indeed we are unsure even of the qualitative correctness of these paths), but ``standard guesses\" for characteristic values would be 1: $\\geq80\\cal M_\\odot$, 2: $60\\cal M_\\odot$, 3: $40\\cal M_\\odot$, and 4: $20\\cal M_\\odot$.  We emphasize that these are purely speculative values, and that the actual ranges should depend upon metallicity. Indeed it is to address this issue that the present series of papers has come about.  Between the initial O-type stage (of luminosity class ``V\") and the He-burning stage (LBVs, W-Rs, RSGs) the star will become a B-type supergiant; most, but not all, of these are expected to still be H-burning objects (Massey et al.\\ 1995b).  The luminous A-F supergiants are very short-lived intermediate stages during He-burning for stars of intermediate high-mass, depending upon the metallicity. The precursor to SN1987A is believed to have been a ``second-generation\" B-type supergiant, a He-burning object of somewhat lower mass than those being discussed here.  In addition to the question of whether or not the above paths are correct, and what masses to assign to each as a function of metallicity, we are also interested in the evolutionary significance, if any, to the Wolf-Rayet spectral subtypes. WN-type W-R stars are classified as ``early\" (WNE) or ``late\" (WNL) depending on whether NV~$\\lambda\\lambda 4602,19$ or NIII~$\\lambda\\lambda 4634,42$ dominates; e.g., WNEs correspond to numerical subtypes WN2, WN3, WN4, while WNLs consist of subtypes WN6, WN7, WN8, and WN9, with the WN5 stars split between the two groups. Similarly WC-type W-R stars are classified as WCE or WCL depending upon whether CIII~$\\lambda 5696$ dominates over CIV~$\\lambda 5808$; e.g., corresponding to spectral subtypes WC4 through WC6 and WC8-9, with some WC7s falling into each camp (Conti \\& Massey 1989).  Do these subtypes mean anything in an evolutionary sense?  Various authors have claimed so (see, for example, Moffat 1982), but this conjecture does not seem borne out by either observational or theoretical studies.  Our understanding of massive star evolution is limited, in part, because of the difficulty of constructing stellar models from first principles.  The physics of massive stars is complicated by strong stellar winds, and the choice of the functional dependence of mass-loss rates on stellar parameters (luminosity, temperature, mass, and metallicity) greatly influences the theoretical tracks (e.g., Meynet et al.\\ 1994), particularly in the later stages of evolution.  In addition, the models are sensitive to the amount of mixing. However, there is little agreement on the treatment of the relevant processes of semi-convection and overshooting (Maeder \\& Conti 1994), while the most recent work has emphasized the significant role that rotation may play in this regard (Maeder 1997, 1999). Nevertheless, empirical studies of massive star evolution provide confidence that the above picture is correct, and are beginning to provide quantitative information on the mass ranges corresponding the various paths.  These studies provide an observational basis against which the models can be evaluated and refined. (For a humorous rendering of this process, the reader is referred to Fig.~5 in Conti 1982.)  \\subsection{An Observational Approach} \\subsubsection{Global Studies}  The galaxies of the Local Group provide perfect laboratories for pursing these studies observationally, as the metallicity differs by almost an order of magnitude (SMC to M31) among the galaxies currently active in forming stars. During the past few years there have been a number of studies of {\\it mixed-age} populations, the relative number of {\\it this} and {\\it that}. The implicit assumption of these studies are that the IMF slopes are identical and cover regions that provide a good sampling of stellar stages over time.  The number ratios provide quantitative criterion for the models to attempt to match. These studies have found the following:  (1) The relative number of RSGs to W-Rs decreases with increasing metallicity (Massey 1998a; Massey \\& Johnson 1998). Histograms of the number of RSGs vs.\\ luminosity reveal that there are proportionately fewer high luminosity RSGs at higher metallicity, while the lack of a sharp luminosity cut-off supports the interpretation that as $Z$ increases, massive stars spend a greater fraction of their He-burning phase as W-Rs, and a smaller fraction as RSGs, rather than there being a difference in the actual mass ranges that go through a RSG phase.  This is why we indicated a RSG phase at high luminosities (path 2 above). Possibly even LBVs will go---or have gone---through a RSG phase, but this is unknown. We also do not know if massive stars go through a RSG phase at the highest metallicity: the relation between the RSG/W-R ratio and metallicity appears to flatten below the high metallicity that characterize M31, but only a few regions have been surveyed in that large galaxy, and more data are being gathered to resolve this.  (2) The relative number of WCs to WNs increases with increasing metallicity, with the notable exception of the star-burst galaxy IC~10 (Fig.~8 in Massey \\& Johnson 1998).  This trend is also in accord with the predictions of the Conti scenario, as increased mass-loss makes it possible for a star of a given luminosity to reach the WC stage sooner, spending more of its W-R stage as a WC than a WN. (The explanation for IC10's peculiarly high WC/WN ratio remains a mystery at present. See discussion in Massey \\& Johnson 1998.)  (3) Significant differences in the spectral subtypes found in the Magellanic Clouds compared to the Milky Way were noted by Smith (1968): no WCLs are found in the LMC or SMC, and most of the ones known in the Milky Way are found inwards of the solar circle. Similar differences are seen for the WNs. Armandroff \\& Massey (1991) showed that the WC line widths (which are correlated with spectral subclass) change systematically with metallicity, extending an important finding by Willis, Schild, \\& Smith (1992) to other galaxies of the Local Group. Massey (1996) proposed that the WC spectral subtypes are nothing more than an atmospheric effect due to metallicity. (See also Massey \\& Johnson 1998).  Recently Crowther (2000) has demonstrated from W-R model atmospheres that the WN subtypes may similarly be a reflection of metallicity rather than other stellar parameters, at least in terms of WN3 through WN6.  \\subsubsection{Coeval Associations} A more direct way exists to attack the problem of understanding massive star evolution.  By using coeval associations that contain evolved, massive stars, we can in fact directly {\\it measure} the mass ranges that correspond to the above evolutionary paths as a function of metallicity. This is the subject of the current series of papers.  In Paper I of this series (Massey, Waterhouse, \\& DeGioia-Eastwood 2000) we established that many Magellanic Cloud OB associations and young clusters are sufficiently coeval ($\\Delta \\tau<$1~Myr) that we can measure a meaningful cluster turn-off mass; e.g., the mass of the most massive star on the main-sequence.  These turn-off masses then place a lower-limit on the mass of the progenitors of the evolved stars in the cluster, to the extent that star-formation proceeded coevally. If the cluster is well populated, than the initial mass of the turn-off star is a good approximation to the initial mass of the progenitor of the W-R star. In addition, the bolometric luminosity of the cluster turn-off sets useful limits on the bolometric corrections for the evolved stars, allowing tests of Wolf-Rayet model atmospheres, such as Hillier's ``standard model\" (Hillier 1987, 1990).  The results were quite revealing. In a study of 19 OB associations in the Magellanic Clouds, we found that at the low metallicities that characterize the SMC, only the highest-mass stars ($>70 \\cal M_\\odot$) become Wolf-Rayet stars, although the sample is  small. This is equivalent to saying that path (3) above occurs only for $M>70\\cal M_\\odot$ for $Z\\leq 0.005$.\\footnote{For convenience in talking about the metallicity $Z$, we adopt $Z=0.018$ for the solar neighborhood, corresponding to log~O/H+12=8.70 (Estebam \\& Peimbert 1995).  If we then simply scale $Z$ relative to the easily-measured oxygen abundance, $Z=0.005$ for the SMC (log~O/H+12=8.13) and $Z=0.008$ for the LMC (log~O/H=8.37), e.g., Russell \\& Dopita (1990).  Although it is well recognized that different metals will have different relative abundances, it is fortuitous that it is oxygen (along with carbon and nitrogen) which are the primary accelerators of the stellar winds at the high effective temperatures appropriate to O-type stars (Abbott 1982).} At the higher metallicity of the LMC ($Z=0.008$), WN W-R stars come from stars with masses as low as $30\\cal M_\\odot$. We also found that WC stars are found in the same clusters as WNEs; e.g., the lowest turn-off mass found for a cluster containing WC stars was 45$\\cal M_\\odot$, suggesting that stars with masses from 30--45$\\cal M_\\odot$ might correspond to path (3), while stars with masses 45-85$\\cal M_\\odot$ correspond to path (2).   The rare ``Ofpe/WN9\" stars\\footnote{The stars may be rarer at higher metallicity than at low; ten are known in the LMC (Bohannan \\& Walborn 1989), while only one is known in the Milky Way (Bohannan \\& Crowther 1999). Six are known in M~33 (Massey et al.\\ 1996), while one is known in the higher metallicity M~31 (Massey 1998b).}, once thought to be a transition type between ``Of\" stars and Wolf-Rayet stars, are only found in clusters and associations with the lowest turn-off masses, 25-35$\\cal M_\\odot$.  Recently the Ofpe/WN9 stars were implicated in the LBV phase of massive stars, after one Ofpe/WN9 star (R127) underwent an ``LBV-like\" outburst.  But, the classical LBVs in our LMC sample, including the archetype itself S Doradus, are found in clusters with the very highest turn-off masses, $>85 \\cal M_\\odot$.  (Similarly the SMC W-R star HD~5980, which many consider to be a ``true\" LBV [Barb\\'a et al.\\ 1995] is found in a cluster with a very high turn-off mass.) We conclude that the Ofpe/WN9 stars are just the lowest-mass versions of Wolf-Rayet stars. True LBVs, on the other hand, are found only in the clusters with the highest mass turn-offs, suggesting that they are indeed stars near their Eddington limit and are a normal stage of the most massive stars.  Our study also shed light on the origin of the different W-R classes and subtypes, at least at the modest metallicity that characterizes the LMC. The early-type WN stars (WNEs) in the LMC are found in clusters with a large range of turn-off masses (from 30$\\cal M_\\odot$ to 100$\\cal M_\\odot$ or more), suggesting that these are a stage that most massive stars go through at LMC-like metallicities.   We turn now to the higher metallicity of our own Milky Way, and pose the question of where do LBVs and Wolf-Rayets of various types come from at a metallicity considerably higher than that of the Magellanic Clouds. ",
        "conclusions": "  \\begin{enumerate} \\item Wolf-Rayet stars in the Milky Way are found in clusters containing a large range of turn-off masses.  The data suggest that at the metallicity that characterizes the Milky Way  some early-type WN Wolf-Rayet stars come from progenitors with masses as low as 20--25$\\cal M_\\odot$.  \\item The WNEs may come from a large range in masses, as they do in the Magellanic Clouds (Paper~I), but this result is uncertain, as the one high-mass WNE star in our sample is a member of a close binary, V444~Cyg.  \\item WN7 stars are found only in clusters with the highest masses. The youngest of these are only 1~Myr old.  Although these could still be H-burning objects, ``guilt by association\" suggests that these are in fact evolved massive stars, as the youngest region also contains the LBV $\\eta$~Car, thought to be an evolved object.   \\item $\\eta$ Car itself is found to be highly coeval with the rest of the Tr~14/16 complex, despite the region's young age.  It, and the LBV candidate VI~Cyg~No.~12, are found in clusters with the highest masses. This is identical to what we found for the Magellanic Clouds in Paper~I, and argues strongly that LBVs are a normal stage in the evolution of the most massive stars.  \\item The Galactic WC stars are found in clusters with turn-off masses $> 70 \\cal M_\\odot$. In the Magellanic Clouds we find WCs occurring in clusters with masses as low as 45~$\\cal M_\\odot$. We argued in Paper~I that most WNs thus evolve to WCs.  The data for the Milky Way might suggest that only the more massive stars become WCs, but the sample size in the Milky Way is small (3 WCs in 3 clusters) and additional data are needed.  \\item The BCs of Galactic W-Rs are hard to determine using the cluster turn-off methods, as considerable luminosity evolution is expected at the higher mass-loss rates that characterize the Milky Way luminous stars. The data are at least consistent with the ``standard model\" of Hillier (1987, 1990) as applied to two of the stars in our sample (Crowther et al.\\ 1995).  In Paper~I we found concluded that there was excellent agreement, with the BCs of early WN and WC stars found to be extreme ($\\approx -6$~mag).  \\end{enumerate}  We note that much recent work has established the need to extend the theoretical evolutionary tracks to masses higher than 120$\\cal M_\\odot$. Stars with masses estimated to be as high as 160$\\cal M_\\odot$ have been found in the R~136 cluster (Massey \\& Hunter 1998), and the Galactic clusters Trumpler~14/16, C1715-387, Pismis~24, and Cyg~OB2 all contain stars whose luminosities place them above the highest evolutionary track computed by the Geneva group (120~$\\cal M_\\odot$).  Future observational work is needed to extend the H-R diagrams of these and other Galactic clusters, and to investigate other coeval regions in the Milky Way and nearby galaxies that can be used to extend these studies."
    },
    "0010/astro-ph0010462_arXiv.txt": {
        "abstract": "We present preliminary results on the effects of mergers on the chemical properties of galactic objects in hierarchical clustering  scenarios. We adopt a hydrodynamical chemical code that allows to describe the coupled evolution of dark matter and baryons within a cosmological context. We found that disk-like and spheroid-like objects have distinctive metallicity patterns that may be the result of different evolution. ",
        "introduction": " ",
        "conclusions": ""
    },
    "0010/astro-ph0010181_arXiv.txt": {
        "abstract": "We have imaged seventeen recently discovered Sirius-like binary systems with {\\it HST/WFPC2} and resolved the white dwarf secondary in eight cases. Most of the implied orbital periods are of order several hundred years, but in three cases (56 Per, $\\zeta$ Cygni and RE~J1925$-$566) the periods are short enough that it may be possible to detect orbital motion within a few years. It will then be possible to derive dynamically determined masses for the white dwarfs, and potentially these stars could be used as stringent tests of the mass-radius relation and initial-final mass relation. ",
        "introduction": "Although universally accepted, the white dwarf mass-radius relation (Chandrasekhar 1931; Hamada \\& Salpeter 1961; Wood 1992) has resisted in-depth examination through direct observations for $\\approx$70 years. The difficulty arises mainly from obtaining accurate, model-independent measurements of the masses and radii of known white dwarfs. A combination of spectroscopically derived temperatures and gravities, and accurate parallaxes can provide the independent means of determining $M_{\\rm WD}$ and $R_{\\rm WD}$, and therefore a new route for examining evolutionary models. The white dwarfs observed by {\\it Hipparcos} (Vauclair et~al. 1997; Provencal et~al. 1998) were entirely consistent with expectations, although they were restricted mainly to a narrow range around $0.6M_\\odot$. Furthermore, even with these new parallaxes the overall uncertainties are still too large to provide a really thorough examination of the differences between theoretical models which assume a variety of core and envelope compositions.  White dwarfs in resolved binary systems can provide the most stringent tests of evolutionary models, since the mass can be determined from the orbital and physical elements of the system. In practise, though, few such systems are available to be studied in sufficient detail (they are: Sirius, Procyon, 40 Eri and Stein 2051). However, as reported at previous workshops (e.g.~Burleigh 1999) $>20$ new Sirius-like binaries have been identified through the {\\it ROSAT WFC} and {\\it EUVE} all-sky surveys. Each system consists of a normal or subgiant star plus a hot white dwarf companion, which is responsible for the soft X-ray and EUV flux detected by {\\it ROSAT} and {\\it EUVE}. Crucially, in most cases, the bright primary has an accurate {\\it Hipparcos} parallax, yielding a precise distance for the white dwarf. Thus this new sample of Sirius-like binaries presents a golden opportunity of extending the sample of well-studied white dwarf binary systems. Our first aim, therefore, is to identify those systems that can be resolved and have their orbits measured within an acceptable timescale. Unfortunately, none of these new binaries can be resolved from the ground, due to the huge difference in brightness between the components ($>5$ mags.), although we know from radial velocity studies that most of these systems are wide with periods $>$few years (e.g.~Vennes, Christian, \\& Thorstensen 1998).  The {\\it Hubble Space Telescope} provides an answer to this problem. Unencombered by the Earth's atmosphere, it allows imaging in the UV where the brightness of the hot white dwarf and its companion are similar. It also delivers a diffraction limited resolution of $\\approx0.05\\arcsec$, making it possible to resolve binary components with separations as small as 0.1\\arcsec. We report here the first results of a snapshot survey with {\\it HST/WFPC2} to image these new Sirius-like binary systems. ",
        "conclusions": ""
    },
    "0010/astro-ph0010148_arXiv.txt": {
        "abstract": "We present large-scale CCD images of a dozen low surface brightness objects in the vicinity of giant Sc galaxy NGC~6946, which have been found recently on POSS-II copies. Six of them: UGC~11583, kk251, kk252, kkr55, kkr56, and kkr59 are dwarf irregular galaxies resolved into stars. We determined their distances from the luminosity of the brightest blue stars. The mean estimated distance to the NGC~6946 group is 5.9$\\pm$0.4 Mpc. Together with Cepheus 1, discovered before, the group consist of 8 late type galaxies. Almost all of them are detected or marginally detected in the HI line. The group is characterized by a linear projected diameter of 420 kpc, a radial velocity dispersion of 95 km~s$^{-1}$, and a virial mass-to-luminosity ratio about 56 $M_{\\sun}/L_{\\sun}$. ",
        "introduction": "NGC 6946 is amongst the seven more massive galaxies in the Local Volume (=LV) with distances  $D <$ 7 Mpc (Karachentsev et al. 1999). 51\\% of the population of the LV are within the gravitational influence of these \"oligarchs\". Each of 7 the giant galaxies, except NGC~6946, has a lot of dwarf companions: from 7--9 members around Sc galaxies M101 and NGC~5236 to 32 members around giant elliptical galaxy NGC~5128. In this sense the absence of accompying dwarf systems around NGC~6946 within a radius of $\\sim$1 Mpc (Tully, 1988) looked like a remarkable anomaly. However, the apparent isolation of NGC~6946 may be related with its location on a border of the Local Void (Tully 1988). The substantial galactic extinction in the direction of NGC~6946 has also an influence on the visibility of faint dwarf galaxies.  Based on copies of the Second Palomar Observatory Sky Survey (=POSS-II), Karachentseva \\& Karachentsev (1998) undertook a search for galaxies of low surface brightness in the vicinity of normal galaxies in the LV. Observations of these galaxies (kk- list) with the 100-m Effelsberg radio telescope revealed three dwarf galaxies: UGC~11583 = kk250, kk251, and kk252, with positions and radial velocities close to NGC~6946 (Huchtmeier et al. 1997). Using the luminosity of the brightest stars in NGC~6946 and UGC~11583, Sharina et al. (1997) derived distance moduli to these galaxies: 28.90 and 29.13 mag, respectively. During the course of a search for compact high velocity clouds Burton et al. (1999) discovered beside NGC~6946 a low surface brightness galaxy with radial velocity $V_h= +58$ km~s$^{-1}$ named by them Cepheus 1. Subsequent searches for nearby dwarf galaxies by Karachentseva et al. (1999) (kkr- list) and Karachentseva et al. (2000) (kkh- list) led to the discovery of some more LSB objects around NGC~6946. In this paper we present results of optical and HI observations of these new objects, particularly data on three new probable companions of NGC~6946. ",
        "conclusions": "Being located at a low galactic latitude, $b = 12\\degr$, in a zone of substantial extinction, NGC~6946 looks at first glance like an isolated giant spiral galaxy on the border of the Local Void (Tully, 1988). Recent searches for low surface brightness galaxies on the POSS-II films revealed about a dozen probable companions to NGC~6946. Subsequent HI observations of the candidates and distance estimates from the brightest stars showed that at least 7 dwarf irregular galaxies can be considered as physical companions of NGC~6946. Therefore, a new nearby group of galaxies has been discovered at a distance of (5.9$\\pm$0.4) Mpc. The mean projected diameter of the group is 420 kpc, its root mean square radial velocity is 95 km~s$^{-1}$, and the virial mass-to-luminosity ratio is $\\sim56 M_{\\sun}/L_{\\sun}$, typical for small galaxy groups. The two nearby giant face-on Sc galaxies: M101 and NGC~6946, and their dwarfs are alike with respect to their structure and kinematics. Being gas-rich galaxies, the members of the NGC 6946 group need to be studied in more detail with radio HI synthesis interferometry."
    },
    "0010/astro-ph0010524_arXiv.txt": {
        "abstract": "We present a detailed analysis of the iron K$\\alpha$ line variabilities in NGC 4151 by using long ASCA observation data obtained in May 1995. Despite the relatively small amplitude variations in the continuum flux, the iron K$\\alpha$ line flux and profile show dramatic variations. Particularly, the line profile changes from single peak to seeming double peaks and back in time scales of a few 10$^4$ sec. The seemingly double-peaked profiles can be well interpreted as line emission from a Keplerian ring around a massive black hole. An absorption line at around 5.9 keV is also marginnaly detected. We discussed current Fe K line models, but none of them can well explain the observed line and continuum variations. ",
        "introduction": "Broad Fe K fluorescence line profiles, detected in many AGNs (Tanaka et al. 1995, Nandra et al. 1997, Wang et al. 1999a, Nandra et al. 1999, and references therein), are thought to arise from the accretion disk near the center massive black hole of AGNs. The line profile, thus, carries important information about the distribution and kinematics of optically material as well as about the space properties in the vicinity of the postulated super massive black hole. Doppler and gravitational shifts would imprint characteristic signatures on the line profile which map the geometric and dynamical distributions of matter surround the black hole.  Additional information concerning the geometry of X-ray source and matter in the active nucleus could in principle be derived by studying the rapid variability (on time scales of several 10$^4$ s or less) of the line profile, intensity and their relationship with the continuum variations. So far, rapid variability in Iron K line has been detected in NGC 7314 (Yaqoob et al. 1996), MCG --6-30-15 (Iwasawa et al. 1996, 1999), NGC 4051 (Wang et al. 1999b) and NGC 3516 (Nandra et al. 1999).  The observed Fe K$\\alpha$ line rapid variabilities in different targets or during different observations of the same target are quite different. Yaqoob el al. (1996) presented the evidence for rapid variability of the Fe K line profile in the narrow-line Seyfert galaxy NGC 7314, which is consistent with a disk-line of constant equivalent width superposed on a constant flux narrow line (presumably from the torus). Iwasawa et al. (1996) discovered that, during the ASCA observation on MCG --6-30-15 in 1994, when the source is bright, the Fe K line is weak and dominated by the narrow core, whilst during a deep minimum, a huge red tail appears. The intensity of broad Fe K line correlates inversely with the continuum flux on the time scales of several 10$^4$ s. And Iwasawa et al. (1999) discovered that during the ASCA observation on the same target in 1997, the Fe K$\\alpha$ line of a flare phase peaks around 5keV and most of its emission is shifted to below 6 keV with no component detected at 6.4keV. While for NGC 4051, during the ASCA observation in 1994, the equivalent width and the width of Fe K line correlate positively with the continuum flux (Wang et al. 1999b), which shows an opposite trend with the ASCA observation on MCG --6-30-15 in 1994. The rapid variabilities of Fe K$\\alpha$ line profiles in NGC 3516 are also quite irregular (Nandra et al. 1999, Wang et al. 2000).  The nearby Seyfert 1.5 galaxy (z = 0.0033), NGC 4151, was first established as a bright X-ray source some thirty years ago (Gursky et al. 1971). Since that time, this archetypal low-luminosity active galaxy has invariably been considered as a prime target by all major X-ray astronomy missions. In 1995, NGC 4151 was observed by ASCA from May 10 to 12 (Leighly et al. 1997). The net exposure time of the observation is about 200 ks. Because of the enhanced sensitivity provided by ASCA and the much longer exposure time, the obtained Fe K$\\alpha$ line profile has very high data quality (Wang et al. 1999a, here after W99). In this letter we report the discovery of unique and rapid variabilities of the iron line profiles in NGC 4151 during the ASCA observation in 1995. ",
        "conclusions": "\\subsection{The seemingly double-peaked line profiles}  Massive black holes are generally thought to exist at the centers of active galaxies (Rees 1984), but unambiguous identification of a black hole has been impeded by lack of evidence for the strong-field relativistic effects expected in the vicinity of black holes. Because of its unique anticipated profiles, the Fe K$\\alpha$ fluorescence line has long been considered as an observational key to the study of the innermost regions of active galactic nuclei (AGN). Fabian et al. (1989) and Laor (1991) pointed out that the line profiles from the accretion disk should be broad, skewed and double-peaked (dependent on the inclination of the disk and the line-producing region in the disk), which are characteristics of the Doppler effects of the accretion disk and the strong gravitational field in the vicinity of the central black hole.  The Fe K$\\alpha$ line profiles detected in many AGNs have been proved to be skewed and extremely broad (Tanaka et al. 1995, Nandra et al. 1997, Wang et al. 1999a, Nandra et al. 1999, and references therein), consistent with the disk line model, so give the ever strongest evidence for the presence of massive black holes and accretion disks in the center of AGNs.  However, some alternative models, such as Comptonization and jet, can also explain the broad and skewed Fe K$\\alpha$ line profiles. Those models do not require strong gravitational field and so give no support to the presence of massive black holes (Fabian et al. 1995, Misra \\& Sutaria 1999). But different to the disk line model, those models predict a smooth line feature, but not a double-peaked one. Though some indirect evidences have shown that they are not satisfactory (Fabian et al. 1995, Misra 2000, Reynolds \\& Wilms 2000), as Misra \\& Kembhavi (1998) have pointed out, we infer that the detection of a double-peaked line profile is the key to confirm the disk line model for the observed Fe K$\\alpha$ lines, and so on confirm the detection of extremely strong gravitational field.  In this paper we find that the Fe K$\\alpha$ line profile of NGC 4151 is seemingly double-peaked during phase I-3 and I-4 in Fig1. Because an additional resonant absorption line at 5.9 keV has also been detected in the line profile in I-4, we only focus on the profile of I-3 in this section. We show the seemingly double-peaked line profile of I-3 in detail in Fig.3, and the best-fit two gaussians and disk-line models are also plotted.  The seemingly double-peaked line profile strongly suggest that the line arises from materials with disk-like geometry. Additional strong gravitational field is also needed to explain the red shift of the profile. Others models for the broad line profiles, such as Comptonization or jet, can not be responsible for a double-peaked line profile, and disk-like outflows or inflows with high velocity can not be responsible for the red shift of the line profile. The best disk line fits show that the line mainly arises from a narrow ring between 10.8 R$_g$ and 18.6 R$_g$.  Currently, perhaps due to the limited energy resolution and data quality, obvious double-peaked Fe K$\\alpha$ line profile has never been detected in any active galactic nuclei. We noticed that the line profile of MCG --6-30-15 in Fig. 1b of Wang et el. (1999a) looks double-peaked, but the plot was folded through the effective area of the detectors which rises sharply towards the low energy, so did not give the true profile of the line. While in plottings corrected for detectors' effective area, we can not see obvious double peaks (see Fig. 7 of Iwasawa et al. 1996 and Fig. 6 of Fabian et al. 2000). We hope that the new generation X-ray satellites, such as XMM, can show us clearly double-peaked iron line profiles of AGNs.  \\subsection{The rapid variabilities of the Fe K$\\alpha$ line profiles}  Additional to the double-peaked line profiles, the rapid variabilities of the Fe K line profiles in NGC 4151 are also detected in this letter. We divided the whole ASCA observation of NGC 4151 in 1995 into five time intervals, and found that, despite the relatively small amplitude variations in the continuum flux, the Fe K$\\alpha$ line flux and profiles show dramatic variations: during the first time interval I-1, the iron line profile is broad and skewed, showing a strong peak at 6.4 keV, and a huge red wing extending to $\\sim$ 4.5 keV, similar to the time-averaged line profile; after several 10$^4$ s, the huge red wing becomes very weak (I-2); for I-3, except the strong peak at 6.4 keV, there is also a seeming peak at around 5.4 keV; the line profile of I-4 is similar to that of I-3, but much weaker; and finally, the line profile change back to single peak (I-5).  The line equivalent widths of the five intervals are 375, 148, 329, 168 and 300 eV respectively, showing obvious variations.  The rapid variabilities of Fe K line profiles discovered previously have suggested that the X-ray continuum flux is generated by magnetic flares above the accretion disk, and at any given time there are only a few flares moving around above the disk (Iwasawa et al. 1996, Fabian 1997, Wang et al. 1999b, Iwasawa et al. 1999, Wang et al. 2000). Flares with different location and different motion can give different fluorescence line profiles.  Can the magnetic flare model also explain the observed rapid variabilities of Fe K$\\alpha$ line in NGC 4151? For I-1 and I-5, the X-ray continuum (in the flare model context) could be dominated by a single or several strong flares located very close to the center black hole, while the inner disk radius from the disk line fits (7.4 and 8.3 R$_g$) gives strong supports to this model. For I-2, the X-ray continuum seems to be dominated by flares far away from the center black hole (R$_i$ = 17 R$_g$). For I-3 and I-4, the X-ray continuum could be dominated by a single or several strong flares, which are  $\\sim r_g$ above and coroating with the disk at around 14 R$_g$, with only the materials in the ring around 14 R$_g$ being illuminated.  The large dramatic decreasing in the line EWs from I-3 to I-4 might be due to different degree of the anisotropy in the X-ray continuum, i.e., larger fraction of the X-ray continuum received by the disk in I-3 than in I-4. However, it requires some fine tune in order to keep the observed continuum more or less constant. Alternatively, Misra (2000) suggested that the changes in the disk structure may explain the observed line profile variations, but, a disk undergoing structure changes is also hard to keep the same radiative spectrum. Finally, the decreasing of the fluorescent yielding may be due to the increasing ionization in the disk skin.  Under flare context, large change in the ionization structure of disk skin is possible even for quite moderate variations in the observed continuum flux if the flares have different heights If this is case, we would expect that the flare is lower in the I-4 than in I-3. This should also reflect in the line profile. However, the uncertainty in the disk parameters do not allow us to verify this.  The observed properties strongly suggest that the rapid line variations may not be due the variations of X-ray continuum source. Instead, it may be due to changes in geometry or ionization level of the reflecting medium and at the same time, these changes should not affect the X-ray producing medium. If so, except for the variations of the X-ray continuum source, what can lead to the changes in geometry or ionization level of the reflecting medium?  An absorption line at 5.8 keV is marginally detected in I-4, but not significant in other intervals. Note a variable absorption line at the same energy has been discovered by Nandra et al. (1999) in NGC 3516. Nandra et al. (1999) interpreted this line as the redshifted Fe K resonant absorption line. The large redshift is either caused by the infalling absorbing material or by the strong gravity in the vicinity of the black hole. Given the fact that the line is highly variable and the large redshift, the agreement of the redshifts in this two objects perhaps is not purely coincidence, but have some physical origin. If so, the gravitational redshift explanation would be favored."
    },
    "0010/astro-ph0010530_arXiv.txt": {
        "abstract": "Relying on his photometry of galaxies, Holmberg (1958) found, more than 40 years ago, that the color indices of paired galaxies were closely correlated. Our deep broad-band BVRI CCD photometry of 45 (S+S) pairs from the Karachentsev (1972) catalogue (see Hern\\'andez Toledo \\& Puerari, this volume), and additional (B-V) color information from the literature (50 extra (S+S) pairs), help us to confirm the effect. This ``Holmberg Effect'' has long been remained unverified and not explained yet. ",
        "introduction": "Hints that galaxy interactions might play an important role in inducing multi-wavelength emission enhancement date back at least 40 years. One of the first quantitative indications came from Holmberg (1958), as a byproduct of his famous photometric survey of nearby galaxies. By comparing the photographic colors of paired galaxies, he found a significant correlation between the colors of pair components. This phenomenon has since been referred to as the ``Holmberg effect''. Although the physical explanation of the Holmberg effect is complex, it reflects in part a tendency for similar types of galaxies to form together (morphological concordance), a possible reflection of the role of local environment in determining galaxy morphology, but it can presumably also reflect mutually induced star formation (Kennicutt 1998). ",
        "conclusions": ""
    },
    "0010/astro-ph0010195_arXiv.txt": {
        "abstract": "We test the methods developed in previous papers for inferring the intrinsic shapes of elliptical galaxies, using simulated objects from $N$-body experiments. The shapes of individual objects are correctly reproduced to within the statistical errors; close inspection of the results indicates a small systematic bias in the sense of underestimating both the triaxiality and short-to-long axis ratio. We also test the estimation of parent shape distributions, using samples of independent, randomly oriented objects. The best results are statistically accurate, but on average the parent distributions are again slightly biased in the same sense. Using the posterior probability densities for the individual objects, we estimate the magnitude of the bias to be $\\lesssim 0.1$ in both shape parameters. The results support the continued use of these methods on real systems. ",
        "introduction": "}  Observational estimates of the true three-dimensional shapes of elliptical galaxies serve as diagnostics of the physics of galaxy formation and evolution. Numerical simulations of protogalactic collapse (Dubinski \\& Carlberg \\markcite{DC91}1991), mergers (Weil \\& Hernquist \\markcite{WH96}1996, Barnes \\& Hernquist \\markcite{BH98}1998), and black-hole growth (Merritt \\& Fridman \\markcite{MF98}1996, Merritt \\& Quinlan \\markcite{MQ98}1998) all suggest that dark halos {\\it should\\/} tend toward axisymmetry with increasing dissipation or central mass concentrations. But whether this has, in fact, happened is difficult to establish. The effort to uncover the intrinsic shape distribution has a long history (reviewed by Statler \\markcite{Sta95}1995, updated by Bak \\& Statler \\markcite{BS00}2000, hereafter BS). Photometric methods, introduced by Hubble \\markcite{Hub26}(1926) and revisited by many authors (most recently Ryden \\markcite{Ryd92}1992, \\markcite{Ryd96}1996; Fasano \\markcite{Fas95}1995), have been effective in constraining the distribution of flattenings, but not in establishing the relative numbers of triaxial and axisymmetric systems. For constraining triaxiality, dynamical information is necessary, with both gas (Bertola et al. \\markcite{Ber91}1991) and stellar (Binney \\markcite{Bin85}1985, Franx et al. \\markcite{FIZ91}1991, Tenjes et al. \\markcite{Ten93}1993) kinematic data having been used for this purpose.  Earlier papers in this series (Statler \\markcite{S94a}1994a, Statler \\& Fry \\markcite{SF94}1994, Statler \\markcite{S94b}1994b, hereafter Papers I, II, and III respectively) presented a method for statistically estimating the intrinsic shapes of elliptical galaxies from observations of their radial velocity fields and surface brightness distributions. Paper III included a test of the method on the end product of an $N$-body collapse simulation (Dubinski \\markcite{D92}1992). Using simulated observations of the object along a single line of sight, the method could recover the true shape only to within a $2\\sigma$ error region; this result was attributed to an unfortunate line of sight. Since then, the method has profitably been applied to real systems, to derive both the shapes of individual galaxies (NGC 3379, Statler \\markcite{S94c}1994c, Statler \\markcite{S00}2001, NGC 1700, Statler et al.\\ \\markcite{SDS99}1999, hereafter SDS) and the parent shape distribution of a sample (Davies \\& Birkinshaw \\markcite{DB88}1998) of radio ellipticals (\\markcite{BS00}BS). Nonetheless, there remains a lingering impression that the performance on realistic simulated systems is ``disappointing'' (Binney \\& Merrifield \\markcite{BM98}1998).  In this paper we present more extensive tests, using a homogeneous set of simulated elliptical galaxies from group-merger experiments (Weil \\& Hernquist \\markcite{WH94}1994, \\markcite{WH96}1996, hereafter WH94 and WH96, collectively WH). We demonstrate that the method performs well, both in estimating the shapes of individual objects and in deriving the shape distribution of small samples. The presentation is ordered as follows: Section 2 briefly describes the test objects and the methods employed, with references to earlier papers where complete expositions can be found. Section 3 describes the tests performed and their results. Section 4 deals with residual systematic bias and its possible origin, and Section 5 sums up. ",
        "conclusions": "}  We would like to understand the origin of the small systematic bias in our results. One possibility is that the small samples used in \\S\\ \\ref{s.samples} could be biased in orientation, but this is not the case. Figure \\ref{f.orientations} shows the distribution of the lines of sight, relative to the principal axes of the observed objects, for each sample. There is no obvious clustering of views. The isotropy of the distribution can be quantified using $|Q_1|$, the largest (in absolute magnitude) eigenvalue of the quadrupole moment tensor, normalized to the number of objects. A Monte Carlo simulation of $10^5$ six-object samples gives a skewed distribution with a mean of $0.44$, a mode of $0.36$, and a standard deviation of $0.22$. The $|Q_1|$ values for Samples 1, 2, and 3 are $0.35$, $0.42$, and $0.24$ respectively, so the samples are neither abnormally anisotropic nor abnormally isotropic.  \\begin{figure}[t]{\\epsfxsize=4.5in\\hfil\\epsfbox{f4.eps}} \\caption{\\footnotesize Distribution of lines of sight for the observed objects in ($a$) Sample 1, ($b$) Sample 2, ($c$) Sample 3. In each panel the origin corresponds to views down the short axis, the top edge to views down the long axis, and the right edge to views down the middle axis. For these models views are uniquely defined over only 1 octant, which is shown in equal-area projection with dotted lines at $30\\arcdeg$ intervals. \\label{f.orientations}} \\end{figure}  At any rate, bias in the 6-object samples, even if present, would not account for the similar systematic error seen in the single-object tests of \\S\\ \\ref{s.single}. Unfortunately, we see no obvious reason why the method should tend to underestimate the triaxiality and overestimate the flattening. But we can speculate on three possibilities:  \\begin{enumerate}  \\item {\\em Overly cautious use of the kinematic data.\\/} The observed rotation curves are averaged over a range of radii where they are well-behaved. By indulging a natural tendency to exclude radii where the data become ``weird,'' we may have reduced the signature of the very kinematic asymmetries that are the hallmarks of triaxiality.  \\item {\\em Coarse dynamical grid.\\/} \\markcite{SDS99}SDS and \\markcite{BS00}BS emphasized the importance of the ``disklike'' or ``spheroid-like'' character of the internal velocity field to the inferred shape. The model grid used here includes only the extreme cases on the assumption that they should bracket the correct result. However, we occasionally encounter situations where shapes inferred from intermediate models are more triaxial than those from either extreme (Statler \\markcite{S00}2001).  \\item {\\em Figure rotation.\\/} Pattern speeds for the merger remnants have not been measured. An addition of a solid-body component to the streaming velocity field of a triaxial object can, in principle, make it appear more axisymmetric than it actually is. This is a long-standing unresolved issue, and there are still no dynamical models for ellipsoidal systems that include figure rotation as an adjustable parameter.  \\end{enumerate}  }  We have tested the methods developed in previous papers for inferring the intrinsic shapes of elliptical galaxies, using a set of six simulated objects formed by group mergers of disk/halo systems. We have modeled two of the objects over a uniform grid of orientations, and find that their true shapes are recovered to within the statistical errors. A more stringent test, using the prior knowledge that the same object is being observed repeatedly, indicates a small systematic bias in the results, in that both triaxiality and short-to-long axis ratio are underestimated by roughly $0.1$. We have also constructed three simulated samples of 6 objects, and estimate the parent shape distribution in each case. The result for one of the samples is statistically unimpeachable, with the correct fractions of the sample falling within the 68\\% and 95\\% HPD regions. The other two samples give results that are worse by an amount consistent with the above bias. Including prior knowledge of the internal dynamics of the objects improves the results for all the samples. Examination of the posterior shape estimates for the individual objects in the samples again suggests a systematic bias $\\lesssim 0.1$ in triaxiality and flattening, in the same sense found in the earlier test. The source of this bias is not understood, but may be related to the handling of the observational data, the coarse grid of dynamical parameters in the models, or figure rotation of the test objects. As a whole, these results support the use of our methods to understand the nature of real elliptical galaxies."
    },
    "0010/astro-ph0010476_arXiv.txt": {
        "abstract": "The BeppoSAX observations of the bright Narrow--Line Seyfert 1 galaxy Ark~564 are presented along with a high quality optical spectrum taken at the 1.5m telescope at La Silla. The 0.1--10 keV X--ray spectrum is characterized by a strong soft component which is best described by blackbody--like emission with a temperature of $\\sim$ 160 eV. At higher energies a steep ($\\Gamma \\simeq$ 2.4) power--law tail is present. There is evidence of an ionized reflector in the form of an iron line and edge. We do not find significant evidence of soft X--ray features if the spectrum is modelled with a two component continuum. The optical and X--ray spectral properties support the hypothesis of a high accretion rate onto a low mass black hole. ",
        "introduction": "In recent years sensitive ROSAT and ASCA observations have revealed that a class of objects classified on the basis of their optical spectra as Narrow--Line Seyfert 1 galaxies (hereinafter NLS1) exhibits peculiar and rather extreme properties in the X--ray band. Large amplitude X-ray variability with timescales as fast as a few minutes (Boller et al. 1997) and extremely soft X--ray spectra in the ROSAT band are common among these objects (Boller et al. 1996). ASCA observations have extended the ROSAT results to higher energies. On average, NLS1 galaxies have steeper X--ray spectra (Brandt et al. 1997) and higher variability amplitude variance (Fiore et al. 1998; Leighly 1999a; Turner et al. 1999a) compared to Seyfert 1s with broad optical lines. The anti--correlation between the X--ray spectral slope and the H$\\beta$ FWHM is now a well--established observational property tested across a wide range of X--ray luminosities (Laor et al. 1997), suggesting that NLS1 represent an extreme of Seyfert activity, possibly linked to an extreme value of a fundamental physical parameter.  The ASCA observation of REJ 1034+393 can be considered a breakthrough in the understanding of NLS1, because it was the first object of this class observed above 2 keV. The 0.5--10 keV spectrum is dominated by a strong soft component below $\\sim$ 2 keV; a steep ($\\Gamma \\simeq$ 2.6) power law tail at higher energies and an ionized iron K$\\alpha$ line at 6.7 keV (although detected only at the 2$\\sigma$ level) are also present (Pounds et al. 1995). Ionized iron lines have been discovered in a few other objects observed by BeppoSAX (Comastri et al. 1998, paper I) and ASCA (Turner et al. 1998; Leighly 1999b), suggesting that the fundamental parameter could be a particularly high accretion rate (Pounds et al. 1995; Laor et al. 1997).  We have started an observational programme to study a sizeable sample of NLS1 with BeppoSAX with the aim of investigating the broad band X--ray spectral and variability properties of these objects. We are taking advantage of the increased effective area at high energies, especially above 5 keV, where the MECS detectors provide the best opportunity to investigate the iron line properties.  Arakelian 564 ($z$ = 0.0247, $V$ = 14.16) is an ideal candidate for such an investigation being the brightest NLS1 in the 2--10 keV band. Previous ROSAT and ASCA observations revealed a complex X--ray spectrum with a few absorption/emission features around 1 keV (Brandt et al. 1994; Leighly 1999b; Vaughan et al. 1999a; Turner et al. 1999b, hereinafter TGN99). More recently, an RXTE/PCA observation (Vaughan et al. 1999b, hereinafter V99) provided the first spectrum of this object up to 20 keV. The most remarkable result is the presence of a deep ionized edge at $\\sim$ 8.5 keV interpreted as reflection from an ionized disc. In the present paper we report the results of 3 BeppoSAX observations of Ark 564 together with a high--quality optical spectrum obtained at La Silla one year before the first X--ray observation. ",
        "conclusions": "We have analysed BeppoSAX data and a high-quality optical spectrum of the bright Narrow-Line Seyfert~1 galaxy Ark~564. Our main results are the following:  $\\bullet$ The optical emission--line intensities and widths clearly confirm the NLS1 nature of Ark~564. The Fe~{\\sc ii} multiplets are not particularly strong and are similar to those of normal Seyfert 1 galaxies.  $\\bullet$ The evidence for soft X--ray spectral features is marginal if the strong and possibly complex soft X--ray excess is modelled with a blackbody or a multicolour accretion disc spectrum.  $\\bullet$ The spectrum hardens at high energies though the power law photon index is rather steep and significantly steeper than the average value measured in broad line Seyfert 1 galaxies.  $\\bullet$ The high--energy iron K$\\alpha$ spectral features (the $\\sim$ 6.8 keV line and the $\\sim$ 9.5 keV edge) strongly suggest reflection from a highly ionized, optically thick accretion disc.  $\\bullet$ The optical to X--ray spectral energy distribution peaks in the soft X--ray band with a behaviour similar to that observed in RE J1034+393. A rapidly accreting, low mass black hole provides the most likely interpretation of the broad band observations."
    },
    "0010/astro-ph0010126_arXiv.txt": {
        "abstract": "We compute precise predictions for the two-point correlation function of local maxima (or minima) in the temperature of the microwave background, under the assumption that it is a random gaussian field. For a given power spectrum and peak threshold there are no adjustable parameters, and since this analysis does not make the small-angle approximation of \\scite{HS99}, it is essentially complete.  We find oscillatory features which are absent in the temperature autocorrelation function, and we also find that the small-angle approximation to the peak-peak correlation function is accurate to better than 0.01 on all scales.  These high-precision predictions can form the basis of a sensitive test of the gaussian hypothesis with upcoming all-sky microwave background experiments MAP and Planck, affording a thorough test of the inflationary theory of the early Universe.  To illustrate the effectiveness of the technique, we apply it to simulated maps of the microwave sky arising from the cosmic string model of structure formation, and compare with the bispectrum as a non-gaussian discriminant.  We also show how peak statistics can be a valuable tool in assessing and statistically removing contamination of the map by foreground point sources. ",
        "introduction": "The cosmic microwave background radiation (CMB) presents an ideal opportunity to test theories of the early Universe.  At the time of last scattering, the Universe is a relatively straightforward, almost uniform, mixture of photons, baryons, electrons and dark matter.  The physics is well-understood, and free from the very complicated effects which make interpretation of the present-day matter distribution more complicated.  The microwave background thus offers the possibility of accurately testing models of structure formation.  A generic test can readily be made between two classes of structure-formation models, based on inflation and cosmic defects respectively.  There are several ways to do this; the power spectrum itself is a useful discriminant of specific models.  We concentrate here on a generic test: most inflationary models predict that the microwave background temperature map will be very close to a random gaussian field, whereas generically defect models predict a non-gaussian temperature map.  It turns out that testing the gaussian nature of the initial fluctuations is easier through analysis of CMB fluctuations than large-scale structure \\cite{VWHK00}, although tests based on number densities of high-redshift objects may also be useful (\\pcite{Robinson00},\\pcite{MVJ00}).  Current evidence from Boomerang \\cite{Boomerang} and MAXIMA (\\pcite{MAXIMA1},\\pcite{MAXIMA2}) favours inflation models, since the power spectrum is acceptable for certain combinations of cosmological parameters.  Indeed, the major scientific goal of these and future experiments such as the {\\em Microwave Anisotropy Probe} (MAP) and {\\em Planck Surveyor} \\cite{PhaseA}, is to derive cosmological parameters from the power spectrum.  To make this interpretation requires that the temperature map is created by inflation or some similar process, not by defects, and that the map is not seriously contaminated by foregrounds.  In both of these areas, the statistics of peaks can be a valuable tool.  The process is quite straightforward: given a power spectrum, the statistical properties of peaks of a gaussian field are fully determined - there are no free parameters.  If the peaks are not consistent with the predictions, then either the CMB temperature map is not gaussian, or it is significantly contaminated by foregrounds, or both.  In either of these cases, the derived cosmological parameters from the power spectrum will be suspect.  In this paper, we compute the predictions for the correlation function of local maxima (and minima) for a gaussian field.  The paper generalises the work of \\scite{HS99} in dropping the small-angle approximation: the results of this paper can be used for all valid separations on the sky.  There are several ways to test the gaussian hypothesis, such as the three-point function (e.g. \\pcite{Hinshaw94}, \\pcite{FRS93}, \\pcite{LS93}, \\pcite{GLMM94}), the genus and Euler-Poincar\\'e statistic (\\pcite{Coles88}, \\pcite{Gott90}, \\pcite{Luo94b}, \\pcite{Smoot94}), the bispectrum (\\pcite{Luo94}, \\pcite{Hea98}, \\pcite{Ferreira98}), studies of tensor modes in the CMB \\cite{CCT94}, excursion set properties (\\pcite{Barreiro98}, \\pcite{Barreiro2000}), peak statistics (\\pcite{BE87}, \\pcite{Kogut95}, \\pcite{Kogut96}, \\pcite{Barreiro97}) and wavelet analyses (e.g. \\pcite{MHL00}, \\pcite{AF99}, \\pcite{FA99}).  One advantage which the method presented here has is the possibility of assessing and removing contamination by foreground point sources.  We return to this in the discussion. Non-gaussian signals have been reported for the COBE map by \\scite{Ferreira98} (see also \\pcite{Pando98}, \\pcite{KJ98}, \\pcite{BT99}, \\pcite{MHL00}, \\pcite{Mag2000}) .  If this nongaussian signal is really present in the microwave background map, and not the result of some artefact \\cite{Banday2000}, then it would be a severe challenge to inflation models, as it is many orders of magnitude larger than expected (e.g. \\pcite{VWHK00} and references therein). ",
        "conclusions": "We have presented calculations of the exact correlation function of peaks in a random gaussian field defined on the surface of a sphere.  No small-angle approximation is made, so the method is an advance on the flat-sky computations of \\scite{HS99} and now effectively complete. The formalism allows very accurate theoretical predictions of the peak-peak correlation function for temperature fluctuations in the microwave background, which is the application considered here. We envisage the main use of this method being as a sensitive test of the gaussian hypothesis. Since inflationary models generically predict a temperature field which is very close to gaussian, this is a consistency test for inflation. Other structure formation models, based for example on strings, predict non-gaussian temperature maps.  Although the visual appearance of string maps is evidently non-gaussian, it is not necessarily easy to find statistics which will unambiguously distinguish them from gaussian fields.  To illustrate this point, we have analysed 12.5-degree square simulated maps of string models, using the bispectrum and the peak-peak correlation function as distinguishing statistics.  We find that, while cosmic variance in the bispectrum makes it difficult to use on a small patch of sky, the peak-peak correlation function clearly rules out a gaussian map.  In practice, maps of the microwave background will be contaminated at some level by point sources, amongst other things.  Peak statistics may be useful in assessing this contribution.  The most straightforward example is that an uncontaminated map has the same average number density of maxima and minima; a significant excess of maxima would be indicative of contamination.  Unfortunately the theory of peaks is not able to tell us the {\\em distribution} of the number of maxima or minima within a finite sky (only its mean), but it is a straightforward matter to determine the distribution by monte carlo realisations.  One can attempt to go further than this, by removing statistically the contribution from the point sources, provided one knows from other observations what their correlation function is. Assuming the point sources are uncorrelated with the microwave background peaks, the correlation function of the combined map is simply a weighted mean of the two.  The point sources will contribute to the power spectrum; one can vary the assumed contribution from point sources and modify the power spectrum and the derived microwave background peak correlation function accordingly.  If consistency can be achieved, one will be confident both of the gaussian nature of the microwave background, and the level of point source contamination.  {\\bf Acknowledgments}  We are grateful to Francois Bouchet for providing the simulated catalogues, and to Ravi Sheth for useful discussions. Computations were made partly using Starlink facilities."
    },
    "0010/astro-ph0010310_arXiv.txt": {
        "abstract": "We present a first emission line analysis of a high resolution X-ray spectrum of the stellar wind of $\\theta^1$ Ori C obtained with the High Energy Transmission grating Spectrometer onboard the Chandra X-ray Observatory. The spectra are resolved into a large number of emission lines from H- and He-like O, Ne, Mg, Si, S, Ar and Fe ions. The He-like Fe XXV and Li-like Fe XXIV appear quite strong indicating very hot emitting regions. From H/He flux ratios, as well as from Fe He/Li emission measure ratios we deduce  temperatures ranging from 0.5 to 6.1$\\times 10^7$ K. The He-triplets are very sensitive to density as well. At these temperatures the relative strengths of the intercombination and forbidden lines indicate electron densities well above 10$^{12}$ cm$^{-3}$. The lines appear significantly broadened from which we deduce a mean velocity of 770 km s$^{-1}$ with a spread between 400 and 2000 km s$^{-1}$. Along with results of the deduced emission measure we conclude that the X-ray emission could originate in dense and hot regions with a characteristic size of less then 4$\\times 10^{10}$ cm. ",
        "introduction": "X-ray spectra from stellar winds have been quite difficult to interpret, since the first observations with \\ein (Seward et al. 1979). The spectral resolution from previous X-ray missions was insufficient for resolving any emission and absorption lines. Models that describe the mass loss from a hot luminous star successfully in the UV (Pauldrach et al. 1994, Lamers et al. 1999) could never correctly predict observed X-ray fluxes. A model involving a possible hot corona seems unlikely due to the lack evidence for absorption at the 20.6 \\AA~ K-shell ionization edge from oxygen (Cassinelli $\\&$ Swank 1983); additionally no coronal line emission (i.e. Fe XIV at 530 nm) was found in optical spectra of $\\eta$ Ori and $\\kappa$ Ori (Nordsieck et al. 1981) nor in the protoptype star for stellar winds $\\zeta$ Pup (Baade $\\&$ Lucy (1987). X-ray emission from shocks emerging in instabilities within a radiatively driven wind forming a forward shock was proposed by Lucy $\\&$ White (1980) and  Lucy (1982). However, reverse shocks that decelerate material as it rams into dense shells ahead seem more likely (Owocki et al. 1988). These models predict X-rays up to a temperature of about kT$\\sim$0.5 keV. Feldmeier et al. (1997) introduced turbulent perturbations into their calculations which ultimatively allowed for higher temperatures.  Several recent observations with the  0.1 to 10.0 keV \\asca bandpass, i.e. of $\\tau$ Sco (Cohen et al. 1997) and $\\eta$ Car (Tsuboi et al. 1997), indicated that there is a hard component to these spectra at $>$ 2 keV. Although these two examples tend to resemble quite extreme cases of massive stars, \\asca observations of other, less massive O-stars (Corcoran et al. 1994), and the Orion Trapezium (Yamauchi et al. 1996) indicated the existence of a similar hard component in their spectra as well. In the most recent analysis of \\chandra CCD spectra of the Orion Trapezium, Schulz et al. (2000) were able to resolve the entire Trapezium into individual sources. Identified O- and B-stars showed a soft component at $\\sim$ 0.8 keV as observed with \\ros; the very early spectral types required an additional  hard component of temperatures above 2 keV as well.  In this letter we present first results from an observation of $\\theta^1$ Ori C with the High Energy Transmission Grating Spectrometer (HETGS, Canizares et al. 2000, in preparation) onboard \\chandra (Weisskopf et al. 1996). We present an X-ray line list of the brightest emission lines detected and investigate emission line properties to determine the range of temperatures and densities of the line emitting regions. Some of the results will have immediate impact constraining models of stellar winds.  \\centerline{\\epsfxsize=7.5cm\\epsfbox{f1.ps}} \\figcaption{HETGS focal plane image of the Orion Trapezium Cluster \\label{focal}} ",
        "conclusions": "A large number of emission lines were detected with a signal to noise ratio of greater than 5$\\sigma$. These lines are from H- and He-like ion species of Fe, S, Si, Mg, Ne, and O; Ca and Ar are not detected. We do not observe any residual line shifts we could attribute to the source.  The X-ray lines appear significantly broadened implying a mean shock velocity of 770 km s$^{-1}$. A similar shock velocity of 500 km s$^{-1}$ has been determined from \\ros fits of $\\zeta$ Pup spectra (Hillier et al. 1993). Within the statistical uncertainities and the current state of the analysis, the line profiles appear symmetric showing blue and red-shifts of equal proportions. We therefore conclude that the wind emission is quite isotropic. Our estimates from the VEM indicates that the emission volume is equivalent to a volume of well within one stellar radius. However, Simulations by Feldmeier et al. (1997) indicated that the X-ray emission must extend far out into the wind. Estimating the distance of the X-ray producing shell collision we use the $\\beta$ velocity law (with $\\beta$= 0.88) and a terminal velocity of $\\sim$ 1000 km s$^{-1}$ (Prinja et al. 1990) to determine the distance where the wind reaches 770 km s$^{-1}$, which gives about 4 stellar radii. In this respect it is plausible that the X-ray emitting regions appear in very thin dense shells as described by Feldmeier et al. (1997). Their model, however, predicts strong variability of the X-ray emission, which, at least in terms of total X-ray flux, is not observed during our observations.  Whatever the actual emission geometry turns out to be, the emitting regions of $\\theta^1$ Ori C appear well confined, hot, and dense with a temperatures ranging between 0.05 and 6 $\\times 10^7$ K (0.06 to 4.5 keV) and densities above $10^{12}$ cm$^{-3}$. CCD spectra of $\\theta^1$ Ori C already indicated that the X-ray emission must have a hot component with a temperature of kT$\\sim$3 keV (Yamauchi et al. 1997, Schulz et al. 2000) or equivalent to $\\sim$ 4$\\times 10^7$ K. A broadband study ranging from UV to \\asca spectra of the B0 V star $\\tau$ Sco Cohen (1996) also suggested that wide range of temperatures up to $10^7$ K is necessary to describe X-ray and UV emission. It is quite hard to explain such high temperature with the line driven shock instability models (Lucy $\\&$ White 1980, Lucy 1982), which predict temperatures more on the order of 5$\\times 10^6$ K. The spectrum in Figure 2 also shows that the hard part of the spectrum cannot be explained by inverse Compton scattering of UV photons (Chen $\\&$ White 1991), since it shows emission lines with similar characteristics than the low energy lines. The approach by Feldmeier et al. (1997) seems to provide promising ingredients to describe the co-existence of low and high temperatures at relatively high densities. A most recent analysis by Kahn et al. (2000) of XMM RGS spectra of $\\zeta$ Pup also resulted in low f/i line ratios and thus high density values, but it was suggested that a high UV radiation field could destroy the He-like forbidden line leading to an overestimation of the density.  It still remains to be explained whether the high temperatures in $\\theta^1$ Ori C and other hot candidates are the exception or the rule. Clearly, we need more high resolution X-ray observations of massive O-Stars in order classify the X-rays in terms of their line emission properties. These properties include temperature ranges, densities, emission volumes and ionization balance."
    },
    "0010/astro-ph0010583_arXiv.txt": {
        "abstract": "A Virtual Observatory (VO) will enable transparent and efficient access, search, retrieval, and visualization of data across multiple data repositories, which are generally heterogeneous and distributed.   Aspects of data mining that apply to a variety of science user scenarios with a VO are reviewed. ",
        "introduction": "What is data mining and why is applicable to scientific research?  Data mining is defined as {\\it{an information extraction activity whose goal is to discover hidden facts contained in databases}}.  Data mining has taken the business community by storm and there is consequently now a vast array of resources and research techniques available for exploitation by the scientific communities.  It is useful therefore to examine a categorization of data mining thrusts and their sub-components, since these are likewise applicable to the scientific exploration of large astronomical databases.  Data mining is used to find patterns and relationships in data by using sophisticated techniques to build models -- abstract representations of reality. A good model is a useful guide to understanding that reality and to making decisions.  There are two main types of data mining models:  {\\it{descriptive}} and {\\it{predictive}}.  {\\it{Descriptive}} models describe patterns in data and are generally used to create meaningful subgroups or clusters.  {\\it{Predictive}} models are used to forecast explicit values, based upon patterns determined from known results.  There is another differentiation of data mining into two categories that we find particularly appropriate to knowledge discovery in large astronomical databases:  {\\it{event-based mining}} and {\\it{relationship-based mining}}. At the risk of trivializing some fairly sophisticated techniques, we classify event-based mining scenarios into four orthogonal categories:  \\begin{itemize}  \\item Known events / known algorithms -- use existing physical models ({\\it{descriptive models}}) to locate known phenomena of interest either spatially or temporally within a large database.  \\item Known events / unknown algorithms -- use pattern recognition and clustering properties of data to discover new observational (in our case, astrophysical) relationships among known phenomena.  \\item Unknown events / known algorithms -- use expected physical relationships ({\\it{predictive models}}) among observational parameters of astrophysical phenomena to predict the presence of previously unseen events within a large complex database.  \\item Unknown events / unknown algorithms -- use thresholds to identify transient or otherwise unique (``one-of-a-kind'') events and therefore to discover new phenomena.  \\end{itemize}  \\noindent For relationship-based mining, we identify three classes of association-driven scenarios that would find application in astronomical research:  \\begin{itemize}  \\item Spatial associations -- identify events (astronomical objects) at the same location in the sky.  \\item Temporal associations -- identify events occurring during the same or related periods of time.  \\item Coincidence associations -- use clustering techniques to identify events that are co-located within a multi-dimensional parameter space.  \\end{itemize}  From this discussion, we thus derive a reduced set of science requirements for data mining that correspond to the following exploratory approaches to mining large databases\\thinspace : {\\it{Object Cross-Identification}}, {\\it{Object Cross-Correlation}}, {\\it{Nearest-Neighbor Identification}}, and {\\it{Systematic Data Exploration}}. (a)~``Object cross-identification'' refers to the classical problem of connecting the source list in one catalog (or observation database) to the source list in another, in order to derive new astrophysical understanding of the cross-identified objects (e.g., gamma-ray burst counterparts). (b)~``Object cross-correlation'' refers to the application of ``what if'' scenarios to the full suite of parameters in a database (e.g., identify distant galaxies as $U$-band dropouts in a color-color scatter plot from the HDF survey). (c)``Nearest-neighbor identification'' refers to the general application of clustering algorithms in multi-dimensional parameter space (e.g., finding the closest known population of young stars -- in the TW Hydrae association -- through their similar kinematics, X-ray emission, $H\\alpha$, and Li abundance). (d)~``Systematic data exploration'' refers to the application of the broad range of event-based and relationship-based queries to a database in the hope of making a serendipitous discovery of new objects or a new class of objects (e.g., finding new types of variable stars, such as ``bumpers'', in the MACHO database). ",
        "conclusions": ""
    },
    "0010/astro-ph0010060_arXiv.txt": {
        "abstract": "Results are presented from a long-term astrometry program on PSR B0919+06 using the NRAO Very Long Baseline Array. With ten observations (seven epochs) between 1994--2000, we measure a proper motion $\\mu_{\\alpha} = 18.35 \\pm 0.06$ mas yr$^{-1}$, $\\mu_{\\delta} = 86.56 \\pm 0.12$ mas yr$^{-1}$, and a parallax $\\pi = 0.83 \\pm 0.13$ mas (68\\% confidence intervals). This yields a pulsar distance of $1.21 \\pm 0.19$ kpc, making PSR B0919+06 the farthest pulsar for which a trigonometric parallax has been obtained, and the implied pulsar transverse speed is $505 \\pm 80$ \\kms. Combining the distance estimate with interstellar scintillation data spanning 20 years, we infer the existence of a patchy or clumpy scattering screen along the line of sight in addition to the distributed electron density predicted by models for the Galaxy, and constrain the location of this scattering region to within $\\sim 250$ parsecs of the Sun.  Comparison with the lines of sight towards other pulsars in the same quadrant of the Galaxy permits refinement of our knowledge of the local interstellar matter in this direction.  \\\\ ",
        "introduction": "Since the discovery of pulsars, it has been recognized that the dispersion and scattering of pulsar signals provide unique information about the intervening medium. The distance to most pulsars is estimated using the observed dispersion measure (DM) and a model for the Galactic electron density distribution \\citep[e.g.][hereafter TC93]{TC93}. Where available, a model-independent distance from annual trigonometric parallax provides crucial calibration information for this model, as well as allowing the absolute luminosity of the pulsar to be derived.  Pulsar astrometry also provides the observational evidence required to investigate several other questions. Proper motion measurements (especially in conjunction with reliable distance estimates) allow verification of pulsar--supernova remnant associations \\citep[e.g.][]{K98}. The inferred speeds of pulsars constrain the minimum asymmetry in supernova core-collapse processes, or other sources of kick velocities. Pulsar population statistics, selection-effect biases, and planetary--extragalactic reference frame ties have also been addressed by astrometry \\citep[e.g.,][]{LL94,CC98,BCR+96}.  At present, there are only a handful of model-independent distances to pulsars. In a recent summary, \\citet{TBM+99} list nine such objects. In \\S\\ref{Sec:VLBI} of this paper, we present results from a long-term astrometry program on PSR B0919+06 using the NRAO Very Long Baseline Array (VLBA). We measure the proper motion and trigonometric parallax for this pulsar, and infer a distance of $\\sim 1.2$ kpc and a transverse speed of $\\sim 500$ \\kms, comparable to the mean population speed.  PSR B0919+06 has a DM of 27.31 pc cm$^{-3}$ \\citep{PW92}, yielding a mean electron density of 0.023 cm$^{-3}$ for the interstellar medium (ISM) along this line of sight. However, interstellar scintillation (ISS) and scattering of the pulsar signals can be used to extract much more information about the line of sight than simply this mean electron density. In \\S\\ref{Sec:ISS}, we review the necessary formalism, and analyze the published data spanning $\\sim 20$ years for this pulsar. Our analysis combines data from very long baseline interferometry (VLBI) and ISS, in order to infer the distribution of scattering material along the line of sight. We find the need for extra scattering material in addition to the TC93 distribution; with the assumption that this material is present in the form of a screen or clump with variable scattering strength, we constrain the scren location to the local interstellar medium, within $\\sim 250$ parsecs of the Sun.  This analysis, along with DM and distance measurements for other pulsars in the third quadrant of our Galaxy, allows refinement of our knowledge of the local electron density distribution. Besides being a situation of interest in its own right, this analysis also serves as a model for the hybrid VLBI--ISS technique applied to the local ISM. ",
        "conclusions": "Parallax distances to pulsars have the prospect of resolving several outstanding questions, both about the pulsar population and about the intervening ISM. In this work, we have presented VLBA astrometry on PSR B0919+06, using phase-referenced observations with an in-beam calibrator only 12\\arcmin\\ from the pulsar. The use of the in-beam source reduces ionospheric effects and other astrometric errors to the 0.1 mas level and allows sub-\\mas\\ astrometry.  With ten observations (seven distinct epochs) distributed over seven years, we derive a proper motion $\\mu_{\\alpha} = 18.35 \\pm 0.06$ mas yr$^{-1}$, $\\mu_{\\delta} = 86.56 \\pm 0.12$ mas yr$^{-1}$, and measure a parallax $\\pi = 0.83 \\pm 0.13$ mas (68\\% confidence intervals). This result implies a transverse speed of $505 \\pm 80$ \\kms\\ at a distance $D = 1.21 \\pm 0.19$ kpc, making PSR B0919+06 one of the farthest objects for which a trigonometric parallax has been obtained. The use of an in-beam calibrator with the VLBA at 1.5~GHz can provide trigonometric parallaxes to twice this distance, or even further (to 10~kpc) with the use of more sensitive antennas and higher frequencies.  For the measured distance of 1.2~kpc and a DM of 27.31 pc~cm$^{-3}$ \\citep{PW92}, the derived mean electron density towards B0919+06 is 0.023 cm$^{-3}$. This line of sight is along the same Galactic longitude as the superbubble proposed by \\citet{H98}, as well as clouds of ionized gas in the local neighbourhood \\citep{DG98}, though at a higher Galactic latitude. We use scintillation data spanning 20 years in a hybrid analysis, equating the derived ISS and interferometric speeds of the pulsar in order to constrain the distribution of scattering material along the LOS. We find that excess scattering material is required compared to the standard TC93 model, and assuming that this excess material is confined to a thin screen or clump, we constrain its location to within $\\sim 250$~pc of the Sun. Comparison with the neighboring lines of sight to pulsars B0950+08, J1024$-$0719, and B0823+06 indicates that this result is consistent with a turbulent interface region between the Local Bubble and the GSH~238+00+09 superbubble, which possibly contains clumps of scale size $< 30$~pc.  We note that using an updated model of the Galactic electron density distribution (Cordes {\\&} Lazio, in preparation) will not materially alter these conclusions."
    },
    "0010/astro-ph0010256_arXiv.txt": {
        "abstract": "We propose and implement a fast, universally applicable method for extracting the angular power spectrum ${\\cal C}_{\\ell}$ from CMB temperature maps by first estimating the correlation function $ \\xi(\\theta)$. Our procedure recovers the ${\\cal C}_{\\ell}$'s using $N^2$ (but potentially $N \\log N$), operations, where $N$ is the number of pixels. This is in contrast with standard maximum likelihood techniques which require $N^3$ operations. Our method makes no special assumptions about the map, unlike present fast techniques which rely on symmetries of the underlying noise  matrix, sky coverage, scanning strategy, and geometry. This enables for the first time the analysis of megapixel maps without symmetries. The key element of our technique is the accurate multipole decomposition of $\\xi (\\theta)$.  The ${\\cal C}_{\\ell}$ error bars and cross-correlations are found by a Monte-Carlo approach.  We applied our technique to a large number of simulated maps with Boomerang sky coverage in $81000$ pixels. We used a diagonal noise matrix, with approximately the same amplitude as Boomerang. These studies demonstrate that our technique provides an unbiased estimator of the ${\\cal C}_{\\ell}$'s. Even though our method is approximate, the error bars obtained are nearly optimal, and converged only after few tens of Monte-Carlo realizations. Our method is directly applicable for the non-diagonal noise matrix. This, and other generalizations, such as minimum variance weighting schemes, polarization, and higher order statistics are also discussed. ",
        "introduction": "Future missions of measuring the Cosmic Microwave Background (CMB) fluctuations will revolutionize our knowledge of cosmology. They will either confirm or refute the basic Big Bang paradigm, will reveal the values of most cosmological parameters within a few percent (e.g., Spergel 1994, Knox 1995, Hinshaw, Bennett, \\& Kogut 1995, Jungman \\etal 1996, Zaldarriaga, Spergel, \\& Seljak 1997, Bond, Efstathiou, \\& Tegmark 1997). The large number of pixels contained in current and future experiments enable these exciting developments, but at the same time, present unprecedented challenges the data analysis. The mainstream maximum likelihood techniques are already pushed to their limits by the largest existing CMB maps, but they are clearly inadequate for future megapixel surveys. Therefore the most important near term task for CMB research is to find techniques which could perform the required analyses with realistic resources, thus fulfill the promise of the high precision experiments.  This Letter proposes a fast, universally applicable method for the estimation of $C_l$'s from any CMB pixel map, based on estimating correlation functions. The recovered errorbars are at most $10\\%$ larger than the theoretically smallest possible ones imposed by cosmic variance and noise. Optimal methods could further decrease these errorbars only slightly, and at an unrealistic cost; thus they  represent a case of diminishing returns. Our procedure scales as $N^2$, and more sophisticated algorithms will improve this to $N \\log N$. This essentially solves the problem for two major upcoming space missions, MAP (Microwave Anisotropy Probe) and the Planck, and at the same time allows more detailed analyses of current and future balloon-borne and ground based experiments. The short analysis turn around time  enables Monte-Carlo (MC) studies of systematics, foregrounds, errors, underlying models etc.  These are essential to the interpretation of the data but inaccessible to present methods.  The standard maximum likelihood methods for analyses were developed and tested on COBE measurements (e.g., G\\'orski 1994, G\\'orski \\etal 1994, 1996, Bond 1995, Tegmark \\& Bunn 1995, Hinshaw \\etal 1996, Tegmark 1996, Bunn \\& White 1996, Bond, Jaffe, \\& Knox 1998, hereafter BJK98, 2000) which have only about $N \\simeq 1000$ pixels. Balloon-borne experiments, such as Boomerang (de Bernardis \\etal 2000), Maxima (Hanany \\etal 2000),  and Tophat (Martin \\etal 1996),  and ground based measurements e.g., TOCO (Miller \\etal 1999) and Viper (Peterson \\etal 2000) with up to $N \\simeq 10^5$ are already pushing present day supercomputer technology. The future missions MAP and Planck with $N = 10^6-10^7$ are estimated to require up to millions of years (Borrill 1999, Bond, Crittenden, Jaffe, \\& Knox 2000) for one iteration of the quadratic maximum likelihood estimator for $C_l$'s. The disk storage and memory usage of these algorithms are prohibitive as well.  To date fast algorithms were presented under two fairly restrictive assumptions: i) the noise is both temporally uncorrelated and spatially axially symmetric ii) the foregrounds can be exactly removed both from the map and the correlation matrix (Oh, Spergel, \\& Hinshaw 1999, Wandelt, Hivon \\& Gorski 1998, 2000). Our approach is significantly different from these, since it does not assume any symmetries. ",
        "conclusions": "We presented a novel method to extract  ${\\cal C}_\\ell$'s from large CMB maps via two-point correlation functions. We have cast our determination of ${\\cal C}_\\ell$ into 3 steps: fine-grained $C(\\theta)$ estimation, the pixel-related $4^\\prime$ Gaussian smoothing of it, followed by $P_\\ell$ multiplication and Gauss-Legendre integration. The resulting expression for ${\\cal C}_\\ell$ can be itself understood as  a sum over pixel pairs, with a pair-weighting proportional to the Gaussian smoothing of $P_\\ell$. In this respect our estimator belongs to the class of quadratic estimators for ${\\cal C}_\\ell$, of which other examples (e.g., BJK98) have been used in the past and are expected to give similar results.   In the present example the measured errorbars were at most $10$\\% larger then the theoretically minimal ones; this negligible suboptimality allowed the reduction of the CPU time from months (Borrill 1999, Table 3) to hours. For future large missions the difference will be even more dramatic: optimal methods would improve the errorbars only slightly at a prohibitive cost.  Our new approach has several advantages in comparison to previous techniques. It scales as $N^2$ even in its most straightforward implementation, compared to typical quadratic estimators which scale as $N^3$. This scaling can be further improved up to $N\\log N$. While in the present implementation we used diagonal noise, no other (e.g., asymuthal) symmetries were assumed about noise, or geometry of the map. Cut out holes around bright sources, galactic cut, any irregularity in the sampling, make no difference in the speed or performance of the algorithm. To illustrate this, Equation~\\ref{eq:estimator} was used to determine the correlation function from COBE DMR data, which has inhomogeneous noise. The recovered $C_l$'s are consistent with those obtained from implementations of  quadratic estimators discussed in BJK98. In contrast with any other previous attempt, our approach is straightforward to generalize for non-diagonal noise matrix (see below).  While our method in its present form is already practical for analyzing megapixel CMB maps, it has  the potential for further generalizations. General non-diagonal noise appears to pose a serious problem. Fast iterative map making methods (Wright \\etal 1996, Prunet \\etal 2000)  capable of handling large data sets only furnish the weight matrix, $w_{ij} = N_{ij}^{-1}$. Since inversion of the noise matrix is again an $N^3$ problem, our estimator of Equation~\\ref{eq:estimator} might not be directly applicable. Instead, we propose a new estimator using MC realizations of artificial noise \\begin{equation} \\tilde\\xi(\\cos\\theta) = \\sum_{ij} f_{ij} (\\Delta_i \\Delta_j - \\frac{1}{M}\\sum_{k=1}^M n_i^k n_j^k), \\label{eq:modestimator} \\end{equation} where $n_i^k$ is one of $M$ realizations of the noise for pixel $k$. This plays the role of a MC inversion of the weight matrix $w_{ij} = N_{ij}^{-1}$. Generation of $n_i^k$ in the time domain, where it has simple correlation structure, is straightforward. Its iterative projection into pixel space is equivalent to map-making. This is feasible even when storing the noise-matrix would be prohibitive, as for Planck.  To further improve the performance of our method, we will implement a  minimum variance weighting scheme  (Feldman Kaiser, \\& Peacock 1994, and Colombi, Szapudi, \\& Szalay 1998). This corresponds to down-weighting measurements with their variance, and might be important in maps, where various pairs contributing to the correlation function have widely differing errors. Otherwise, the uniform weighting scheme is nearly optimal (Colombi, Szapudi, \\& Szalay 1998). The present simple weighting scheme can be improved heuristically via individual pixel weighting reflecting differences in sky coverage. More complex pair weighting can be defined iteratively, using the MC estimates of the variances.  The present Letter used a simple $N^2$ code to calculate the correlation function, and this is perfectly adequate for LDB surveys with $N\\simeq 10^5$ pixels, and, with supercomputers, even for MAP. Nevertheless, the development of an $N\\log N$ code is under way (Szapudi \\& Colombi 2000, Connolly, Nichol, Moore, \\& Szapudi 2000).  Our technique has reestablished the utility of correlation functions for CMB studies. This approach has further potential applications: it is naturally generalizable for the assessment of non-Gaussianity in CMB maps via $k$-point correlation functions, with implementations as fast as $N(\\log N)^{k-1}$ (Sunyaev-Zeldovich effect, lensing of CMB); it is equally useful for polarization correlation functions and for obtaining the appropriate $C_l$'s from them. We have found that the statistical information is condensed into singular features of the correlation function, (see also Bashlinsky \\& Bertschinger, in prep.). This suggests that direct parameter estimation from the CMB two-point correlation function might be fruitful as well.    Other applications include correlations of the infrared (SCUBA, BLAST, SIRTIF, FIRST) and optical background, and weak gravitational lensing."
    },
    "0010/astro-ph0010604.txt": {
        "abstract": "We use a multi-color classification method introduced by Wolf, Meisenheimer \\& R\\\"oser (2000) to reliably identify stars, galaxies and quasars in the up to 16-dimensional color space provided by the filter set of the Calar Alto Deep Imaging Survey (CADIS). The samples of stars, galaxies and quasars obtained this way have been used for dedicated studies which are published in separate papers.  The classification is good enough to detect quasars rather completely and efficiently without confirmative spectroscopy. The multi-color redshifts are accurate enough for most statistical applications, e.g. evolutionary studies of the galaxy luminosity function. Also, the separation between stars and galaxies reaches deeper than with morphological criteria, so that studies of the stellar population can be extended to fainter levels.  We characterize the dataset presently available on the CADIS 1h-, 9h- and 16h-fields. Using Monte-Carlo simulations we model the classification performance expected for CADIS. We present a summary of the classification results on the CADIS database and discuss unclassified objects. More than 99\\% of the whole catalog sample at $R<22$ (more than 95\\% at $R<23$) are successfully classified matching the expectations derived from the simulations. A small number of peculiar objects challenging the classification are discussed in detail.  Spectroscopic observations are used to check the reliability of the multi-color classification (6 mistakes among 151 objects with $R<24$). From these, we also determine the accuracy of the multi-color redshifts which are rather good for galaxies ($\\sigma_z \\approx 0.03$) and useful for quasars. We find that the classification performance derived from the simulations compares well with results from the real survey. Finally, we locate areas for potential improvement of the classification.   %Black \\& white seems to be out, now that color vision is established in astronomical households. More and more stars prefer to be featured in color presentations, dressing in all thermal colors nowadays. Even black becomes fashionable, being still colorful in innovative near infrared. A brief explanation of modern color trends follows. % ",
        "introduction": "The {\\it Calar Alto Deep Imaging Survey} (CADIS) is an extragalactic key project at the Max-Planck Institut f\\\"ur Astronomie (MPIA), Heidelberg, which is aiming at two types of objectives: CADIS investigates whole samples of different object classes using statistical tools, but it also searches for individual rare and faint objects, which will be studied in detail with coming large telescopes. As a pencil beam survey, it probes seven different fields at galactic latitudes $b \\ga 45\\degr$ with a total area of $\\sim 0.25\\degr$.  The final object catalog will arise from two fundamentally different survey techniques: \\begin{itemize} \\item   a multi-color survey with B, R, J and K$^\\prime$ plus 13 medium-band filters from 400\\,nm to 1000\\,nm, practically resembling low-resolution imaging spectroscopy and giving a complete list of objects with $R \\la 23$, \\item   and an emission-line survey using an imaging Fabry-Perot interferometer to probe emission line galaxies down to a limiting line flux of $\\sim 3 \\times 10^{-20} Wm^{-2}$. \\end{itemize}  Presently, the data for three fields are reduced and have been used for a number of application studies published already or to be published this year. For many applications, objects of concern are selected by our multi-color classification which uses the many bands to sort the objects into {\\it stars}, {\\it galaxies} and {\\it quasars}.  Also, multi-color redshifts are estimated for the extragalactic objects. This classification scheme was originally developed on the basis of CADIS data, but is meanwhile used in a range of different survey activities. It uses a library of $\\sim 65000$ templates and achieves high classification reliability ($>90\\%$ correct classification in each class) and a high redshift accuracy of $\\sigma_z \\approx 0.03$ for galaxies and $\\sigma_z \\approx 0.1$ for quasars. The methodogical background for the classification was published by Wolf, Meisenheimer \\& R\\\"oser (2000), hereafter paper I. There, the classification method is derived from statistical principles, the libraries are defined, the performance expected with different filter sets is compared and where conclusions are drawn for optimum survey strategies.  The purpose of this paper is to characterize the present data of the CADIS multi-color survey and to discuss the classification performance which was checked by a subsample of objects with spectroscopic identifications. The paper is organized as follows: Section 2 lists CADIS goals for which the classification is relevant and discusses what kind of objects the classifications should be prepared for. Section 3 defines the present CADIS dataset and characterizes its photometry and calibration. Section 4 outlines the classification method and presents statistics of classified and unclassified objects. Section 5 discusses the classification performance using a spectroscopic cross check sample. Finally, Section 6 summarizes the quality of the classified catalogs and evaluates the practical performance of the CADIS multi-color classification. A few identified peculiar objects challenging the classification are discussed in the appendix. ",
        "conclusions": "In paper I an innovative method for identifying stars, galaxies and quasars in multi-color surveys was presented, which uses a library of $\\ga 65000$ color templates for comparison with observed objects. The method aims for extracting the information content of object colors in a statistically correct way, and performs a classification as well as a redshift estimation for galaxies and quasars in a unified approach based on the same probability density functions.  The three basic ingredients to this method are \\begin{enumerate} \\item  accurately measured and calibrated color data for the objects to be classified including the color errors \\item  accurate and representative color template libraries covering the range of objects expected in the dataset obtained \\item  a statistical classifier and parameter estimator which can potentially be trimmed for best performance in particular applications. \\end{enumerate}  Also, in paper I it was concluded that medium-band surveys are expected to deliver a performance superior to pure broad-band surveys even under the constraint of equally limited telescope time. Based on survey simulations this method should be capable of \\begin{enumerate} \\item  separating stars from galaxies down to deeper limits than possible by using morphology only \\item  finding quasars also at redshifts where its colors overlap with the stellar locus in two-color projections, particularly in the range of $2.2<z<3.5$ \\item  selecting quasar candidates without reference to their morphology so that objects of low-luminosity with resolved hosts would not be excluded from the sample obtained \\item  using multi-color redshifts from medium-band surveys directly and without spectroscopic follow-up for statistical investigations, e.g. for studying the evolution of galaxy populations with redshift. \\end{enumerate}  In this paper, we applied the classification scheme to a real multi-color dataset provided by the Calar Alto Deep Imaging Survey (CADIS) investigating its reliability and accuracy. The dataset is 80\\% complete on three CADIS fields presented here, the 1h-, the 9h- and 16h-field. Some filters on some fields are still missing and observations of some others have not reached the desired depth, yet. In any case, the CADIS filterset was tailored to the requirements of the emission line survey and is not an optimal choice for a general multi-color classification.  The libraries of stars, galaxies and quasars presented in paper I are sufficient to classify all but a handful of unusual objects in the CADIS dataset as the fields do not contain significant numbers of objects missing in the libraries. This is demonstrated by the low fraction ($<1\\%$) of {\\it strange} objects in the sample at $R<24$. Our classification differentiates between {\\it unclassified} objects that can not securely be decided among the alternatives and {\\it strange} objects that are outliers in color space inconsistent with any library object. We note that unclassified objects are typically not strange and vice versa.  The classified subsamples of stars and galaxies agree with expected numbers and the unusually high content of quasars at $z>2$ was confirmed by spectroscopic observations \\cite{Wolf99}. The fraction of unclassified objects is less than 1\\% at $R<22$ and reaches about 50\\% at $R=24$ due to increasing photon noise that tends to make different original SEDs equally likely sources of the observed colors. At some level in between the classification becomes incomplete and also features an increasing fraction of actual mistakes. These are basically all members of the rich galaxy population spilling over into regions of color space that is usually occupied by stars and quasars.  A spectroscopic cross check using 162 identifications confirmed the multi-color classification to work essentially free of errors at $R<22$ (two mistakes among 103 objects). We do not have proper knowledge about where and how the classification collapses exactly, which is particularly important for the rare quasar class which will become incomplete and dominated by contaminating galaxies at some level. To settle this uncertainty, dedicated spectroscopic observations are required. These would be a valuable and not too time-consuming investment, if a large number of spectra from a CADIS size field could be taken simultaneously with an instrument like VMOS at the VLT.  The findings on the classification performance are consistent with the expectations raised by the Monte-Carlo simulations of the CADIS filter set. They suggest the classification to work nearly perfect down to $R \\approx 23$ except for some contamination of quasar candidates by emission line galaxies towards that limit. The simulations actually appear to be too optimistic about the working depth of the classification by about a third of a magnitude. Of course, a real survey will perform worse than a simulation, that can not account for differences between the real world and our library and does not contain artifacts and variable or blended objects. The latter issues have to be dealt with better data analysis.  For galaxies, our multi-color redshifts are useful down to $R \\sim 24$. The statistics on the redshift errors are dominated by $\\sim 10\\%$ catastrophic mistakes, where the estimator decides for the wrong one among alternative values with comparable probability. The core of the error distribution has a zero mean error and $\\sigma_z \\approx 0.03$ rms width. Quiescent galaxies tend to work better than starburst types.  Half of the quasars receive remarkably correct estimates with an average variance of $\\sigma_z \\approx 0.1$, preferentially the $z>2$-objects and those of higher luminosity. In contrast, a large amount of redshift confusion is expected at lower redshift. Also, simulations for quasars work better than the real redshift estimation, probably mostly due to variability.  Eventually, the classified multi-color catalog has been used for several dedicated studies published in separate papers: \\begin{enumerate}  \\item The star-galaxy separation with the multi-color data reaches deeper than a morphological separation. Phleps et al. (2000) could therefore use the sample of stars to probe the stellar content and the Galactic structure along the pencil beams established by the CADIS fields, where they find strong evidence of a thick disk.  \\item The multi-color redshifts of galaxies are sufficiently accurate for most statistical studies. Also, the classification includes a substantial fraction of compact galaxies into the sample. Using that, Fried et al. (2000) investigated the evolution of galaxies within $0.3<z<1$, finding evidence that {bf strong} evolution takes place only among starburst objects.  \\item The multi-color data allow to identify quasars rather free of contaminants. Spectroscopy is not required to clean a sample obtained by the classification. The selection should also be rather complete, since CADIS finds an unexpectedly high density of $z>2$-quasars. Many of these might reside close to the stellar locus and be overlooked in broad-band surveys \\cite{Wolf99}. Thus, quasars can be identified with a much more uniform completeness across the accessible redshift range and more homogeneous samples can be obtained. At $z>2$, it will be possible to constrain the evolution of the luminosity function from a large enough multi-color sample.  \\end{enumerate}"
    },
    "0010/hep-ph0010280_arXiv.txt": {
        "abstract": "The initial condition problem of inflation is examined from the perspective of both spacetime embedding and scalar field dynamics. The spacetime embedding problem is solved for arbitrary initial spatial curvature $\\Omega$, which generalizes previous works that primarily treat the flat case $\\Omega=1$. Scalar field dynamics that is consistent with the embedding constraints are examined, with the additional treatment of damping effects. The effects of inhomogeneities on the embedding problem also are considered.  A category of initial conditions are identified that are not acausal and can develop into an inflationary regime. ",
        "introduction": "\\label{introduction} Inflation is the foremost idea for explaining the large scale homogeneity and near flatness of the universe, which are the two main unexplained observational features in the standard hot big-bang model.  The large scale homogeneity or horizon problem amounts to the fact that under hot big-bang evolution, due to a decelerating scale factor ${\\ddot a} < 0$, sufficiently separated regions of the present day observable universe would never have been in causal contact.  Nevertheless, it appears the universe we observe looks very much the same, and in particular very smooth, in all directions.  This fact is best seen in the cosmic microwave background radiation (CMB) which has  temperature fluctuations of only one part in $10^5$ when measured from any direction in the sky \\cite{Bennett:1996ce}.  The inflation solution to this horizon problem is to picture the universe during an early epoch to undergo an accelerated expansion, ${\\ddot a} > 0$.  Such expansion can take an initially small causally connected patch and enlarge it to a size that comfortably encompasses our present day observed universe, thereby solving the horizon problem. To realize inflation, the equation of state within the inflationary patch must be of a very special form, possessing negative pressure. The potential energy of a scalar field has an equation of state that satisfies this requirement. This fact has been a key link towards a dynamical realization of inflation and thereby has further motivated the inflation solution.  As inflation is meant to solve the horizon problem, it is important that it does not require acausal initial conditions. In particular, the picture of inflation considered here is for the universe to emerge from an initial singularity and then enter into a hot big-bang radiation dominated evolution.  At some time $t_i$ after the initial singularity, the conditions appropriate for inflation should occur within a small patch that is contained within the causal horizon at that time. Chaotic inflation \\cite{ci,Linde:1990nc}  does not fall into this picture as there inflation is thought to start at the Planck epoch with homogeneity assumed on the Planck scale. To realize the picture of a ``local'' inflation, two requirements must be satisfied.  {}First a physically sensible embedding must be demonstrated of the inflating patch immersed within a non-inflating background \\cite{embedd,Vachaspati:2000dy,Trodden:1999wc}. By embedding we mean matching the inflationary space time with the background space time at the boundary of the patch. Second, it must be shown that for $t> t_i$, the patch is dynamically stable to sustain inflation \\cite{piran181,kb,kb90,gpprl,gold43,other,Goldwirth:1992rj}. {}For example, large fluctuations, which conceivably could enter the inflating patch at a maximum rate limited only by causality, should not destroy the inflationary conditions within the patch.  Recently, a convenient methodology has been developed in \\cite{Vachaspati:2000dy,Trodden:1999wc} for analyzing the embedding problem, based on the null Raychaudhuri equation \\cite{ray,Wald:1984rg}.  In Sec. \\ref{embedding_conditions} the flat spacetime ($\\Omega=1$) formulation in \\cite{Vachaspati:2000dy} is generalized to arbitrary spacetime curvature. Then, an alternative solution from \\cite{Vachaspati:2000dy} to the embedding problem will be identified, which is especially attractive for an inflating patch with an open geometry. In Sec. \\ref{dynamic_conditions}, initial conditions for scalar field dynamics are presented, which are consistent with causality and our embedding solution and which evolve into successful supercooled or warm inflationary regimes. In Sec. \\ref{dynamical_effects_on_embedding} we combine the results in Sec. \\ref{embedding_conditions}  and Sec. \\ref{dynamic_conditions} to evaluate the effect of inhomogeneities on the embedding problem. Finally, Sec. \\ref{conclusion} presents our conclusions. ",
        "conclusions": "\\label{conclusion} This paper has investigated the initial condition problem of inflation from the perspective both of spacetime embedding and inflaton dynamics.  Our study has highlighted two attributes of this problem which have not been addressed in other works.  {}First, from the perspective of spacetime embedding, we have observed that the global geometry can play an important role in determining the size of the initial inflationary patch that is consistent with the weak energy condition.  Second, from the perspective of inflaton dynamics, we have noted that a $\\Gamma {\\dot \\phi}$ damping term could alleviate several problems which traditionally have led to large scale homogeneity requirements before inflation.  The purpose of this paper was to note for both these attributes, their salient features with respect to the initial condition problem. In the wake of this, several details emerge that must be understood. Below, we will review the main result we found for both attributes and then discuss the questions that must be addressed in future work.  {}For a causally generated patch a successful embedding can be achieved if the patch does not contain an anti-trapped region.  We have shown that the MAS size can be arbitrarily larger than the Hubble scale provided $\\Omega$ is made small enough.  So if the patch Hubble horizon is taken as the minimum stable patch size, then the patch does not have to contain an anti-trapped region if $\\Omegainf < 1$. This generalizes the analysis of \\cite{Vachaspati:2000dy} which was only for  $\\Omega=1$. However, without the effects of damping, it appears that the event horizon, not the Hubble horizon, is the minimum stable patch size. {}For the de Sitter case one can see analytically that the event horizon is equal to the MAS size regardless of $\\Omega$ and numerical calculations suggest the same is true for power law inflation. However, radiation damping of perturbations could stabilize a patch smaller than the event horizon. In this case an open geometry for the patch would allow the patch not to contain an anti-trapped region and thus allow a causal embedding in an expanding background which does not violate the weak energy condition. Eventually the patch should develop a MAS within it, but by then it could have expanded to be larger than the background MAS.   {}For the dynamics problem, with respect to the scalar field the new consideration was the effect of a $\\Gamma {\\dot \\phi}$ damping term.  We found that such a term could suppress many of the effects from initial inhomogeneities of the inflaton, which in studies traditionally done without this term lead to important impediments to entering the inflation regime.  It appears evident that inclusion of such a damping term will lead to qualitative differences in the initial condition problem.  The most interesting outcome is initial inflationary patches smaller than the Hubble radius $1/{\\rm H}$ may be able to inflate.   This paper examined the consequences of damping terms but did not delve into their fundamental origin. {}For the cosmological setting, such damping terms are typically associated with systems involving a scalar field interacting with fields of a radiation bath.  In this case, such damping terms have been found in first principles calculations for certain warm inflation models \\cite{wifp}, although more work is needed along these lines. It is worth noting here that in the early stages of certain supercooled inflaton scenarios where radiation is present, in particular new \\cite{ni} and thermal \\cite{ti} inflation, a careful examination of the dynamics may reveal damping terms similar to this. Since the initial stages are the crucial period for the initial condition problem, if further study supports the importance of such damping terms, it may be useful to better understand damping effects also in such scenarios.  Since the most suggestive situation for the damping terms is where in addition a radiation component is present in the universe, in Subsect. IIIB we also studied the effects of this component on the initial condition problem. Specifically we studied the case most suggestive for the initial stages of new and warm inflation, where a small inflation patch is submerged inside a larger radiation dominated spatial region.  Our main observation has been that the minimal condition for the patch to inflate is that its vacuum energy density must be larger than the background radiation energy density.  Provided the vacuum energy sustains itself, since expansion of the universe will dilute the radiation energy density, it is not clear-cut that the external radiation energy can act with sufficient magnitude to impede inflation inside the patch.   We have shown that the gradient terms, due to the inhomogeneous scalar field in the background space time, make it possible for the patch boundary to overtake the background MAS. However the kinetic energy of the scalar field must not be dominant for this to happen.  This can be ensured by including a damping term in the scalar field equation."
    },
    "0010/astro-ph0010382_arXiv.txt": {
        "abstract": "Spectroscopic observations of a proto-planetary nebula He 3-1475 with the Space Telescope Imaging Spectrograph (STIS) reveal the kinematics of its high (1200 km s$^{-1}$) velocity jets. The jets are formed at a large (0.15 pc) distance from its central star by collimation of an asymmetric stellar wind in a pair of conical shocks seen in Wide Field Planetary Camera (WFPC2) images. The jets consist of several pairs of knots symmetrically distributed with respect to the central star, with most knots exhibiting a head-tail morphology. Large (up to 650 km s$^{-1}$) radial velocity gradients are seen within the knots on subarcsec scales, with velocities decreasing from the knot heads toward their trailing tails. These large velocity gradients are a sign of efficient deceleration of jets by a much slower bipolar outflow. The inclination angle of the bipolar outflow is equal to 40\\degr, based on Doppler shifts of the scattered stellar H$\\alpha$ line. Its velocity is equal to 140 km s$^{-1}$ at a distance of 0.23 pc from the star, and increases monotonically with the radial distance from the star. A comparison of new WFPC2 [N II] $\\lambda$6584 images with older WFPC2 images reveals expansion of the jets. The measured jet proper motions in combination with their radial velocities imply that He 3-1475 is a Galactic Bulge star at a distance of 8 kpc, located 800 pc above the Galactic plane. Its very high luminosity (25,000 $L_\\odot$) implies that He 3-1475 must be significantly more massive than a typical AGB star within the Galactic Bulge, perhaps because of a past mass transfer and/or a merger event in an interacting binary system. ",
        "introduction": "He 3-1475 is a B[e] type star located in the direction of the Galactic Bulge, at $l = 9$\\fdg4, $b = +5$\\fdg8. Stars of this spectral type do not show typical absorption line spectra, but instead numerous permitted and forbidden emission lines are superposed on featureless continua (for further information see a recent review on B[e] stars by Zickgraf 1998). He 3-1475 is an unusual B[e] star because it is surrounded by a dense expanding torus of circumstellar matter (CSM) seen in OH maser emission (Bobrowsky et al. 1995), and also as a prominent dark torus in Hubble Space Telescope (HST) images (Bobrowsky et al. 1995; Borkowski et al. 1997). A bipolar outflow and spectacular jets, perpendicular to the torus, were detected in the optical by Riera et al. (1995) and Bobrowsky et al. (1995). Radial velocities of bright knots in jets span 1000 km s$^{-1}$, which is an order of magnitude more than what is typically seen in both young and evolved stars surrounded by dense CSM. The morphology of the jets is also unusual, because they do not originate in the central star. Instead, a fast bipolar stellar outflow is apparently collimated far from the star in conical shocks seen in the HST images (Borkowski et al. 1997). The jets above the tips of conical shocks consist of a series of bright pairs of knots, located symmetrically with respect to the central star. Their spectra show the presence of substantial ($\\sim 450$ km s$^{-1}$) velocity gradients within the knots, which indicates their strong deceleration by the ambient, more slowly moving material in the bipolar lobes (Riera et al. 1995). There is also evidence for deceleration of knots on large scales, as the knot radial velocities decrease with increasing distance from the cone tips to the periphery of the bipolar outflow.  He 3-1475 is clearly an unusual source whose nature is not understood at present. It's distance has been estimated at 1 -- 5 kpc (Riera et al. 1995; Bobrowsky et al. 1995), which together with its high infrared flux (Parthasarathy \\&\\ Pottasch 1989) suggests that He 3-1475 is probably a post-Asymptotic Giant Branch (AGB) proto-planetary nebula (PPN). (Knapp et al. 1995 detected 8.4 GHz continuum radio emission, which they interpreted as coming from a compact, newly formed PN close to the central star.) This conclusion should be considered preliminary because a reliable distance determination is lacking. The geometry of the circumstellar medium is also poorly known, with an uncertain (30\\degr\\ -- Riera et al. 1995) inclination of the jets with respect to the line of sight. The velocity of the bipolar outflow and its age are unknown. This lack of fundamental knowledge about He 3-1475 makes it difficult to understand this puzzling object.  We report here on imaging and spectroscopic observations of He 3-1475 with the HST, which allow us to determine its distance, inclination and kinematics of its jets, and the velocity of the bipolar outflow. Imaging and astrometry with the Wide Field Planetary Camera (WFPC2) and the Space Telescope Imaging Spectrograph (STIS) data are presented and discussed in \\S~2 and \\S~3, respectively. The jet inclination and the distance to He 3-1475 are determined in \\S~4 and \\S~5, and the nature of He 3-1475 and of its jets are discussed in \\S~6. ",
        "conclusions": "Our distance determination places He 3-1475 in the Galactic Bulge, at a much larger distance than previously anticipated. At this distance its infrared luminosity is equal to 25,000 $L_\\odot$ (Riera et al. 1995), which might be less that the stellar luminosity if the dust optical depth is low along the bipolar outflow so that a substantial fraction of stellar luminosity escapes along the poles. He 3-1475 is apparently a high-luminosity Galactic Bulge star at a height of 800 pc above the Galactic plane. Its location suggests that He 3-1475 is an old Population II star, but an old, low-mass post-AGB star cannot be so luminous -- a more massive star is required. Luminous, more massive AGB stars are indeed found in old stellar populations (e. g., in the globular cluster NGC 6553 -- Guarnieri et al. 1997), although their luminosities may be several times lower than that of He 3-1475. It is also feasible that He 3-1475 is a member of a trace intermediate-age population of Galactic Bulge.  Irrespective of its age and origin, He 3-1475 appears to be an evolved star which lost a substantial amount of mass in a slow asymmetric wind at least 1,500 yr ago, possibly in a superwind at the AGB tip or during a binary interaction, and which is now seen in the optical as a dense torus. The presence of broad P-Cygni profiles in the stellar spectrum shows that the transition from the slow to the fast wind has occurred since then. The B[e] spectral classification of the central star implies that this fast wind is asymmetric, according to a disk wind model for B[e] stars by Zickgraf et al. (1985). This wind is perhaps driven by stellar UV radiation from a single fast rotating B type star (e. g., Pelupessy, Lamers, \\& Vink 2000), or more generally from a star + a disk system (Oudmaijer et al. 1998). This asymmetric wind is then collimated into 1200 km s$^{-1}$ jets during its interaction with the more slowly moving material ejected earlier, maybe purely by hydrodynamical means (Borkowski et al. 1997). The time variability often seen in many B[e] stars suggests that a ``knotty'' jet morphology seen in He 3-1475 could be caused by temporal variations in the stellar wind properties.  Our STIS observations show how this high-velocity, time-dependent jet is decelerated by the circumstellar medium on both large and small scales. We see the large scale deceleration in the decrease of the maximum velocity as we move down the jet and out to the knots. But we also see at all points along the flow, and especially at the knots, local ($\\sim 0\\farcs05 = 6 \\times 10^{15}$ cm) decelerations of over 500 km s$^{-1}$. Consider that a plane-parallel shock of 500 km s$^{-1}$ would produce gas with a temperature of $3 \\times 10^6$ K, while we see a spectrum which seems more appropriate for a cooling shock of far lower velocity, about 100 km s$^{-1}$ (Riera et al. 1995). Clearly, there is much hydrodynamical structure which is still unresolved by HST.  It is still a mystery why B[e] star+disc systems are present in evolved post-AGB stars such as He 3-1475. There are a dozen such systems in compact PNe or proto-PNe (Lamers et al. 1998), albeit generally of lower luminosity than He 3-1475. The best known of these systems is the ``Butterfly'' PN M2-9, a low luminosity (550 L$_\\odot$) object at a distance of 650 pc (Schwarz et al. 1997). M2-9 is clearly very much different than He 3-1475, suggesting that compact PN B[e] central stars form a heterogeneous class, just like B[e] stars as a whole (Lamers et al. 1998). M2-9 shows a complex kinematics, with young ($< 10$ yr) micro-jets close to the star with velocities up to 195 km s$^{-1}$ and much older (1,300 yr) bipolar outflows with velocities up to 140 km s$^{-1}$ (Solf 2000). The 1200 yr old outermost reflection knots are expanding with velocity of 165 km s$^{-1}$ (Schwarz et al. 1997). Doyle et al. (2000) find evidence for $\\sim 1000$ km s$^{-1}$ outflows. High velocity (500 km s$^{-1}$) outflows were also found by Redman et al. (2000) in another PN with a B[e] central star, Mz3, while spectacular jets were recently imaged by the HST in yet another PN on Lamers' et al. PN and PPN list, He 2-90 (Sahai \\& Nyman 2000). Both M2-9 and He 2-90 are suspected to be interacting binaries. An interacting binary progenitor hypothesis also seems attractive for He 3-1475, because its very high luminosity excludes a typical low-mass AGB star in the Galactic Bulge. A mass transfer binary and/or a merger event provides a reasonable solution to the presence of a more massive star in the Galactic Bulge.  In this paper, we concentrated our attention on the large-scale jet kinematics and on the global properties of He 3-1475. The next step in analysis of the extensive STIS dataset should involve detailed studies of individual knots and of the jet collimation regions. Such studies are essential for providing answers to questions about the mechanism of the jet collimation, about the nature of the observed shock emission, and about the temporal and spatial variability of the flow. For example, a simple (but perhaps naive) interpretation would be to attribute the observed shock emission to shocks driven into the jets, such as encountered in working surfaces of continuous jets. Because of the high jet velocities, the bow shocks ahead of the knots should in principle be nonradiative, X-ray emitting shocks, with little or no optical emission. The presence of distinct knots symmetrically distributed with respect to the star suggests that the stellar outflow varies on timescales from tens to hundreds of years, leading to noncontinuous, episodic, and wobbling jets. It is also not certain whether the jets can be collimated by purely hydrodynamical processes, or whether magnetic fields must be involved. We hope that the continuing analysis of the HST He 3-1475 dataset will provide answers to some of these questions in the near future."
    },
    "0010/astro-ph0010457_arXiv.txt": {
        "abstract": "We model the population of double white dwarfs in the Galaxy and find a better agreement with observations compared to earlier studies, due to two modifications. The first is the treatment of the first phase of unstable mass transfer and the second the modelling of the cooling of the white dwarfs.  A satisfactory agreement with observations of the local sample of white dwarfs is achieved if we assume that the initial binary fraction is $\\sim$50 \\% and that the lowest mass white dwarfs ($M < 0.3 \\msun$) cool faster than the most recently published cooling models predict.  With this model we find a Galactic birth rate of close double white dwarfs of 0.05 yr$^{-1}$, a birth rate of AM CVn systems of 0.005 yr$^{-1}$, a merger rate of pairs with a combined mass exceeding the Chandrasekhar limit (which may be progenitors of SNe~Ia) of 0.003 yr$^{-1}$ and a formation rate of planetary nebulae of 1 yr$^{-1}$.  We estimate the total number of double white dwarfs in the Galaxy as 2.5\\,$\\times\\,10^8$.  In an observable sample with a limiting magnitude $V_{\\rm lim} = 15$ we predict the presence of $\\sim$855 white dwarfs of which $\\sim$220 are close pairs. Of these 10 are double CO white dwarfs of which one has a combined mass exceeding the Chandrasekhar limit and will merge within a Hubble time. ",
        "introduction": "Close double white dwarfs\\footnote{Throughout this work we'll use the term double white dwarf instead of double degenerate, which is commonly used, because the term double degenerate is sometimes used for white dwarf - neutron star or double neutron star binaries.}  form an interesting population for a number of reasons. First they are binaries that have experienced at least two phases of mass transfer and thus provide good tests for theories of binary evolution.  Second it has been argued that type Ia supernovae arise from merging double CO white dwarfs \\citep{web84,it84a}.  Thirdly close double white dwarfs may be the most important contributors to the gravitational wave signal at low frequencies, probably even producing an unresolved noise burying many underlying signals \\citep{eis87,hbw90}.  A fourth reason to study the population of double white dwarfs is that in combination with binary evolution theories, the recently developed detailed cooling models for (low-mass) white dwarfs can be tested.  The formation of the population of double white dwarfs has been studied analytically by \\citet{it86a,it87} and numerically by \\citet{lp88,ty93,ty94,ylt+94,hpe95}; \\citet[hereafter \\citetalias{ity97}]{ity97}, and \\citet[hereafter \\citetalias{han98}]{han98}.  Comparison between these sudies gives insight in the differences that exist between the assumptions made in different synthesis calculations.  Following the discovery of the first close double white dwarf \\citep{slo88}, the observed sample of such systems in which the mass of at least one component is measured has increased to 14 \\citep{mm99,mmm+00}. This makes it possible to compare the models to the observations in more detail.  In this paper we present a new population synthesis for double white dwarfs, which is different from previous studies in three aspects. The first are some differences in the modelling of the binary evolution, in particular the description of a common envelope without spiral-in, in which the change in orbit is governed by conservation of angular momentum, rather than of energy (Sect.~\\ref{binev}). The second new aspect is the use of detailed models for the cooling of white dwarfs (Sect.~\\ref{cooling}), which are important because it is the rate of cooling which to a large extent determines how long a white dwarf remains detectable in a magnitude-limited observed sample. The third new aspect is that we use different models of the star formation history (Sect.~\\ref{SFH}).  Results are presented in Sect.~\\ref{results} and discussed in Sect.  \\ref{discussion}. The conclusions are summarised in Sect.~\\ref{conclusion}. In the Appendix some details of our population synthesis are described . ",
        "conclusions": "We computed a model of the population of close binary white dwarfs and found good agreement between our model and the observed double white dwarf sample. A better agreement with observations compared to earlier studies is found due to two modifications.  The first is a different treatment of unstable mass transfer from a giant to a main sequence star of comparable mass.  The second is a more detailed modelling of the cooling of low mass white dwarfs which became possible because detailed evolutionary models for such white dwarfs became available. Our main conclusions can be summarised as follows.  \\begin{enumerate} \\item Comparing the mass distribution of the white dwarfs in close pairs with the observations, we find a lack of observed white dwarfs with masses below 0.3 \\msun. This discrepancy can be removed with the assumption that low-mass white dwarfs cool faster than computed by \\citet{dsb+98}.  The same assumption removes discrepancies between observed and derived ages of low-mass white dwarfs that accompany recycled pulsars, as shown by \\citet{kbk+00}. Faster cooling is expected if the hydrogen envelopes around low-mass white dwarfs are partially expelled by thermal flashes or a stellar wind.   \\item Our models predict that the distribution of mass ratios of double white dwarfs, when corrected for observational selection effects as described by \\citet{mmm00}, peaks at a mass ratio of unity, consistent with observations. The distributions predicted in the models by \\citet{ity97} and \\citet{han98} peak at mass ratios of about 0.7 and above 1.5 and agree worse with the observations even after applying selection effects.  \\item Our models predict a distribution of orbital periods and masses of close double white dwarfs in satisfactory agreement with the observed distribution.   \\item Amongst the observed white dwarfs only a small fraction are members of a close pair. To bring our models into agreement with this, we have to assume an initial binary fraction of 50\\%\\ (i.e.\\ as many single stars as binaries).  \\item In our models the ratio of the local number density of white dwarfs and the planetary nebula formation rate is a sensitive function of the star formation history of the Galaxy. Our predicted numbers are consistent with the observations.   \\item Using detailed cooling models we pridict that an observed sample of white dwarfs near the Sun, limited at the magnitude $V=15$, contains 855 white dwarfs of which 220 are close pairs. Of these pairs only 10 are double CO white dwarfs and only one is expected to merge having a combined mass above the Chandrasekhar mass. The predicted merger rate in the Galaxy of double white dwarfs with a mass that exceeds the Chandrasekhar mass is consistent with the inferred SN~Ia rate.  \\citetalias{ity97} estimated, depending on $\\alpha_{\\rm ce}$, to find one such pair in a sample of $\\sim$200 to $\\sim$600 white dwarfs.  Reversing this argument, when the statistics become more reliable, the observed number of systems with different types of white dwarfs could provide constraints on the cooling models for these white dwarfs.  \\end{enumerate}"
    },
    "0010/astro-ph0010331_arXiv.txt": {
        "abstract": "To build upon the goals of the upcoming INTEGRAL mission, the next generation soft $\\gamma$-ray (0.2-20~MeV) observatory will require improved sensitivity to nuclear line emission while maintaining high spectral resolution. We present the simulated performance of a germanium Compton telescope (GCT) design, which will allow a factor of ten improvement in sensitivity over INTEGRAL/SPI. We also discuss a number of issues concerning reconstruction techniques and event cuts, and demonstrate how these affect the overall performance of the telescope. ",
        "introduction": "Combining the success of COMPTEL/CGRO and the spectral resolution of INTEGRAL/SPI, a number of researchers \\citep{johns96, jean96, boggs98} have discussed the merits of a germanium Compton telescope (GCT). Compton telescopes work on a well-known principle: by measuring the positions and energies of the photon interactions the initial photon direction can be reconstructed to within an annulus on the sky using the Compton scatter formula (Figure 1). The uncertainty, or width, of this annulus depends on the spatial and spectral resolution of the detectors, but also has a fundamental limit set by Doppler broadening due to Compton scattering off of bound electrons. Reconstruction of the event annulus requires that the first and second photon interaction locations in the instrument are spatially resolved, and their order properly determined.  Germanium detectors pose two major complications for Compton telescope designs. Photons above $\\sim$0.5~MeV predominantly scatter multiple times in germanium before being photoabsorbed. Also, the expected event timing resolution in germanium detectors ($>$10~ns) is not adequate to determine the interaction order for reasonable GCT configurations.  \\begin{figure} \\centering \\includegraphics[width=1.0\\linewidth]{figure1.eps} \\caption{Germanium Compton telescope configuration analyzed in this work.\\label{figure1}} \\end{figure}  In a previous paper \\citep{boggs00}, hereafter Paper I, we introduced two \\textit{reconstruction techniques} to accurately determine the photon interaction order in GCTs. The first technique, Compton Kinematic Discrimination (CKD), takes advantage of redundant information for photons which interact three or more times in the instrument (3+ site events) to determine the most probable interaction order. CKD additionally allows efficient rejection of background events, including photons which scatter out of the instrument before fully depositing their energy (Compton continuum photons), non-localized $\\beta^{-}$-decays, $\\beta^{+}$-decays, and pair-production events. The second reconstruction technique, Single Scatter Discrimination (SSD), allows good determination of the interaction order in 2-site events, but without the benefit of background rejection. (Also in Paper I we summarized additional techniques for rejecting $\\beta^{-}$-decays, $\\beta^{+}$-decays, and pair-production events.) In addition to CKD and SSD, other reconstruction techniques can be imagined. As we discussed in Paper I, for a given GCT configuration the performance will depend on the reconstruction techniques employed.  Telescope performance will also depend on \\textit{event cuts}, which can be made on the initial direction of the photon scatter, the number of interaction sites, and the minimum separation between the first and second interaction sites (\\textit{minimum lever arm}). The tradeoffs are generally higher efficiency at the expense of degraded angular resolution, and hence increased background. First, the uncertainty in the Compton scatter angle (angular resolution) is smaller for forward scatter events than backscatter events (Paper I, Equation 4). Second, events with only 2 interaction sites (2-site) do not permit CKD background rejection, and also have a larger fraction of backscatter events that 3+ site events. Finally, a larger minimum lever arm will minimize the effects of spatial uncertainty in the detectors, improving angular resolution and hence background, but at the expense of lower efficiency.   \\begin{figure} \\centering \\includegraphics[width=0.9\\linewidth]{figure2.eps} \\caption{Angular resolution for on-axis sources as a function of photon energy. Also shown is the limit set by Doppler broadening of bound electrons.\\label{figure3}} \\end{figure}   Here we present detailed simulations of a GCT configuration in an effort to determine the optimized sensitivity to nuclear line emission, as well as demonstrate the variation in performance for several different combinations of reconstruction techniques and event cuts. Our selections range from utilizing most of the event information to physically reconstruct the event (Case 1, with CKD), to using a purely empirical, but highly efficient approach (Case 4), and should fairly represent the range of performance characteristics possible within a GCT.  \\begin{figure} \\centering \\includegraphics[width=0.9\\linewidth]{figure3.eps} \\caption{Photopeak spectral resolutions, which show little dependence on the reconstruction technique or event cuts. For comparison is shown the assumed single-site resolution.\\label{figure4}} \\end{figure} ",
        "conclusions": "GCTs offer an attractive option for a soft $\\gamma$-ray observatory following INTEGRAL. The GCT presented here will allow a factor of 10 improvement in nuclear line sensitivity over INTEGRAL, which is required for new scientific goals such as the systematic study of Type Ia SNe, as well as improved imaging of the positron annihilation line, $^{26}$Al, $^{60}$Fe, and $^{44}$Ti. We have also demonstrated how a judicious choice of reconstruction techniques (CKD) and event cuts (3+ site, forward scatters) results in a 3-fold improvement in sensitivity near 1~MeV, while maintaining angular resolution near the Doppler-broadening limit. Our next goal is to study the performance of several different GCTs configurations, varying the detector-plane spacings, to determine how the geometry affects performance."
    },
    "0010/astro-ph0010107_arXiv.txt": {
        "abstract": "The ability of the neutron star surface to supply all or only part of the charges filling the pulsar magnetosphere is crucial for the physics prevailing within it, with direct consequences for the possible formation of pair creation regions. We evaluate the conditions for $e^-$ emission from pulsar surfaces for  a simple Goldreich-Julian geometry taking both thermal and field emission processes into account. Using recently published estimates for the equation of state at the neutron star's surface, we show, that for a large range of $T_{\\rm surf}, B$ and $P$, the liberated charges will fully screen the accelerating B-parallel electric field  ${\\rm E_\\|}$. For surface temperatures $T_{\\rm surf}<2\\cdot 10^5 $K a balance between field emission of electrons and shielding of the field will occur. Even in the overidealised case of  $T_{\\rm surf}=0$ one can expect a prodigious supply of electrons which will weaken the accelerating ${\\rm E_\\| }$. We calculated the motion of electrons along selected polar field lines numerically for the low temperature, field emission scenario yielding their Lorentz factors as well as the produced radiation densities. Inverse Compton  and later curvature losses are seen to balance the acceleration by the residual electric fields.  We found that the conditions for magnetic pair production are not met anywhere along the field lines up to a height of 1500 pulsar radii. We did not {\\it a priori} assume an \"inner gap\", and our calculations did not indicate the formation of one under realistic physical conditions without the introduction of further assumptions. ",
        "introduction": "A proper assessment of the  conditions very close to the surface of a neutron star is a vital ingredient for any model that attempts to explain the enigmatic origin of the observed electromagnetic radiation from pulsars.  Numerous studies of the polar cap acceleration have been made within a particular framework: A steady acceleration of electrons within a space charge limited particle flow originating at the surface of the neutron star will produce particles emitting magnetically convertible curvature $\\gamma-$rays (e.g. Arons \\& Scharlemann 1979). These high energy photons are converted into $e^+e^-$-pairs with high enough efficiency (Daugherty \\& Harding 1982) so that in the end, they and their daughter products will exceed the initial (Goldreich-Julian density) by a factor of $10^4$ (e.g. Zhang \\& Harding 2000). An important {\\it a priori } assumption is inherent in all these calculations: Above a critical height $h_{\\rm gap}$ a region of high conductivity will quench any electric field ${\\rm E_\\|}$ parallel to the guiding B-field lines, so that the full potential across the polar cap will be active between the outer surface of the neutron star and the inner surface of the high conductivity region.  Our approach differs in that respect from other workers, we do not make that initial assumption, but try to calculate the amount of charges produced by the surface, how they affect and are affected by  ${\\rm E_\\|}$, their energy losses and the possibility of pair production. Should one find suitable conditions for such a pair production region to exist, the methods quoted above will be the prime choice for its description.  The conclusion that electron-positron pairs with high densities exist close to the neutron star has recently been questioned by several authors. Kunzl et al. (1998) considered this widespread assumption of dense pair plasmas for the problem of radio wave propagation within the neutron star magnetosphere. On hand of the most recent emission height estimates (50 - 100 $R_{\\rm NS}$) for the radio emission (Kijak \\& Gil (1997); Kramer et al. (1997, 1998)) they found that the existence of low frequency radio emission for example from the Crab pulsar is incompatible with the scenario of extended dense pair production. These findings were fully confirmed by the theoretical investigation of Melrose \\& Gedalin (1999), who treated the plasma emission mechanisms in detail and came to the conclusion that only low energy particles with Goldreich-Julian density could exist in the radio emission regions in order to alow for escaping radio waves (see also Melrose (2000) and references therein). Furthermore recent simulations of the pair cascade by Arendt \\& Eilek (2000) for strong and weak magnetic fields shed some light on the difficulties for an efficient production of secondary pairs by high energetic primary particles via $\\gamma$-ray curvature photons. Although their findings need some more investigation they could show that the field strength at the surface has the most important effect on the cascade. For strong fields $B\\sim 10^9\\,{\\rm T}$ they find almost no secondary photons and the cascade terminates in one generation. The weak fields ($B\\sim 10^7\\,{\\rm T}$) had either no cascades or quite weak ones. Only for moderate fields ($B\\sim 10^8\\,{\\rm T}$) they observed some secondary particle generations. Arendt \\& Eilek concluded that it is not yet clear whether pulsars can really produce a large number of pairs at the surface gaps.  This tempted us to question the canonical {\\it a priori} assumption of the existence of an inner pair creating gap and investigate its possibility given a reasonable estimate of the surface characteristics and the fields involved.  In the literature there are three fundamentally different ideas for pair production in the inner magnetosphere: A \"starved\" magnetosphere where the neutron star surface provides only an insufficient supply of charged particles (e.g. Ruderman and Sutherland (1975)), a magnetosphere where the Goldreich-Julian density is reached at the surface but due to curvature effects of the field lines deviations occur close to the surface (e.g. Arons \\& Scharlemann (1979), Barnard \\& Arons (1982)) and Shibata's model which predicts electric fields caused by {\\em overdense} regions, that means in the opposite sense than in the Arons model (Shibata, 1997).  The latter two ideas suffer from  serious weaknesses since it turned out that Arons type models do not work for slow pulsars (Beskin (2000) gives an upper limit of 0.1 to 0.3 s for the period). The Shibata model in turn neglects the back reaction of an overdense region which reduces the particle flow from inside. Therefore we will restrict our discussion to models of Ruderman-Sutherland type and fix the parameter range in which this kind of models is reasonable.  For the scope of this article we consider a strong magnetic dipole anchored in a heavy ($1.4 M_{\\rm sun}$) sphere with a radius $r= 10\\,{\\rm km}$ that rotates twice per second ($\\Omega=2 \\pi \\cdot 2 s^{-1}$). Let the B-field have a strength of $|B_0|=10^{8}~{\\rm T}$ at the surface and for simplicity`s sake let the magnetic and the rotational axes be aligned\\footnote{We use spherical coordinates $(r,\\theta,\\phi)$ centered in the neutron star with $\\theta$ being the directional angle referred to the pole which coincides with the rotation axis.}. \\begin{equation} {\\rm B}(r,\\theta) = B_0\\cdot \\left( {r_{\\rm NS} \\over r}\\right) ^{-3} \\cdot (\\cos\\theta ,{1\\over 2}\\sin\\theta , 0) \\end{equation} Such objects have been discussed as idealised cases of radio pulsars by Goldreich \\& Julian (1969) and many authors since then. Of course one has to realise that the above mentioned idealisations limit the applicability of such simple models for the interpretation of observational facts from the growing number of pulsars that is being discovered. We will however try to extend the scope of these models by constraining the number of charged particles available and analysing the fate of individual charged particles (limiting ourselves to electrons and positrons) in the polar regions of such heavy rotating dipoles under the constraints given by some of the more recent observations.  Although a pulsar magnetosphere is characterized by extreme physical conditions in terms of strong electric and magnetic fields the fundamental processes that can supply charges are reasonably well known from terrestial experiments and technical applications and can be extrapolated to the surface conditions of a neutron star. Such an extrapolation provides constraints on the initial charge density which is vital for any model of cascading particle creation and the radiation processes. ",
        "conclusions": ""
    },
    "0010/astro-ph0010277_arXiv.txt": {
        "abstract": "The first VLBI images of H$_2$O maser emission in the Circinus Galaxy AGN show both an accretion disk and outflow 0.1 to 1 pc from the central engine.  The maser traces a warped, edge-on accretion disk between radii of about 0.1 and 0.4 pc that is bound by a $1.3\\times10^6$ M$_\\odot$ central mass.  The rotation curve is somewhat flatter than a Keplerian rotation law and is consistent with a disk mass on the order of $10^5$ M$_\\odot$. Clumpy substructure may display a kinematic signature consistent with spiral arms. Away from the disk, a second population of water masers traces broadly a bipolar, wide-angle outflow that contains (bullet-like) clumps ejected from a region centered on the central engine and $<0.1$ pc in radius.  Out to a radius of $\\sim 0.4$ pc, the warp of the accretion disk appears to channel the outflow, the orientation of which on the sky coincides with the orientation of the known kiloparsec-scale ionization cone.  Beyond this radius, the flow crosses the disk and truncates it. This suggests that the current accretion event and associated disk-outflow geometry have a lifetime on the order of $10^7$ years. ",
        "introduction": "The Circinus galaxy is one of the nearest Seyfert\\,II galaxies that hosts an extragalactic H$_2$O maser. (Freeman \\etal\\ 1977 estimate a distance of 4~Mpc.) The $\\sim 10^{42}$ erg s$^{-1}$ (2-10 keV) central engine is obscured at energies below 10 keV by a large gas column, n$_{\\rm H}\\sim4\\times10^{24}$ cm$^{-2}$ (\\cite{matt99}).  The X-ray spectrum also exhibits a prominent Fe K$\\alpha$ line that probably comprises  reflected radiation (\\cite{matt96}). These characteristics suggest some  similarity to another maser-host galaxy, NGC\\,1068.  The Circinus AGN exhibits an ionization cone (\\cite{marconi94}) within which Veilleux \\& Bland-Hawthorn (1997) observed linear optical filaments and compact knots reminiscent of Herbig-Haro objects. The cone opening angle is at least $90^\\circ$, at a mean position angle (PA) of roughly $290^\\circ$. The AGN also drives a kpc-scale nuclear outflow betrayed by bipolar radio lobes (\\cite{elmouttie98}) at PA $\\sim 295^\\circ$, which are largely aligned with the minor axis the galactic H\\,I disk (\\cite{jones99}) and a nuclear $^{12}$CO ring (\\cite{curran98}).  There is strong evidence that at least several extragalactic H$_2$O masers trace edge-on accretion disks bound by central engines $\\ga 10^6$ M$_\\odot$. These masers display (1) roughly linear structure in projection on the sky, (2) emission symmetrically bracketting the galactic systemic velocities, and (3) declining rotation curves that indicate more or less Keplerian differential rotation.  Discovery of similar  symmetry in the Circinus H$_2$O maser spectrum (Nakai\\ \\etal\\  1995; Greenhill\\ \\etal\\  1997) led to speculation that there too the masers trace an accretion disk (\\cite{greenhill97}). Though the observation of an accretion disk in Circinus confirms that prediction, the discovery of qualitatively new, nondisk emission is equally interesting.  In NGC\\,1068, disk emission is accompanied by secondary maser emission $\\gg1$ pc downstream in a jet, where material in the narrow line region deflects it (\\cite{gallimore96}). However, the nondisk emission in Circinus lies $<1$ pc from the central engine and is more closely related to it. ",
        "conclusions": ""
    },
    "0010/astro-ph0010041_arXiv.txt": {
        "abstract": "A field of $\\sim 38^\\prime \\times 38^\\prime$ around the supernova remnant (SNR) G349.7+0.2 has been surveyed in the CO J=1--0 transition with the 12 Meter Telescope of the NRAO, using the On-The-Fly technique. The resolution of the observations is $54\\pp$. We have found that this remnant is interacting with a small CO cloud which, in turn, is part of a much larger molecular complex, which we call the ``Large CO Shell''. The Large CO Shell has a diameter of about 100 pc, an H$_2$  mass of 9.3$\\times 10^5$ M$_\\odot$ and a density of 35 \\c3 . We investigate the origin of this structure and suggest that an old supernova explosion ocurred about $4 \\times 10^6$ years ago, as a suitable hypothesis. Analyzing the interaction between G349.7+0.2 and the Large CO Shell, it is possible to determine that the shock front currently driven into the molecular gas is a non-dissociative shock (C-type), in agreement with the presence of OH 1720 MHz masers. The positional and kinematical coincidence among one of the CO clouds that constitute the Large CO Shell, an IRAS point-like source and an ultracompact H II region, indicate the presence of a recently formed star. We suggest that the formation of this star was triggered during the expansion of the Large CO Shell, and suggest the possibility that the same expansion also created the progenitor star of G349.7+0.2. The Large CO Shell would then be one of the few observational examples of supernova-induced star formation. ",
        "introduction": " ",
        "conclusions": ""
    },
    "0010/astro-ph0010088_arXiv.txt": {
        "abstract": "We reveal the role of clustering of sub-clumps, which is expected in the cold dark matter (CDM) universe, in forming a bright elliptical galaxy (BEG) from a low-spin seed galaxy. This can be done by comparing the evolution of a low-spin seed galaxy including small-scale density fluctuations expected in the CDM universe (Model 1) with that of a completely uniform one (Model 2), using numerical experiments. We show that Model 2 cannot reproduce the properties of BEGs and forms a disk which is too compact and too bright due to the conservation of the initial-small angular momentum. In Model 1 clustering of the sub-clumps caused by initial small-scale density fluctuations leads to angular momentum transfer from the baryon component to the dark matter and consequently a nearly spherical system supported by random motions is formed. Moreover the collisionless property of the stars formed in the sub-clumps prevents the dissipative contraction of the system, leading to a large measured half-light radius. As a result, the end-product is quite well reproduces the observed properties of BEGs, such as the de Vaucouleurs light-profile, typical color and metallicity gradients, the large half-light radius, the small ratio of the rotational velocity to the velocity dispersion ($V/\\sigma$). We conclude that the clustering of sub-clumps, i.e., the hierarchical clustering, plays a crucial role in the formation of BEGs from a low-spin seed galaxy. ",
        "introduction": "Observed bright elliptical galaxies (BEGs) have the common interesting properties, which are expected to provide strong constraints on their formation scenario. For example, Davies et al.\\ (1983) found that the ratios of the rotation velocity to the velocity dispersion, $V/\\sigma$, of BEGs are much smaller than those of faint ellipticals. In photometric properties, the light profile of BEGs is well described by the de Vaucouleurs law, whereas faint ellipticals usually have an exponential light profile (e.g., Caon, Capaccioli, \\& D'Onofrio 1993).  On the other hand, Kawata (1999) showed that in the CDM universe, i.e., the hierarchical clustering scenario, elliptical galaxies can be formed from a low-spin overdense region, using numerical simulation which includes almost all the important physical processes in galaxy formation self-consistently. He adopted a semi-cosmological model proposed by Katz \\& Gunn (1991). In this model, the process of galaxy formation is mimicked by a collapse of a top-hat over-dense sphere (a seed galaxy) which initially follows a Hubble flow expansion and has a solid-body rotation as an effect of the external tidal field. The amount of the initial solid body rotation is specified by a dimensionless spin parameter, $\\lambda$ (Peebles 1971). In addition, small-scale density fluctuations expected in the CDM universe are superposed on this sphere which consists of dark and baryonic matter. Using this model, Katz \\& Gunn (1991), Katz (1992), and Steinmetz \\& M\\\"uller (1994, 1995) studied only the case of $\\lambda=0.08$ and seemed to succeed in forming a system which has similar properties to an observed disk galaxy in several respects (see also Koda, Sofue, \\& Wada 2000a; 2000b, who presented, in the same way, that disk-like systems are formed for a range of $\\lambda=0.10-0.04$). However, the spin parameter of the virialized dark matter falls in a range 0.02--0.11 with a median value of 0.05, according to the results of N-body simulations (e.g., Barnes \\& Efstathiou 1987; Warren et al.\\ 1992) and analytical works (e.g., Steinmetz \\& Bartelmann 1995; Eisenstein \\& Loeb 1995), in which $\\lambda=0.08$ is near a high-end value. Kawata (1999) studied the evolution of a seed galaxy which has a slow rotation corresponding to $\\lambda=0.02$ and showed that the end-product reproduces the observed properties of elliptical galaxies very well. Cosmologically, a small spin-parameter is preferred in a galactic halo collapsing at a high redshift (Heavens \\& Peacock 1988). Moreover, elliptical galaxies are considered to be a system whose stellar component is old and rotates slowly. Thus, it is a quite natural consideration that most elliptical galaxies were formed in a halo collapsing at high redshift and rotating slowly.  \\begin{figure*} \\epsscale{.9} \\plotone{f1.eps} \\figcaption[f1.eps] { Time evolution of the system in Model 1. The upper, middle, and lower panels show the distributions of the dark matter, gas, and stars respectively.  Each panel measures 150 kpc across and shows the $x$-$z$ projection of the particles, where we set the $z$-axis to be the initial rotation axis. \\label{anim1}} \\end{figure*}  Kawata's study was based on the hierarchical clustering scenario, because in his simulation the galaxy forms through mergers of sub-clumps caused by initial small-scale density fluctuations (see Fig.\\ 1 of Kawata 1999). The question we have to ask here is how the clustering of sub-clumps acts on the formation of elliptical galaxies from a low-spin seed galaxy. In order to understand further the role of the clustering, we compare the evolution of a low-spin seed galaxy including small-scale density perturbations with that of a uniform one, as the opposite case. Then, we focus on massive seed galaxies, and examine whether both types of seed galaxy can reproduce the observed properties of BEGs quantitatively, because the properties of BEGs, as mentioned above, are expected to be able to provide constraints on their formation scenario. If the uniform low-spin seed galaxy also succeeds in forming a system similar to BEGs, small-scale density fluctuations superposed on the initial slow-rotating sphere will not be indispensable to the model. The present study will thus clarify the role of the clustering of the sub-clumps induced by small-scale density fluctuations in galaxy formation from a massive low-spin seed galaxy.  The plan for the remainder of the paper is as follows. In Section 2, the numerical method and the model are described. Section 3 presents results of numerical simulations. Conclusion is described in Section 4.  \\begin{figure*} \\epsscale{.9} \\plotone{f2.eps} \\figcaption[f2.eps] { The same as Fig.\\ 1 but for Model 2. \\label{anim2}} \\end{figure*} ",
        "conclusions": "We showed that the luminosity profile, color and metallicity gradients, half-light radius, and $V/\\sigma$ of the end-product formed by clustering of sub-clumps are in very good agreement with those of the observed BEGs quantitatively. Since the CDM cosmology naturally predicts the existence of massive low-spin seed galaxies including small-scale density fluctuations, the CDM scenario is quite suitable for the formation of BEGs from a massive low-spin seed galaxy, in which the clustering of sub-clumps plays an essential role.  On the other hand, in a massive low-spin seed galaxy including no initial small-scale density fluctuation, the small initial spin leads to a compact disk-like object dominated by the rotation owing to the angular momentum conservation. The disk has an exponential light profile, which is obviously inconsistent with the light profiles of the observed BEGs. Furthermore, the central surface brightness is too high to be in agreement with that of disk galaxies. No object like this has been observed in our universe despite its high surface brightness. This result suggest that a massive low-spin seed galaxy is not allowed to have a uniform density."
    },
    "0010/astro-ph0010263_arXiv.txt": {
        "abstract": "Seventeen southern sky BL Lacs were observed in UBVRI using the CCD Camera on the 1.0m telescope at the South African Astronomical Observatory (SAAO) in Aug and Nov 1999. The analyses of all the seventeen sources are now complete, and are available via anonymous ftp (ftp pukrs1.puk.ac.za/pub/Blazars). A few examples of our results are however given in this paper. Whereas PKS 2005-489 and 2155-304 appear to have been in a high state, PKS 0048-097 and PKS 0521-365 showed evidence of variability on a time-scale of a few days, with the amplitude of variability increasing towards short wavelengths. This is consistent with observations of gamma-ray BL Lacs, which show similar behaviour in optical and X-rays. ",
        "introduction": "BL Lacs are the only members of the population of extragalactic sources which have been observed at TeV energies. They emit their energy over a broad range of the spectrum (radio to $\\gamma$-rays), with signatures of variability in optical, X-rays and $\\gamma$-rays, which are large compared to other AGNs. BL Lacs are highly polarized non-thermal sources with compact flat-spectrum radio cores. Most BL Lacs were identified in radio and X-ray surveys. Whereas those identified in X-rays (XBLs) are mostly nearby, the radio selected BL Lacs (RBLs) have a larger distance scale. Whereas a few XBLs were detected as TeV sources, we find that the RBLs were mostly detected as EGRET sources (see e.g. Lin et al. 1997, 1999).  The number of BL Lacs known presently are mostly northern sources, which constitute $\\sim$75\\% of all known BL Lacs. Convincing detections of VHE photons from BL Lacs have mainly been made from Northern Hemisphere sources, which include Mrk 421 (Punch et al. 1992), Mrk 501 (Quinn et al. 1996) and 1ES 2344+514 (Catanese et al. 1998). TeV observations of Southern Sky sources mostly gave upper limits (e.g. Roberts et al. 1999). Besides PKS 2155-304, which has also been recently confirmed as a TeV source (Chadwick et al. 1999), none of the southern BL Lacs have been well monitored. Therefore, it is justifiable to claim that there are a number of BL Lacs that have not been identified from the southern skies, and probably with advances in VHE ground-based telescopes, some are likely to be identified as TeV sources in future. Our objective is to monitor Southern Sky BL Lacs in optical, and such information can be combined with contemporary X-ray observations for modeling purposes to identify further TeV candidates. ",
        "conclusions": "Whereas BL Lac variability is a common feature, our visual multicolor observations provide important information concerning the relative flux change vs. frequency. The observation of an increasing amplitude of flux variability towards higher frequencies (in this case for PKS 0048-097 and PKS 0521-365) should be combined with simultaneous X-ray observations. This synchrotron feature in the optical/X-ray region of the spectrum should also imply a similar behaviour in the inverse Compton ($\\gamma$-ray)  component of the spectrum."
    },
    "0010/astro-ph0010113_arXiv.txt": {
        "abstract": "The straylight contamination due to the Galactic emission (GSC, Galaxy Straylight Contamination) entering at large angles from the antenna centre direction may be one of the most critical sources of systematic effects in observations of the cosmic microwave background (CMB) anisotropies by future satellite missions like {\\sc Planck} and MAP. For the Low Frequency Instrument (LFI), this effect is expected to be particularly crucial at the lowest frequency channels. We describe here different methods to evaluate the impact of this effect and compare it with other systematics of instrumental and astrophysical origin. The results are presented in terms of simulated data streams and maps, Fourier series decomposition and angular power spectrum. The contributions within few degrees from the beam centre dominate the GSC near the Galaxy plane. The antenna sidelobes at intermediate and large angles from the beam centre dominate the GSC at medium and high Galactic latitudes. We find a GSC peak at $\\sim 15\\mu$K and a GSC angular power spectrum above that of the white noise for multipoles $\\ell \\lsim 10$ albeit smaller than that of CMB anisotropies by a factor larger than $\\sim 10$. At large multipoles, the GSC affects the determination of CMB angular power spectrum significantly less than other kinds of instrumental systematics, like main beam distortions and $1/f$ noise. Although the GSC is largest at low Galactic latitudes, the contamination produced by far pattern features at medium and high Galactic latitudes, peaking at $\\sim 4\\mu$K, has to be carefully investigated, because the combination of low amplitude of Galaxy emission in those regions with the extremely good nominal {\\sc Planck} sensitivity naturally makes high Galactic latitude areas the targets for unprecedentedly precise estimation of cosmological CMB anisotropy. This paper is based on {\\sc Planck} LFI activities. ",
        "introduction": "After the great success of \\COBE-DMR (Smoot \\etal 1992, Bennet \\etal 1996a, G\\'orski \\etal 1996) which probed the gravitational instability scenario for structure formation through the detection of CMB anisotropies at few degree scales, and the recent balloon-borne experiments at high sensitivity and resolution on limited sky regions (De~Bernardis \\etal 2000, Hanany \\etal 2000), supporting a universe model with $\\Omega_{tot} \\sim 1$ (Lange \\etal 2000, Balbi \\etal 2000, Jaffe \\etal 2000), ultimately, future progresses of the CMB anisotropy cosmology will be based on two space missions, MAP (Microwave Anisotropy Probe) (see Bennet \\etal 1996b) by NASA and {\\sc Planck} by ESA, planned to be launched respectively in the years 2001 and 2007.  In particular, the Low Frequency Instrument (LFI, Mandolesi \\etal 1998) and the High Frequency Instrument (HFI, Puget \\etal 1998) on-board {\\sc Planck} will cover together a wide frequency range (30$\\div$900 GHz) which should significantly improve the accuracy of the subtraction of foreground contamination from the primordial CMB anisotropy, providing at the same time a gold mine of cosmological as well as astrophysical information (e.g. De~Zotti \\etal 1999a and references therein).  To fully reach these scientific goals, great attention has to be devoted to properly reduce and/or subtract all the possible systematic effects. Detailed simulation codes have been developed and are continuously implemented to analyse the impact of several classes of instrumental effects related to the behaviour of the optics, instruments and environment for a wide set of possible scanning strategies (e.g. Burigana \\etal 1998, Delabrouille 1998, Maino \\etal 1999, Mandolesi \\etal 2000a). Ultimately this effort leads to the optimization of the mission design and the production of realistic data streams and simulated maps for data analysis tools as well as a realistic definition of {\\sc Planck} scientific performance.  In particular, the behaviour of the {\\sc Planck} antenna patterns both at intermediate and large angles from the directions of beam centres have to be carefully considered. The requirement on the rejection of unwanted radiation coming from directions far from the optical axis (straylight) is stringent for {\\sc Planck} and does not pertain only the telescope itself, but the entire optical system, including also solar panels, shielding, thermal stability and focal assembly components. The primary sources of error for the LFI are those due to imperfect off-axis rejection by the optical system of radiation from the Sun, Earth, Moon, planets, Galaxy and the spacecraft itself (de~Maagt \\etal 1998). The variations of the spurious straylight signal during the mission introduce contaminations in the anisotropy measurements.  The antenna response features at large angular scales from the beam centre (far sidelobes) are determined largely by diffraction and scattering from the edges of the mirrors and from nearby supporting structures. Therefore they can be reduced by decreasing the illumination at the edge of the primary, i.e. increasing the edge taper, defined as the ratio of the power per unit area incident on the centre of the mirror to that incident on the edge. Of course, the higher is edge taper, the lower is the sidelobe level and the straylight contamination. On the other hand, increasing the edge taper has a negative impact on the angular resolution for the fixed size of the primary mirror (e.g. Mandolesi \\etal 2000a). A trade off between angular resolution and straylight contamination has to be found.  The main aim of this work is to evaluate the impact of the Galactic emission as a source of straylight for {\\sc Planck} LFI. We will then compare it with the effects generated by other kinds of systematics, the main beam distortion introduced by optical aberrations and the $1/f$ noise related to gain fluctuations in LFI radiometers, and with the astrophysical contamination from the Galaxy and the extragalactic sources in the main beam. At LFI frequencies, the Galaxy straylight contamination (GSC) is expected to be particularly crucial at the lowest frequencies, due to the increase of synchrotron absolute emission and anisotropies with the wavelength. For simplicity, we limit our analysis to the case of the 30~GHz channel, but the methods presented here can be extended to higher frequencies.  In Sect.~\\ref{simul} we briefly describe the basic recipes of our simulation code, discussing the geometrical aspects relevant for the full sky convolution, the format of the optical simulation output performed by the ESA (de~Maagt et~al.~1998) that are adopted in the present work, the conversion from data streams to maps and the computation of the Fourier modes and of the angular power spectra, two different estimators of the GSC impact. We estimated the expected GSC on the basis of the antenna integrated response from angular regions at different angles from the beam centre, and on the level of Galaxy emission. In Sect.~\\ref{acc_test} we focus on the integration accuracy of our computations and test the consistency of the code by assuming simplified input maps and antenna patterns. The main results concerning the evaluation of the GSC are presented in Sect.~\\ref{results}. In Sect.~\\ref{comparison} they are compared with the effects introduced by other kinds of instrumental effects and several sources of astrophysical contamination. Finally, we draw our main conclusions in Sect.~\\ref{conclusions}. ",
        "conclusions": "\\label{conclusions}  We have studied the impact of GSC on {\\sc Planck} observations at 30~GHz, by considering different and complementary evaluation approaches: absolute and relative quantification of the impact on scan circle data streams, Fourier decomposition of scan circle signals, computation of maps of GSC and evaluation of their angular power spectra. These different methods allow us to focus on different aspects of GSC.  No relevant differences are found by varying the angle $\\alpha$ between the spin axis and the telescope line of sight from $80^{\\circ}$ to $90^{\\circ}$.  Our simulations show that the GSC peaks at values of about $15 \\mu$K (a value comparable with the sensitivity per pixel), mainly due to the signal entering at few degrees from the beam centre. Such values are found in the regions quite close to the Galactic plane, where in any case the ``direct'' (i.e. observed by the main beam) contamination from the Galaxy prevents an accurate knowledge of CMB fluctuations, as it is next to impossible to remove the Galactic signals to accuracy better than $\\sim$~ \\%. These large contamination values, although critical for CMB anisotropy measurements near the Galactic plane, are not crucial for the determination of Galaxy emission, which is several order of magnitude larger.  By considering all the pixels in the sky, the typical values of GSC are less than the 50\\% of the white noise sensitivity per pixel.  The most crucial contamination derives from the signal entering in the far pattern, in spite of its peak values, of about 4$\\mu$K, smaller than those obtained for the intermediate pattern regions. In fact, although this effect does not seem to be very large in amplitude (indeed, being nominally subdominant to power spectrum of expected receiver noise) it does dominate the GSC at medium and high Galactic latitudes, which are critical regions for the extraction of the best quality results on CMB anisotropy. This could be also critical for the {\\sc Planck} polarization measurements which will take advantage from the two patches close to the ecliptic poles where the sensitivity will be several times better than the average, according to the scanning strategy and the feed array arrangement.  As expected on the basis of the behaviour of Galaxy emission angular power spectrum, the GSC affects the determination of the CMB angular power spectrum mainly in the low multipole region and much less at large multipoles, particularly when compared with the other instrumental effects considered here, the main beam distortion and the $1/f$ noise. The additional noise introduced by the main beam distortion can be in principle subtracted in the data analysis, provided that the beam pattern is accurately reconstructed.  Of course, a substantial improvement in the data analysis is necessary to jointly treat all the systematics, of instrumental and astrophysical origin. From the telescope design point of view, the best optimization of the edge taper requires a trade off between the main beam distortion and the GSC effects."
    },
    "0010/astro-ph0010439_arXiv.txt": {
        "abstract": "We observed the energetic binary Cygnus X-3 in both quiescent and flaring states between 4 and 16 $\\mu$m using the {\\it ISO} satellite. We find that the quiescent source shows the thermal free-free spectrum typical of a hot, fast stellar wind, such as from a massive helium star.  The quiescent mass-loss rate due to a spherically symmetric, non-accelerating wind is found to be in the range (0.4--2.9)$\\times 10^{-4}$ $\\rm M_{\\odot}$ yr$^{-1}$, consistent with other infrared and radio observations, but considerably larger than the $10^{-5}$ $\\rm M_{\\odot}$ yr$^{-1}$ deduced from both the orbital change and the X-ray column density.  There is rapid, large amplitude flaring at 4.5 and 11.5 $\\mu$m at the same time as enhanced radio and X-ray activity, with the infrared spectrum apparently becoming flatter in the flaring state.  We believe non-thermal processes are operating, perhaps along with enhanced thermal emission. ",
        "introduction": "The Infrared Space Observatory {\\it ISO} (Kessler et al.\\ 1996) with its on-board complement of cameras and spectrometers, provided a rare opportunity to study sources at wavelengths not accessible from the ground. We present photometry and images of the Cygnus X-3 field made using ISOCAM (Cesarsky et al.\\ 1996), the camera on {\\it ISO}, in the wavelength range 4 to 16 $\\mu$m.  We report the first spectrum of this source in this wavelength range.  Cygnus X-3 is an enigmatic and unusual X-ray binary which was discovered $>$30 years ago (Giacconi et al.\\ 1967). Observations of it have been made from radio to gamma rays and in every wavelength bands the system exhibits unusual behaviour.  In X-rays and infrared, the emission is partially modulated by a 4.8 h period (Mason, C\\'{o}rdova \\& White 1986), and this is believed to be the orbital period of the system.  In 1991, van Kerkwijk et al. observed Cyg X-3 in the infrared with the CGS4 spectrometer on UKIRT and discovered He\\,{\\sc i} and He\\,{\\sc ii} together with N emission lines which were modulated at the 4.8 h period (van Kerkwijk et al.\\ 1992; van Kerkwijk et al.\\ 1996).  They concluded that the lines originated from the companion star, either a Wolf--Rayet of the WN7 class, or an earlier WN4--6 in an outburst phase.  This was investigated in more detail by Fender, Hanson \\& Pooley (1999, hereafter FHP99), who took spectra during a variety of activity states and confirmed that the underlying quiescent spectral type is of an earlier WN4--5 Wolf--Rayet star.  Galactic Wolf--Rayet stars have high velocity winds, up to 2500 km s$^{-1}$ (Crowther, Hillier \\& Smith 1995; Bohannan \\& Crowther 1999). The Wolf--Rayet wind from Cyg X-3 is believed to be the main contributor in providing mass and angular momentum loss which increase the orbital period ($\\dot{P} = -6.6 \\times 10^{-10}$ s s$^{-1}$, Kitamoto et al.\\ 1995).  This period increase is consistent with a mass-loss rate of $10^{-6} M_{\\rm T}$ $\\rm M_{\\odot}$ yr $^{-1}$, where $M_{\\rm T}$ is the total mass in the system (Kitamoto et al.\\ 1995; Matz 1997).  Kitamoto et al.\\ (1994) modelled changes in the radio activity, infrared and X-ray states caused by changes in the density of the WR wind (Table \\ref{wind}).  The multi-wavelength predictions of Kitamoto et al.\\ (1994) were tested by FHP99, who find that during an outburst the infrared {\\it K} band spectrum becomes dominated by strong twin-peaked \\hbox{He\\,{\\sc i}} emission lines. These lines show strong {\\it V}/{\\it R} variability correlated with the orbital phase.  Due to the small size of the binary ($\\sim 5\\;R_{\\odot}$ separation, 4.8 hour period) the observed line flux could not arise from a region of this size as the environment would be too hot.  FHP99 conclude that the He\\,{\\sc i} emission lines are emitted from a region which is significantly larger than the binary separation, and that the variability of these lines indicate an asymmetric emission region, possibly from a flattened disc-like wind.  \\begin{table} \\caption{Radio, infrared and X-ray variability based on a changing density wind, from Kitamoto et al.\\ (1994).  The term `modulation' refers to the continuum.  The states marked with an asterisk (*) have been expanded on and confirmed by Fender et al.\\ (FHP99).} \\begin{tabular}{llll}\\hline Regime \t\t& Emission \t& Rare wind \t& Dense wind \\\\ \\hline Radio\t\t& Continuum*\t& Quiescence\t& Flaring \\\\ Infrared\t& Continuum*\t& Weak\t\t& Strong \\\\ & Line strength*& Weak\t\t& Strong \\\\ & Ionisation*\t& High\t\t& Low\t\\\\ & Modulation\t& Large\t\t& Small\t\\\\ X-ray\t\t& Luminosity*\t& Low\t\t& High\t\\\\ & Spectrum\t& Hard\t\t& Soft \t\\\\ \\hline \\end{tabular} \\label{wind} \\end{table}  The compact object in Cyg X-3 is less well defined.  Cherepashchuk \\& Moffat (1994) suggest that the high X-ray luminosity (greater than the Eddington luminosity for a 1 ${\\rm M_{\\odot}}$ object) implies that the compact object is a black hole.  Schmutz, Geballe \\& Schild (1996) related time-variations of infrared spectral lines to an orbital velocity and derived a mass for the compact object in the range of 7--40 $\\rm M_{\\sun}$, assuming realistic values for the mass of the Wolf--Rayet star and the inclination of the system.  However, their model does not account for all the relative phasing of the X-rays and infrared lines.  Ergma \\& Yungelson (1998) point out that if the progenitor to the binary is two massive helium stars then the mass-loss and common envelope phase of these two stars prevents the formation of a black hole \\& Wolf--Rayet binary with an orbital period less than several days.  This ambiguity does not resolve itself if the compact object is a neutron star; Ergma \\& Yungelson (1998) also comment that the propeller motion of a spinning neutron star would prevent wind accretion in the system, thus removing the jet-creation mechanism.  With a high mass, high mass-loss (wind) object in close proximity to a compact object, a high degree of interaction takes place.  Radio observations at wavelengths ranging from 1--20 GHz have shown Cyg X-3 to undergo many states of emission and are used to define the state of Cyg X-3.  For most of its time, the system occupies a {\\it quiescent} radio state, with 8.3 GHz flux around 100 mJy (Waltman et al.\\ 1994). {\\it Major flares} (8.3 GHz flux $>$ 1 Jy) occur at random intervals every 18 months or so and relaxation after a flare can last up to several weeks (Waltman et al.\\ 1995).  More common are {\\it minor flares} (8.3 GHz flux 200--1000 mJy) and these usually occur during a flaring period typically several weeks duration, with several small flares occurring in close proximity to each other (Waltman et al.\\ 1996).  Minor flares can reach peak emission in a few hours and decay over several hours (Ogley 1998).  The minor and major flares have been shown to be intrinsically different in their production, rise and relaxation times (Newell 1996).  Preceding a major flare, the system enters a quenching state with fluxes at 8.3 GHz dropping to around 10 mJy.  The system returns to quiescence then flares (Waltman et al.\\ 1996).  The main mechanism for radio flares is thought to be associated with the acceleration of relativistic electrons in the base of jets. Synchrotron emitting plasmons have been detected, travelling at $\\simeq$ 5 mas d$^{-1}$ on both arcsec and mas scales (Spencer et al.\\ 1986; Mioduszewski et al.\\ 1998).  There is no confident determination of the velocity of ejecta: while a velocity of 0.3$c$ is generally taken as the speed on the larger arcsec scale (e.g. Spencer et al.\\ 1986), other highly-relativistic interpretations are possible on the mas scale (e.g. Ogley, Bell Burnell \\& Newell 1997; Newell, Garrett \\& Spencer 1998).  The Wolf--Rayet contributes not only to the enigmatic qualities of the system, but also a significant and varying opacity.  The true nature of the secondary is a top priority for the understanding of the system.  Observations at the {\\it ISO} wavelengths provide a unique opportunity to study the Wolf--Rayet wind, and spectra obtained can track its true nature. ",
        "conclusions": "We report the first quiescent and flaring 4-16 $\\mu$m observations of Cygnus X-3.  The quiescent spectrum is a continuation of the shorter wavelength IR spectrum, and fits well the WB75 model for thermal free-free emission from a wind.  We obtain a mass loss rate, assuming spherical symmetry of $\\sim 10^{-4}$ $\\rm M_{\\odot}$ yr$^{-1}$, and note that while this is consistent with other determinations, it is an order of magnitude larger than both the mass loss rate required to derive the change in orbital period (assuming $M_{\\rm T}$ of the order $10\\;\\rm M_{\\odot}$), and the mass loss rate expected from the hydrogen column density to the X-ray source.  These measurements are based on assumptions, which have been scrutinised carefully; while we can identify a number of areas where factors of a few can be gained, we have been unable to identify a single area where a factor of ten can be found.  We agree with the conclusion of FHP99 that the wind is most likely spherically asymmetric, but if this is the only effect operating, large disc density enhancement factors are required.  There are several phenomena which could cause $P_{\\rm orb}$ to be underestimated; we regard this determination of $\\dot{M}$ to be less reliable.  During radio and X-ray outburst, the broadband infrared spectrum shows rapid variability and a spectral flattening associated with increased flux densities. This is in agreement with what is seen in the near-IR during rapid (minutes) flaring events (Fender et al.\\ 1996). Similar (but not identical) near-IR events observed from GRS 1915+105 appear to be associated with jet formation episodes. The emission mechanism responsible for this flat spectral component remains uncertain; both free-free and synchrotron emission have been proposed already in the literature, and await detailed studies to test their viability. Whatever the origin of this component, it seems likely that it is associated with hot gas in the jet from the system, which naturally explains its emergence during periods of enhanced radio (jet) activity."
    },
    "0010/astro-ph0010169_arXiv.txt": {
        "abstract": "{A 0.242 $M_{\\sun}$ object that finally becomes a helium white dwarf is  evolved from Roche  lobe detachment down  to very low luminosities.  In doing  so, we  employ our  stellar code  to which we have  added a set  of routines that  compute the effects due  to   gravitational  settling,   and  chemical   and  thermal diffusion. Initial model is constructed by abstracting mass  to a 1 $M_{\\sun}$ red giant branch  model up to the moment at  which the  model  begins  to  evolve  bluewards.  We  find that element diffusion introduces noticeable changes in the internal structure of  the   star.  In   particular,  models   undergo  {\\it  three} thermonuclear  flashes  instead  of  one  flash as we found with the standard treatment.  This fact  has a  large impact  on the total mass fraction of hydrogen left in the star at entering the  final cooling track. As  a result, at  late stages of  evolution models with diffusion are characterized by a much smaller nuclear energy release, and  they evolve  significantly faster  compared to those found with  the standard treatment.  We  find  that  models  in  which diffusion is considered predict evolutionary  ages   for  the   white  dwarf   companion  to  the millisecond  pulsar  PSR  B1855+09  in  good  agreement  with the spin-down age of the pulsar. \\endabstract ",
        "introduction": "\\label{sec_intro}  Low mass, helium white dwarfs  (WDs) have drawn the attention  of many researchers who have devoted a large effort to their  study. Currently, it  is well  established that  helium WDs  should have been formed  in a  binary system  because an  isolated star would reach such a configuration in  a time-scale much longer than  the present age of our Universe. Examples of recent studies on helium WD evolution are Althaus \\& Benvenuto (1997; 2000), Benvenuto  \\& Althaus (1998), Hansen  \\& Phinney (1998),  Driebe et al.  (1998) and Sarna, Antipova \\& Ergma (1999).  A crucial  issue in  the study  of helium  WDs is  the amount  of hydrogen  left  atop  the  WD  after  the  end  of  mass transfer episodes.  Sarna  et  al.  (1999)  have  presented  detailed calculations of the pre -  WD evolution during the mass  transfer stage  in  a  close   binary  system.  They  found   that,  after detachment,  helium  WDs  are  surrounded  by  massive   hydrogen envelopes of  0.01 -  0.06 $  M_{\\sun}$ with  a surface  hydrogen abundance  by  mass  of  $X_{H}=$  0.35  -  0.5. Massive hydrogen envelopes have  also been found  by  Driebe et  al. (1998), who have mimicked the binary evolution by abstracting mass from a  $1 M_{\\sun}$ model at an adequate rate.  Binary systems composed by a  helium WD and a millisecond  pulsar offer us  a  good  opportunity  to  test  the  predictions  of evolutionary calculations. Millisecond pulsars are thought to  be recycled during the mass transfer stage (Bhattacharya \\& van  den Heuvel 1991). When mass transfer ends, the pulsar begins to  spin down. If we set zero ages at the end of such mass transfer,  then the ages of the WD and  the pulsar component should be the  same. If the pulsar radiates as a dipole and its current period ($P$) is much longer than the initial one, then the spin-down of the pulsar  is given by $t_{PSR}= P/2\\dot{P}$, where $\\dot{P}$ is the rate of change of $P$.  Recently, van Kerkwijk et al. (2000) have optically detected  the low-mass component of the PSR B1855+09 system and determined  its effective temperature to  be $4800 \\pm  800$ K. In  addition, the mass of the WD is accurately  known thanks to the measure of  the Shapiro  delay   of  the   pulsar  signal   and  results   to  be $0.258^{+0.028}_{-0.016} M_{\\sun}$. The age  derived for the  WD component results to be twice as long as the pulsar spin-down age  ($\\approx 5  Gyr$)  when  using the models without diffusion of Driebe et al. (1998).  Helium WD evolution  in presence of  element diffusion has  been considered by Iben \\&  Tutukov (1986), who computed  the evolution of a $0.296 M_{\\sun}$  object throughout the entire  evolutionary stages  and  found  two  thermonuclear  flashes. Remarkably, they found that after flashes the  amount of hydrogen in the  envelope of the model was unable to support an appreciable amount of  hydrogen burning reactions, enforcing a  fast evolution for the  resulting WD star.  It is the  purpose of this  work to study  the impact of  element diffusion on the  structure and evolution  of a $0.242  M_{\\sun}$ helium WD star. This mass  value is  representative of  the companion to  PSR B1855+09.  In doing  so, two  sequences are computed: one  with and  the other  without diffusion.  Thus, the resulting differences should be due  to the allowance for such  a process. ",
        "conclusions": "\\label{secc:conclu}  From the results  presented in the previous section,  it is clear that element  diffusion  introduces   important  changes  in  the  internal structure of  the star.  Our  present detailed results  indicate that, contrary  to Driebe  et al.  (1998) and  Sch\\\"onberner et  al.  (2000) assertions,  nuclear  reactions are  not  an  important ingredient  in determining the  age of the  star during the final  cooling evolution. This conclusion is valid at  least for the stellar mass value analyzed in this paper.  Although the hydrogen mass fraction at  the end of the computed stages  is very  similar in both  sequences with  and without diffusion,  there is a  large difference  in the  total time  spent in reaching such  a low  luminosity in each  case. While  standard models without diffusion  spend about  30 Gyr in  reaching $\\log{L/L_{\\sun}}= -5$, models with diffusion get such a luminosity in about 15 Gyr.  As stated  in the introductory  section, previous models  available in the  literature (those in  which hydrogen  burning is  dominant during most of the  cooling stages) lead to a  strong discrepancy between the ages of PSR  B1855+09 and its helium WD  companion.  These ages should be, on very general grounds,  the same.  The results presented in this paper  strongly indicate  that {\\it  such  a discrepancy  is a  direct consequence  of  ignoring  element  diffusion  in  the  stellar  model calculations}. If we  include it properly, ages naturally  come into a nice agreement. Consequently, we do not have to invoke any ad-hoc mass loss as proposed  by Sch\\\"onberner et al. (2000)  or exotic mechanisms to  account  for the  observational  situation  of  the binary  system containing PSR B1855+09.  It is worth  noting that the structure of the  models belonging to the final cooling branch we have found  in the present paper for the 0.242 $M_{\\sun}$ helium  WD strongly resemble  those we have assumed  in our previous  work on  helium WDs  with hydrogen  envelopes  (Benvenuto \\& Althaus 1998).  In fact, if  we assume $M_{H}/M_{*}= 2 \\times 10^{-3}$ the  agreement  with  the  present  results  is  excellent,  not  only regarding the cooling  evolution but also with respect  to the nuclear activity.   In  our  opinion,  the present  detailed  results  largely justify the assumptions we have made in our quoted previous paper.   We  want  to  emphasize  that  our  results  strongly  resemble  those presented sometime ago  by Iben \\& Tutukov (1986)  (IT86) for the case of a  $0.296 M_{\\sun}$  model.  They found  two flashes  and suspected that the  second one was  due to diffusion,  but because they  did not compute models without diffusion,  they were not conclusive about this point. Here we confirm that diffusion is the reason for the occurrence of  additional flashes.  IT86  also  found that  the  final amount  of hydrogen left in  the WD is small enough to  ultimately enforce a fast cooling.  Although  their  model  is  more massive  than  ours,  which prevents from a quantitative comparison between both set of calculations, the role of  hydrogen burning at the final cooling  stages is the same we have found.  There  is nevertheless an important difference between the work of IT86 and our study.  In fact, IT86 considered that shortly after the  second flash,  when the model  gets its maximum  radius, it again overfills its  Roche lobe and losses a tiny  part of its stellar mass. Obviously, this mass belongs to the hydrogen envelope and such a loss has a large impact on the final value of $M_{H}/M_{*}$ in the WD. This is ultimately responsible for  the fast cooling occurring at late stages of  evolution. The point to  emphasize here is that  we did not consider such a mass transfer  during our calculations, and some words about this  point are  in order. If  such a transfer  actually occurs, $M_{H}/M_{*}$  should  drop  below   the  values  we  presented  above (eventually  to  zero).  This  would  have  an  impact on  the  global structure of the  WD. Such a situation may be  considered in the frame of the  models presented in  Benvenuto \\& Althaus (1998).   Apart from the structural differences we do not expect an important change in the time-scale  of  cooling.  Notice  that  in  the  models  presented  in Benvenuto \\&  Althaus, the time  spent in reaching a  given luminosity value is rather independent from  the exact value of $M_{H}/M_{*}$ (if it is in the range of $1 \\times 10^{-3}$ to zero). Thus, the occurence of such a  late mass transfer should not  modify the conclusions about the  age of  the final  WD  model and  its better  agreement with  the observations related to the ages of the PSR B1855+09 system."
    },
    "0010/astro-ph0010443_arXiv.txt": {
        "abstract": "Quintessence models with a dark energy generated by pseudo Nambu--Goldstone bosons provide a natural framework in which to test the possibility that type Ia supernovae luminosity distance measurements are at least partially due to an evolution of the sources, since these models can have parameter values for which the expansion of the Universe is decelerating as well as values for which it is accelerating, while being spatially flat in all cases and allowing for a low density of clumped matter. The results of a recent investigation \\cite{NW} of current observational bounds which allow for SNe Ia source evolution are discussed. It is found that models with source evolution still favour cosmologies with an appreciable amount of acceleration in the recent past, but that the region of parameter space which is most favoured shifts significantly. ",
        "introduction": " ",
        "conclusions": ""
    },
    "0010/astro-ph0010505_arXiv.txt": {
        "abstract": "We present the results of a wide-field survey for H$\\alpha$-emitting galaxies in the  cluster AC\\,114 at $z=0.32$. Spectra centred on H$\\alpha$ at the cluster redshift have been obtained for 586 galaxies to $I_{tot}\\sim 22$ out to a radius of $\\sim 2\\,h_{50}^{-1}$\\,Mpc. At most, only $\\sim 10$\\% of these were found to be H$\\alpha$--emitting cluster members. These objects are predominantly blue and of late-type spiral morphology, consistent with them hosting star formation. However, $\\sim 65$\\% of the cluster members classified morphologically as spirals (with {\\it HST}), have no detectable H$\\alpha$ emission; star-formation and morphological evolution in cluster galaxies appear to be largely decoupled. Changes in the H$\\alpha$ detection rate and the strength of H$\\alpha$ emission with environment (as traced by local galaxy density) are found to be weak within the region studied. Star formation within the cluster members is also found to be strongly {\\it and} uniformly suppressed with the rates inferred from the H$\\alpha$ emission not exceeding 4\\,M$_{\\odot}$\\,yr$^{-1}$, and AC\\,114's H$\\alpha$ luminosity function being an order of magnitude below that observed for field galaxies at the same redshift. None of the galaxies detected have the high star formation rates associated with `starburst' galaxies; however, this may still be reconciliable with the known ($8\\pm 3$\\%) fraction of `post-starburst' galaxies within AC\\,114, given the poorly determined but short lifetimes of starbursts and the possibility that much of the associated star formation is obscured by dust. ",
        "introduction": "Clusters of galaxies provide a powerful laboratory for studying the evolution of galaxies. In the hierarchical picture of galaxy formation, clusters grow in mass by accreting galaxies from the surrounding field. Once these galaxies enter the dense environment of the cluster, star formation is suppressed and the galaxies evolve to become early-type galaxies.   Many processes operate to make the cluster environment hostile to star forming galaxies. The most important processes include:  mergers and violent encounters between galaxies (Lavery \\& Henry, 1994); tidal stripping and `harassment' (Byrd \\& Valtonen 1990, Moore et al.\\ 1996); ram pressure stripping of the gas disk (Gunn \\& Gott 1972, Abadi et al.\\ 1999, Quilis et al.\\ 2000); and simply the removal of any gas reservoir surrounding each galaxy (thus preventing replenishment of the gas disk; Larson, Tinsley \\& Caldwell 1980, Balogh et al.\\ 1999).  Since clusters can be observed out to $z>1$, we can use the look-back times of the distant systems to directly investigate the evolution of cluster galaxies. Butcher \\& Oemler (1978a) and many subsequent authors have noted that the fraction of blue galaxies in cluster cores increases with redshift. Spectroscopic surveys have confirmed that these blue galaxies are indeed cluster members (e.g.\\ Dressler \\& Gunn 1982, 1983; Couch \\& Sharples 1987, hereafter CS; Lavery \\& Henry 1988; Dressler et al.\\ 1999) and have identified a conspicuous population of galaxies with no detectable emission (e.g.\\ [O{\\sc ii}]$\\lambda$3727) but with abnormally strong Balmer absorption lines (termed `E+A' galaxies by Dressler \\& Gunn, 1982 and `k+a' or `a+k' galaxies by Dressler et al.\\ 1999). It has also been proposed that the systematically higher blue galaxy fractions in distant clusters is coupled with a reduction in the S0 galaxy populations in these systems [identified from imaging with the {\\it Hubble Space Telescope} ({\\it HST})] in comparison to their local counterparts (Dressler et al.\\ 1997; Couch et al.\\ 1998, hereafter C98).  The evolution of the galaxy populations in distant clusters may be explained as the result of the increased activity of field galaxies at intermediate redshift (Lilly et al.\\ 1996), combined with a higher accretion rate in these more distant systems (Bower 1991, Kauffmann 1996, Diaferio et al.\\ 2000): not only are the galaxies forming stars more rapidly when they arrive in the cluster, but they also arrive at a higher rate.  However, while an overall picture of the evolution of cluster galaxies is emerging, we have no detailed understanding of the physical mechanism responsible for the suppression of star formation and its effect on other properties such as mass and morphology.  The key issues can be summarised by three questions:  (1)\\,By how much and how quickly is star formation suppressed in the infalling field galaxies? (2)\\,What is the radial dependence of the star formation suppression? (3)\\,Does the cluster simply suppress star formation or does it first promote a burst of star formation?  The ability of current observations to address these issues is, however, seriously hindered by most distant cluster studies having largely concentrated on the central {\\it core} ($r<1\\,h_{50}^{-1}$\\,Mpc) regions in which the BO-effect was originally discovered: spectroscopic data of the quantity and quality needed to track in detail the star formation characteristics of galaxies well outside the core remain in short supply. Progress towards redressing this problem has been made by Balogh et al.\\ (1997, 1998), who used the CNOC Survey data (Yee, Ellingson \\& Carlberg 1996) to trace the spectral properties of galaxies out to and beyond the virial radius in 15 X-ray bright clusters at $0.18<z<0.55$. Using the equivalent width of the [O{\\sc ii}]$\\lambda$3727 emission line as a star formation indicator, they found a smooth decline in the star formation rate from the field into the interior of their clusters, an effect which they claimed was likely to be independent of the morphology--density relationship established by Dressler (1980). That this decline should extend so far out in their clusters is surprising, since simple cluster mass models would suggest that galaxies at and beyond the virial radius should have only become recently bound to the cluster.  While the CNOC study is an important first step to understanding the star formation patterns at large radii in distant clusters, a number of issues remain. In particular, Poggianti et al.\\ (1999; hereafter P99) and Smail et al.\\ (1999) have questioned the reliability of [O{\\sc ii}]$\\lambda$3727 as a star formation indicator. P99 have noted that there are insufficient bright, star-forming galaxies in their clusters to explain the abundance of a+k/k+a objects. Instead, they suggest that the progenitors of the a+k/k+a galaxies are e(a) emission line objects (which show both weak [O{\\sc ii}] emission and strong $H\\delta$ absorption) in which the bulk of the star formation is obscured by dust (see also Poggianti \\& Wu 2000).  Furthermore, Smail et al.\\ (1999) suggest,  on the basis of radio continuum observations, that a significant number of the apparently post-starburst (a+k/k+a) galaxies in the core of Abell~851 ($z=0.41$) are in fact on-going star burst galaxies in which the star formation (as measured by the [O{\\sc ii}] emission line) is completely hidden by dust.  Motivated by these issues and the clear need to explore further the star formation behaviour at large cluster-centric radii, we have embarked on an intensively--sampled, wide--field ($\\sim 4\\,h_{50}^{-1}$\\,Mpc) H$\\alpha$ survey of galaxies in 3 {\\it optically--selected} clusters at $z\\sim 0.3$. These clusters -- AC103, AC118, and AC\\,114 -- have been subjected to extensive imaging and spectroscopy both from the ground and from space ({\\it HST} and {\\it ROSAT}), the results of which have shown them to be quite diverse in their structural and X-ray properties. They therefore not only provide useful targets for studying the cluster-to-cluster variations in star forming galaxies, but also provide an important test of how the global star formation history is dependent on cluster morphology. By exploiting recent innovations in telescope instrumentation, our program is able to survey large numbers of ($\\sim 10^3$) galaxies in each cluster field for H$\\alpha$ emission -- the most direct indicator of star formation -- down to very low, sub--1\\,M$_{\\odot}$\\,yr$^{-1}$ levels. Hence it provides a highly complete inventory of star-forming galaxies with sufficient statistical sampling at large cluster-centric radii to robustly determine cluster-to-cluster trends.  In this paper we present the first observations to be made as part of this program: H$\\alpha$ spectroscopy of $\\sim 600$ galaxies in the field of AC\\,114 at $z=0.32$. In the next section we describe our observations and outline the novel techniques that allow us to obtain spectra for this number of galaxies in just 4\\,hrs on a 4-m telescope. In section 3, we provide details of our reduction techniques and the detection and measurement of H$\\alpha$ emission in our spectra. The numbers of H$\\alpha$ emitting galaxies found, their distribution within the cluster, and the star formation rates and luminosity function associated with this population are then examined in section 4. We then discuss our results in section 5 and summarise our conclusions in section 6. ",
        "conclusions": "In this paper we have presented  H$\\alpha$ spectroscopy for 586 unique galaxies to $I\\sim 22$ over a $3.1\\times 3.1\\,h_{50}^{-1}$\\,Mpc field centred on the rich galaxy cluster AC\\,114 at $z=0.32$. The major conclusions to be drawn from this survey of H$\\alpha$ emitting galaxies within this cluster can be summarised as follows:  \\begin{itemize} \\item The incidence of H$\\alpha$ emission in cluster galaxies is {\\it low} with 4\\% of the galaxies surveyed here having highly secure detections and the total being no more than 9\\% if less secure detections are included.  \\item The galaxies in which H$\\alpha$ emission is observed, have morphologies and colors that are largely consistent with them hosting star formation activity. From the small sub-sample of H$\\alpha$ emitters which have {\\it HST} morphologies, the majority (67\\%) are found to be spirals predominantly of later type. H$\\alpha$ emission is preferentially found in {\\it bluer} galaxies although we note that the detection rate does not drop to zero for the reddest galaxies in the cluster.  \\item In the central, $r\\leq 1\\,h_{50}^{-1}$\\,Mpc region of AC\\,114 where our {\\it HST} imaging overlaps with this study, approximately two-thirds of the galaxies with spiral morphology were undetected in H$\\alpha$. This provides further evidence that morphological and star-formation evolution in clusters, is largely decoupled.  \\item Apart from the complete absence of H$\\alpha$--emitting galaxies within a radius of 250\\,$h_{50}^{-1}$\\,kpc from the centre of AC\\,114, the H$\\alpha$ detection rate shows no detectable radial variation out to $\\sim 2\\,h_{50}^{-1}\\,$Mpc. However, any radial trend will clearly be weakened by the highly elongated structure of this cluster. A weak anti-correlation between detection rate and {\\it local} galaxy density is found, indicating that the incidence of star--forming cluster galaxies increases by roughly a factor of 2 in going from the highest ($\\sim 250\\,h_{50}^{2}$\\,gals\\,Mpc$^{-2}$) to the lowest ($\\sim 50\\,h_{50}^{2}$\\,gals\\,Mpc$^{-2}$) density regions probed within AC\\,114.  \\item The rate of star formation inferred from the H$\\alpha$ luminosities measured for the detected galaxies is found to be uniformly low across the entire region of the cluster studied here. The maximum SFR observed is $\\sim 4$M$_{\\odot}$\\,yr$^{-1}$, with the majority of galaxies having rates $< 1.5$M$_{\\odot}$\\,yr$^{-1}$. No discernible variation in SFR with `environment' (as traced by local galaxy density) was found.  \\item A comparison of the H$\\alpha$ luminosity function for galaxies in AC\\,114 with those in the field at a similar redshift shows the former to be highly suppressed. At the same H$\\alpha$ luminosity, AC\\,114's luminosity function falls below that of the field by {\\it an order of magnitude}. Furthermore, galaxies with luminosities $L({\\rm H}\\alpha)>10^{41}\\,h_{50}^{-2}$\\,erg\\,s$^{-1}$ (and hence a SFR$>4$M$_{\\odot}$\\,yr$^{-1}$), are absent from the cluster environment.  \\end{itemize}"
    },
    "0010/astro-ph0010415_arXiv.txt": {
        "abstract": "We fit the BOOMERANG, MAXIMA and COBE/DMR measurements of the cosmic microwave background anisotropy in spatially flat cosmological models where departures from standard recombination of the primeval plasma are parametrized through a change in the fine structure constant $\\alpha$ compared to its present value. In addition to $\\alpha$ we vary the baryon and dark matter densities, the spectral index of scalar fluctuations, and the Hubble constant. Within the class of models considered, the lack of a prominent second acoustic peak in the measured spectrum can be accomodated either by a relatively large baryon density, by a tilt towards the red in the spectrum of density fluctuations, or by a delay in the time at which neutral hydrogen formed. The ratio between the second and first peak decreases by around 25\\% either if the baryon density $\\Omega_bh^2$ is increased or the spectral index $n$ decreased by a comparable amount, or if neutral hydrogen formed at a redshift $z_*$ about 15\\% smaller than its standard value. We find that the present data is best fitted by a delay in recombination, with a lower baryon density than the best fit if recombination is standard. Our best fit model has $z_*= 900$, $\\Omega_bh^2=0.024$, $\\Omega_mh^2=0.14$, $H_0=49$ and $n=1.02$. Compatible with present data at 95\\% confidence level $780< z_*<1150$, $0.018<\\Omega_bh^2<0.036$, $0.07< \\Omega_m h^2 < 0.3$ and $0.9<n<1.2$. ",
        "introduction": "Measurements of the Cosmic Microwave Background (CMB) anisotropy provide information about the physical conditions in the universe right before decoupling of matter and radiation, and allow for a precise determination of many cosmological parameters. The recently published results of the BOOMERANG~\\cite{deBernardis} and MAXIMA~\\cite{MAXIMA} experiments provide strong confirmation of the presence of the first acoustic peak in the anisotropy angular power spectrum at the degree scale, which suggests that the Universe is very close to spatial flatness. BOOMERANG and MAXIMA also provided high quality data at smaller angular scales, which do not reveal the presence of a prominent second acoustic peak.  The location, height and width of the first peak is in excellent concordance with the simplest spatially flat cosmological models with a nearly scale invariant spectrum of adiabatic density fluctuations (such as those motivated by generic inflationary models) and values of all cosmological parameters in good agreement with independent observations.  The lack of a prominent second acoustic peak requires instead some sort of departure from the simplest models or most likely values of some cosmological parameters. Three possibilities have been highlighted~\\cite{Hu,TZ,Peebles,White,Lange}: i) the power spectrum of primordial density perturbations is tilted in the direction that suppresses small scale fluctuations; ii) the baryonic matter density is slightly above the upper value expected from big-bang nucleosynthesis; iii) the formation of atomic hydrogen was delayed by some mechanism that perturbed the standard ionization history of the Universe.  The first mechanism reduces the height of the second peak simply because a spectrum tilted towards the red has less power at smaller angular scales. The second and third mechanisms work because the ratio between the heights of the second and first peaks decreases when the ratio $R=\\frac{3}{4}  \\rho_ b/\\rho_\\gamma$ between baryon and radiation energy-densities at decoupling increases~\\cite{Hu95}. Either if $\\rho_b$ is increased or if recombination is shifted to lower redshifts $R(z_*)$ increases and the relative height of the second  peak decreases. The shift of $z_*$ to lower redshifts also shifts the location of the peaks in the angular power spectrum towards larger angular scales.  Peebles, Seager and Hu~\\cite{Peebles} developed a picture where very early sources of  Ly $\\alpha$ resonance radiation provide a delay in recombination rapid enough to avoid excessive dissipation of the first acoustic peak, and cause a 10\\% reduction in the relative height of the second peak along with a 5\\% shift in the location of the first peak. Their model is essentially parametrized by the rate of excess of radiation by the early sources from that in the primeval plasma.  In this paper we further explore the compatibility of present CMB anisotropy measurements with a non standard recombination history. We parametrize departures from standard recombination through a change in the fine structure constant around decoupling compared to its present value. We then compute constraints in four cosmological parameters plus the fine structure constant from the COBE/DMR, BOOMERANG and MAXIMA data on CMB anisotropy.  In section \\ref{alfa} we discuss the effects upon the CMB anisotropy spectrum of a variation in the fine structure constant at decoupling compared to its present value, and argue that they  mimic the effects of other physical mechanisms that may change the recombination history. In section \\ref{data} we fit the COBE/DMR, BOOMERANG and MAXIMA data within spatially flat cosmological models allowing for variations of the fine structure constant, the baryon and dark matter energy densities, the Hubble constant, and the scalar fluctuations spectral index. We place bounds on the allowed variation of the fine structure constant, which translate into bounds on departures from standard recombination. In section \\ref{conclusiones} we discuss the results. ",
        "conclusions": "We have modeled non standard recombination through a variation of the fine structure constant $\\alpha$ around decoupling compared to its present value $\\alpha_0$. We performed a maximum likelihood analysis of the BOOMERANG, MAXIMA and COBE/DMR measurements of the cosmic microwave background anisotropy in a grid of spatially flat cosmological models with five parameters varied independently: $\\alpha$, $\\Omega_bh^2$ $\\Omega_mh^2$, $h$ and $n$.  The recombination redshift scales approximately as $z_*=1080 [1+2(\\alpha - \\alpha_0)/\\alpha_0]$.  The lack of a prominent second acoustic peak in the data suggests that the best fit should have, within the family of models considered, either a relatively large baryon density compared to the nucleosynthesis value $\\omega_b=0.019$, a tilt away from scale invariance towards the red in the spectrum of density fluctuations, or a delay in the time at which the CMB decoupled from matter. The best fit is that recombination was largely delayed to a redshift around $z_*=900$, with a nearly scale invariant spectrum of fluctuations, a low value of the Hubble constant, $H_0=49$, and a baryon density $\\Omega_bh^2=0.024$. A delay in recombination allows a good fit to the data with a lower baryon density as compared to a standard ionization history. Acceptable fits with a baryon density as low as $\\Omega_bh^2=0.018$ can be achieved with longer recombination delays. Good fits with such low baryon densities can not be achieved with just a tilt in the spectral index and a standard recombination history. Recombination can not be delayed beyond $z_*\\approx 780$ at 95\\% confidence level and within the cosmological models considered. Notice that in our grid of models $H_0>45$. Recombination may eventually be delayed to lower values of $z_*$ at the same confidence level if this (realistic) prior assumption upon $H_0$ were lifted.  It is instructive to discuss these results from a semianalytic approach. It has been shown~\\cite{Hu00} that the power spectrum of CMB anisotropies up to the third acoustic peak can be conveniently characterized by four observables, namely: the position of the first acoustic peak $l_1$, the height of the first peak relative to COBE normalization $H_1$, the height of the second peak relative to the first $H_2$ and the height of the third peak relative to the first $H_3$. We have performed a semianalytic fit around our best fit model of eq. (\\ref{BF}) assuming a linear dependence on the parameters, with the approximate result:  \\begin{eqnarray} \\frac{\\Delta l_1}{l_1}&=& 1.4 \\frac{\\Delta \\alpha}{\\alpha} + 0.05 \\frac{\\Delta \\omega_b}{\\omega_b} - 0.14 \\frac{\\Delta \\omega_m}{\\omega_m} -0.17  \\frac{\\Delta H_0}{H_0}+0.19 \\Delta n \\label{l1}\\\\ \\frac{\\Delta H_1}{H_1}&=& 0.9 \\frac{\\Delta \\alpha}{\\alpha} + 0.44\\frac{\\Delta \\omega_b}{\\omega_b} -0.47 \\frac{\\Delta \\omega_m}{\\omega_m}-0.13 \\frac{\\Delta H_0}{H_0}+2.6 \\Delta n \\label{H1}\\\\ \\frac{\\Delta H_2}{H_2}&=&3.2 \\frac{\\Delta \\alpha}{\\alpha} - 0.85\\frac{\\Delta \\omega_b}{\\omega_b} -0.04 \\frac{\\Delta \\omega_m}{\\omega_m}+ 0.96\\Delta n \\label{H2}\\\\ \\frac{\\Delta H_3}{H_3}&=&2.3 \\frac{\\Delta \\alpha}{\\alpha} - 0.35 \\frac{\\Delta \\omega_b}{\\omega_b}+0.38 \\frac{\\Delta \\omega_m} {\\omega_m}+1.2 \\Delta n \\label{H3} \\end{eqnarray}  The peaks location and heights of the best fit model are given by $l_1=206.5$, $H_1=7.8$, $H_2=0.37$, $H_3=0.43$.  Equation~(\\ref{H2}) condenses most of the discussion we already made in sections~\\ref{intro} and~\\ref{alfa} about the effects that can accommodate the absence of a prominent second peak in the present CMB anisotropy data. Indeed, the relative height of the second acoustic peak compared to the first decreases by about 10\\% if $\\alpha$ decreases around 3\\%. Alternatively, the same reduction in $H_2$ can be obtained by an increase of around 10\\% in the baryon density, or a 10\\% tilt towards the red in the spectral index of density fluctuations. $H_2$ is instead quite insensitive to $\\omega_m$ and $H_0$.  Notice from eq.~(\\ref{l1}) that a  decrease in $\\alpha$ by 3\\% not only reduces $H_2$ by around 10\\% but also shifts the location of the first acoustic peak by almost 5\\% towards larger angular scales. This coincides with the results of Ref.~\\cite{Peebles}, for a specific model of delayed recombination based in early Ly $\\alpha$ sources of ionizing radiation.  The data from BOOMERANG and MAXIMA does not have the accuracy and angular resolution to test for the presence of a third acoustic peak. Consequently, we should not use $H_3$ as a relevant parameter. Eqs.~(\\ref{l1}-\\ref{H2}) alone are insufficient to analyse degeneracies as a function of just one parameter. It is however quite apparent from the two-dimensional contours of the previous section that the weakest degeneracy is that on the spectral index $n$, which is quite close to scale invariance. If we assume $\\Delta n=0$ we derive, imposing $\\Delta l_1=\\Delta H_1 = \\Delta H_2 = 0$,  \\begin{equation} \\frac{\\Delta\\omega_b}{\\omega_b}=3.6~\\frac{\\Delta\\alpha}{\\alpha} \\ , \\ \\  \\frac{\\Delta\\omega_m}{\\omega_m}=3.5~\\frac{\\Delta\\alpha}{\\alpha} \\ , \\ \\  \\frac{\\Delta H_0}{H_0}=6.4~\\frac{\\Delta\\alpha}{\\alpha} \\label{deg} \\end{equation} which is in very good agreement with the shape of the two-dimensional likelihood contours of the previous section.  Variations in  $\\omega_b$, $\\alpha$ and $n$ modify $H_2$. Neither the variation of $\\omega_b$ nor of $n$ have a strong impact upon the location of the first acoustic peak. Variation of $\\alpha$ does, as we already discussed. Compensation of the peak shift as $\\alpha$ decreases  requires a large decrease either of $H_0$ or of $\\omega_m$. Something similar happens with $H_1$, which is also very sensitive to changes in the spectral index $n$. Our results reveal a  strong degeneracy in the plane ($\\alpha, H_0$), which appears to be quite robust, in the sense that it persists with similar strength even under variations of $n$. The degeneracy in the plane ($\\alpha,\\omega_m$) is instead quite sensitive to the precise degeneracy in the  $(\\alpha,n)$ plane. Information on the structure of the angular power spectrum of CMB anisotropies around the location of the third acoustic peak is necessary to more accurately pinpoint this issue and eventually break some of the present degeneracies.  A decrease in $\\alpha$, as well as any delayed recombination mechanism in which there is an increase in the width of the visibility function, makes diffusion damping more effective, which increasingly reduces the heigths of the peaks at smaller angular scales. This is already noticeable around the third acoustic peak. A decrease in $\\alpha$ that reduces $H_2$ as efficiently as an increase in $\\omega_b$ has a stronger effect in reducing $H_3$, as can be seen from eqs.~({\\ref{H2},\\ref{H3}). Notice, for instance, that the best fit model with delayed recombination of Fig.~{\\ref{comp} has a third accoustic peak suppressed compared to the best fit with standard recombination, which has a larger baryon density. Measurements of the spectrum around and beyond the third accoustic peak can potentially break the degeneracy between the effects of an increase in $\\omega_b$ and delayed recombination.  Our results can also be seen as a bound on departures from standard recombination, which can happen at redshifts as low as $z_*=780$ at 95\\% confidence level, or as a bound on the time variation of the fine structure constant between decoupling and the present time. The bound we obtain, $0.02>\\Delta\\alpha/\\alpha>-0.14$, is an order of magnitude weaker than expected in \\cite{Turner} from future satellite measurements. The weakness of this bound is a consequence of the large present degeneracies. More accurate measurements and higher angular resolutions will certainly improve this bound. Whether a full order of magnitude improvement is or is not achieved will depend on how much this degeneracy is actually broken.  Our result that there is a preference for a smaller value of $\\alpha$ at decoupling compared to its present value is in agreement with similar conclusions of recent work \\cite{Battye00,Avelino00} that also analysed BOOMERANG and MAXIMA data allowing for a variation in the fine structure constant. In the present work we used the time varation of the fine structure constant as an example of how to model non-standard recombination histories. A preprint that develops a two-parameter model for non-standard recombination histories \\cite{Hannestad00} appeared after submission of our paper, with compatible results."
    },
    "0010/astro-ph0010309_arXiv.txt": {
        "abstract": "We  analyse the BOOMERANG and MAXIMA results in the context of models with $\\Omega_{\\rm Total} = 1$ and $n_s = 1$.  We attempt to constrain three other parameters---$h$, $\\Omega_B$, and $\\Omega_m$---from these observations. We show that:  (a) the value of $\\Omega_B h^2$ is too high to be compatible with primordial nucleosynthesis observations at 95\\% confidence level (b) universe with age greater than $ 12 \\, \\rm Gyr$ is ruled out at 95\\% confidence level and (c) the value of $\\Omega_m h$ is too high to be compatible with the shape of the power spectrum of gravitational clustering. In effect, our analysis shows that models with $\\Omega_{\\rm Total} = 1$ and $n_s =  1$ are ruled out by  BOOMERANG and MAXIMA  observations. ",
        "introduction": "Most generic models of inflationary scenario give two unique predictions: $\\Omega_{\\rm Total} = 1$ and $n_s \\simeq  1$ (see e.g. Steinhardt 1995).  The first of these predications is confirmed by BOOMERANG and MAXIMA. The analysis of COBE-DMR data is consistent with $n_s = 1$ (Bennett {\\it et al \\/} 1996, Bunn \\& White 1996).  Therefore it is reasonable to believe that the current data is in good agreement with these predictions. We fix the value of these two parameters based on these considerations and use $\\Omega_{\\rm Total} = 1$ and $n_s \\simeq 1$ throughout.    (It should be pointed out that $\\Omega_{\\rm Total} = 1$ is a stricter requirement of inflation than $n_s = 1$; one gets $n_s = 1$ only for exponential inflation; see e.g. Steinhardt 1995.) Also note that we are not concerned with the origin of the values $\\Omega_{\\rm Total} = 1$ and $n_s = 1$. For example, these could arise merely from the requirement of scale invariance for the background universe (giving $\\Omega_{\\rm Total}=1$) and the perturbations (giving $n_s=1$) without invoking inflation --- as was originally done by Harrison and Zeldovich, years before inflation was invented (Harrison 1970, Zeldovich 1972). But, of course, inflationary models made these parameter values fairly well accepted. Other parameters like $\\Omega_\\Lambda$, $h$ and $\\Omega_B$ cannot be fixed by theoretical considerations alone. In our analysis we assume a non-zero $\\Omega_\\Lambda$ because recent high-z SN1a observations suggest a non-zero cosmological constant (Perlmutter {\\it et al.} 1999, Riess {\\it et al.} 1998). We do not consider CMBR anisotropies from tensor perturbations as most models of inflation give negligible contribution from tensor perturbations for $n_s =1$ (Steinhardt 1995). Re-ionization of universe can also alter primary CMBR anisotropies significantly. Present observations suggest that the universe is ionized up to $z \\simeq 5$. For the CMBR anisotropies to be significantly altered by re-ionization, the minimum re-ionization redshift should lie between $10$ and $20$ (see e.g. Bond 1996). Therefore, keeping the constraints from present observations in mind, we neglect the effect of re-ionization on the measured CMBR anisotropies.  We  fix the value of  $\\Omega_m  + \\Omega_\\Lambda = \\Omega_{\\rm Total} = 1$, and compute the confidence levels on the best-fit values of $\\Omega_B$ and $h$, and $\\Omega_m$ from MAXIMA and BOOMERANG observations. The range of parameters in which the minimum of $\\chi^2$ is searched is: $0.01 \\le \\Omega_B  \\le 0.15$, $0.4 \\le h \\le 1.1$ and $0.1 \\le \\Omega_m \\le 0.95$.  The $\\chi^2$ for the model comparison with observations is given by: \\begin{equation} \\chi^2 = \\sum_{i = 1}^N \\left ({{\\cal C}_\\ell^{\\rm obs} - {\\cal C}_\\ell^{\\rm th} \\over \\Delta {\\cal C}_\\ell^{\\rm obs}} \\right) \\label{eq:a1} \\end{equation} Here $N = 22$ (10 points from MAXIMA and 12 points from BOOMERANG), ${\\cal C}_\\ell^{\\rm obs}$ are the measured band-powers and ${\\cal C}_\\ell^{\\rm th}$ are the theoretical band-powers, $\\Delta {\\cal C}_\\ell^{\\rm obs}$ are the errors on measured band-powers. We do not take into account the calibration uncertainties in our analysis. The theoretical band-powers are calculated using the CMBR Boltzmann code CMBFAST (Seljak \\& Zaldarriaga 1996), which  gives COBE-normalized (normalized according to the fitting formula of Bunn \\& White (1996)) angular power spectra. Eq.~(\\ref{eq:a1}) assumes that different band-powers are uncorrelated. This assumption is valid only if the anisotropies are measured over the entire sky (see e.g. Peebles 1993, Padmanabhan 1993). We assume lack of correlation in this work also because the covariance matrix of band-powers has not been made public yet.    Eq.~(\\ref{eq:a1}) implicitly assumes that the likelihood function is Gaussian in band-powers near its maximum. While this assumption is true in principle, in practice there can be significant deviation from Gaussianity near the maximum. Bond, Jaffe, and Knox (2000) advocate using another variable instead of band-powers for doing the maximum likelihood analysis. We do not use it here. However,  while quoting errors we do not use the Fisher matrix approach which can give meaningful results only for the Gaussian case. Instead  we directly give the confidence levels on $\\Delta \\chi^2$. ",
        "conclusions": ""
    },
    "0010/astro-ph0010623_arXiv.txt": {
        "abstract": "Asteroseismology involves using the resonant frequencies of a star to infer details about its internal structure and evolutionary state.  Large efforts have been made and continue to be made to measure oscillation frequencies with both ground- and space-based telescopes, with typical precisions of one part in $10^3$--$10^4$.  However, oscillation frequencies are most useful when accompanied by accurate measurements of the more traditional stellar parameters such as luminosity and effective temperature.  The Hipparcos catalogue provides luminosities with precisions of a few percent or better for many oscillating stars.  I briefly discuss the importance of Hipparcos measurements for interpreting asteroseismic data on three types of oscillating stars: $\\delta$~Scuti variables, rapidly oscillating Ap stars and solar-like stars. ",
        "introduction": "Because my talk was the last of the day, I chose to begin it with some controversy.  I suggested that we heed the urgings of Trimble (1995) to adopt the spelling `astroseismology.'  My basis was a search using the ADS Abstract Service\\footnote{\\tt http://adsabs.harvard.edu/abstract\\_service.html} that showed this spelling in the titles of the first three papers on the subject, by Christensen-Dalsgaard (1984), Gough (1985) and Kurtz (1986).  Since all three authors are well-known experts on both stellar oscillations and the English language, I was inclined to accept Trimble's suggestion.  I have since checked the original papers and found the `e' to be present in all three, so the ADS made a mistake (which they have since corrected).  I am happy to retract my suggestion, and refer the interested reader to Gough (1996) for an eloquent and scholarly defence of the `e'.  Having tracked down the seminal paper on this subject, namely that by Christensen-Dalsgaard (1984), I can do no better than quote his definition of asteroseismology as ``the science of using stellar oscillations for the study of the properties of stars, including their internal structure and dynamics.''  Hipparcos luminosities are important because asteroseismology is most powerful in constraining stellar models when the oscillation frequencies are accompanied by more traditional observables.  I will concentrate on three classes of stars for which Hipparcos parallaxes are particularly relevant. ",
        "conclusions": ""
    },
    "0010/astro-ph0010079_arXiv.txt": {
        "abstract": "The combined effect of a dissipative fluid and quintessence energy can simultaneously drive an accelerated expansion phase at the present time and solve the coincidence problem of our current Universe. A solution compatible with the observed cosmic acceleration is succinctly presented. ",
        "introduction": "In recent years observation of Type Ia supernovae has lend strong support to an accelerated expansion of the Universe at present time [\\cite{acc}]. This unexpected feature could be explained by resorting to a small cosmological constant $\\Lambda$, which on the one hand would provide enough negative pressure to account for this acceleration, and on the other hand would contribute an energy density of the same order of magnitude than the energy density of the matter content (baryonic plus dark) of today's Universe -say $\\rho_{m} \\simeq 0.3$ and $\\rho_{\\Lambda} \\simeq 0.7$.  This seemly straightforward ``solution\" poses, however, a serious problem. Since $\\rho_{m}$ redshifts as $a^{-3}$ while $\\Lambda$ is a constant, why they turn out to be comparable today? i.e., why we happen to live in a very special (and rather short) phase of cosmic expansion? This is the so--called {\\it coincidence problem} [\\cite{stein}].  To avoid it a new form of dark energy (quintessence energy, or Q--matter) has been introduced. This energy corresponds to a scalar field $\\phi$ that slowly rolls down its potential with the key property of having a negative pressure. In some respects it mimics the scalar field that suppossedly drived inflation at the very early Universe -see e.g. [\\cite{quintessence}]. All these models -formulated for spatially flat Friedmann--Lema\\^{\\i}tre--Robertson--Walker (FLRW) universes- overlook the fact that the matter component of the Universe (baryonic and non--baryonic) might not be very well approximated by a perfect fluid since, in general it should behave as a dissipative fluid and therefore it must have a non--equibrium pressure that might be non--negligible.  In this short report we shall show that the combination of quintessence energy (an evolving scalar field with negative pressure) and a perfect matter fluid cannot simultaneously drive the current accelerated expansion and solve the coincidence problem. However, when the matter fluid is no longer assumed perfect this difficulty disappears altogether for open and flat FLRW universes [\\cite{chimento}].  At this point it is fair to say, however, that attempts to solve the coincidence problem have been also made by using scalar--tensor theories of gravity rather than general relativity [\\cite{orfeu}]. We will not deal with these here. ",
        "conclusions": ""
    },
    "0010/astro-ph0010553_arXiv.txt": {
        "abstract": "We discuss results from sensitive, wide-field imaging of the millimeter extragalactic background using the Max-Planck Millimeter Bolometer array (MAMBO) at the IRAM 30~m telescope. ",
        "introduction": " ",
        "conclusions": ""
    },
    "0010/astro-ph0010286_arXiv.txt": {
        "abstract": "We present a new collapse condition to describe the formation of dark halos via nonspherical gravitational clustering.  This new nonspherical collapse condition is obtained by the logical generalization of the spherical model to the nonspherical one. By solving  a diffusion-like random matrix equation with the help of the Monte Carlo method,  we show that this nonspherical collapse condition yields the mass function derived by Sheth \\& Tormen (1999) which has been shown to be in excellent agreement with the recent N-body results of high resolution. We expect that this nonspherical collapse condition might provide us a deeper insight into the structure formation, and suggest that it should be widely applied to various cosmological issues such as the galaxy merging history, the galaxy bias, and so forth. ",
        "introduction": "The mass function $n(M,z)$ in cosmology is defined to give the comoving number density of dark halos with mass $M$ at redshift $z$. It provides a useful analytical tool to understand the formation and evolution of the large-scale structure in the universe. The excursion set approach to the mass function provides the most direct and solid way to count the number density of dark halos. Bond et al. (1991) applied for the first time the excursion set theory to the Gaussian random density field, and recovered the popular Press-Schechter mass function (Press \\& Schechter 1974, hereafter PS) with a correct normalization factor of $2$.  Sheth, Mo, \\& Tormen (1999, hereafter SMT) suggested an extension of the excursion set approach to a nonspherical dynamical model. Although the PS mass function works fairly well at the high-mass section, recent hight-resolution N-body simulations have yielded less intermediate-mass and more low-mass halos than the PS prediction \\citep{lac-col94,tor98,gov-etal99,jen-etal00}. It has been suspected that this discrepancy of the numerical mass functions with the PS prediction must be due to the departure of the true dynamics from the idealistic spherical one \\cite{gov-etal99}. The work of SMT was in fact so motivated, attempting to justify the empirical  mass function derived by Sheth \\& Tormen (1999, hereafter ST). Indeed, current numerical results from high-resolution N-body simulations agree with the ST formula much better than the standard PS mass function \\cite{jen-etal00}.  Yet it is hard to claim that the ST mass function is anything beyond a  phenomenological fitting formula. Although SMT claimed that the excursion set approach associated with their nonspherical collapse condition does produce the ST mass function to excellent approximation, a more careful analysis of the nonspherical collapse condition from a different perspective ought to be encouraged. In particular,  the perspective of symmetry and smoothness of the collapse condition  in the $\\Lambda$-space (see $\\S 2$) should be seriously considered, and can serve as a guiding principle in constraining the correct collapse condition.  In this paper, we investigate the idea of SMT in a more sophisticated manner to find a physically motivated ellipsoidal collapse condition which yields almost the same ST mass function. The new collapse condition demonstrates in a clear-cut manner how the initial nonspherical properties of proto-halos affect the gravitational process, yielding the halo abundance substantially deviating from the PS predictions based on the spherical model. ",
        "conclusions": "Motivated by the inspiring practical success of the ST mass function in recent N-body tests \\cite{jen-etal00},  we have attempted here to provide a more sophisticated and robust way to determine the nonspherical collapse condition which produces the ST mass function to good approximation.  In fact, various nonspherical approaches to the mass function were already attempted by several authors in the past decades \\citep{mon95,aud-etal97,lee-sh98}. Strictly speaking, however, their approaches were not appropriate in the sense that they all used the original PS formalism which always yields ill-normalized mass functions.  Although the PS mass function can be properly normalized by multiplying a constant normalization factor of $2$ \\cite{bon-etal91},  the mass function from a nonspherical model can no longer be corrected just by a constant normalization factor.  The normalization factor is scale-dependent in any nonspherical dynamical model,  due to the complicated pattern of the scale-dependent occurrence of the cloud-in-clouds. For a detailed description of the cloud-in-cloud problem, see Bond et al. (1991), Jedamzik (1995), and Lee \\& Shandarin (1998). Due to this scale-dependent manner of the cloud-in-cloud occurrence in the nonspherical models, the shape of the mass function could be very different from the one obtained without considering the cloud-in-cloud occurrence correctly. For example, we have tested by our diffusion algorithm the condition that the bound objects form at the local maxima of the smallest eigenfield of the deformation tensor.  We found that the resulting mass function deviates considerably from the one originally given by Lee \\& Shandarin (1998) who had used a constant normalization factor of $12.5$.  In order to correctly solve the normalization problem, one should not rely on the original PS formalism. The only viable alternative for the evaluation of the mass function without extra efforts of concerning about the normalization is the excursion set approach, or equivalently the diffusion approach. SMT employed the excursion set approach to justify the ST mass function in terms of the nonspherical collapse condition. They related the nonspherical collapse condition with a moving barrier and approximated its shape with the help of the peak-patch prescriptions for the ellipsoidal dynamics \\cite{bon-mye96}.  Their moving barrier (eq. [3] in SMT) is similar to our curved boundary in concept,  but differs significantly in practice since the SMT boundary is not a rotationally invariant smooth surface in the $\\Lambda$-space and has kinks.  Given equation (3) in SMT, one can see that the kinks arise at $p = 0$ ($p$: the prolateness of the given ellipsoidal region, see Bardeen et al. 1986). As mentioned in SMT, however, $p = 0$ on average in a Gaussian random field. Thus, the kinks of the SMT boundary occur in so high probability regions that one may not ignore the presence of those kinks.  Here we did not attempt to provide a better ellipsoidal dynamical model to describe the gravitational collapse process. Rather we retained the general framework of the peak-patch theory and try to improve the SMT boundary collapse condition into a more physically meaningful one by the logical generalization of the spherical collapse model. Two superior features of our collapse condition can be summarized as follows: First, it has a sound physical meaning, in that equation (2) is expressed in terms of only one single variable $r$ which is in fact directly proportional to the halo angular momentum square \\citep{hea-pea88, cat-the96}.  Any physically meaningful nonspherical collapse condition should be expressed in terms of such quantities as obviously represent the nonshperical nature of halo formation. As mentioned in $\\S 3$, the rotational motion of dark halos is a unique consequence of the nonspherical collapse. In this respect, our nonspherical collapse condition shows explicitly and quantitatively how the nonspherical gravitational collapse leads to the acquisition of the angular momentum of dark halos. Second it has the desirable analytical property, {\\it rotational invariance and smoothness}. This feature of our collapse condition, in common with the PS collapse condition, will make it easy to extend various important cosmological issues, such as the halo merging, the light-to-mass bias, and so on, from the spherical dynamics to a nonspherical one."
    },
    "0010/hep-ph0010066_arXiv.txt": {
        "abstract": "The flux of cosmic rays beyond the GZK cutoff ($\\sim 10^{20}$ eV) may be explained through their production by ultra high energy cosmic neutrinos, annihilating on the relic neutrino background, in the vicinity of our galaxy.  This process is mediated through the production of a $Z$ boson at resonance, and is generally known as the $Z$-Burst mechanism.  We show that a similar mechanism can also contribute to the super-GZK spectrum at even higher, ultra-GZK energies, where the particles produced at resonance are the Kaluza-Klein gravitons of weak scale mass and coupling from the Randall-Sundrum (RS) hierarchy model of localized gravity model. We call this mechanism Gravi-Burst.  We discuss the parameter space of relevance to Gravi-Bursts, and comment on the possibility of its contribution to the present and future super-GZK cosmic ray data and place bounds on the RS model parameters. Under certain assumptions about the energy spectrum of the primary neutrinos we find that cosmic ray data could be potentially as powerful as the LHC in probing the RS model. ",
        "introduction": "About 25 years ago, Greisen, Zatsepin, and Kuzmin (GZK) noted that the observed spectrum of proton, photon, and nucleus cosmic rays must virtually end at energies above $\\sim 10^{20}$ eV, the GZK cutoff \\cite{GZK}.  Their key observation was that Ultra High Energy Cosmic Rays (UHECR's) deplete their energy through various interactions with the 2.7$^{\\, ^\\circ}$ K Cosmic Microwave Background Radiation (CMBR), over distances of order $10-100$ Mpc.  Above $10^{19}$ eV, nuclei are photo-dissociated by interactions with the CMBR, and a $10^{20}$ eV proton loses most of its energy over a distance of $\\sim 50$ Mpc.  The analogous distance for a photon of the same energy is $\\sim 10$ Mpc, due to $e^+ e^-$ pair production on the radio background \\cite{radio}.  However, over the past three decades, different experiments have observed a total of about 20 events at or above this $10^{20}$ eV bound\\cite{20sgzk}.  Since there seem to be no feasible candidates for the sources of these cosmic rays, such as Active Galactic Nuclei, within a GZK distance $\\sim 50$ Mpc of the earth, the observation of these events poses a dilemma.  A number of proposals have been made to resolve this puzzle \\cite{Olinto}.  One such proposal for the origin of the super-GZK events, due to Weiler, is based on the observation that UHECR neutrinos can travel over cosmological distances, with negligible energy loss{\\cite{Zburst,Weiler}}.  Therefore, if these neutrinos are present in the universe they could in principle produce $Z$ bosons on resonance through annihilation on the relic neutrino background, within a GZK distance of the earth.  The highly boosted subsequent decay products of the $Z$ will then be observed as primaries at super-GZK energies, since they do not have to travel cosmological distances to reach us.  This mechanism for producing super-GZK cosmic rays is referred to as $Z$-Burst.  The $Z$-burst mechanism has the advantage that it does not assume physics beyond the Standard Model (SM) and is, therefore, minimalistic.  However, any extension of the SM that provides a particle $X$ which couples to $\\nu {\\bar \\nu}$ and decays into the usual primaries can in principle contribute to the super-GZK spectrum beyond the range presently observed. Assuming a mass $m_\\nu \\sim 10^{-2}-10^{-1}$ eV for neutrinos as suggested by atmospheric oscillation data, the particle $X$ must have a mass of order the weak scale ($\\sim 1$ TeV) to be relevant to the spectrum near the GZK cutoff.  In this paper, we will show that the massive Kaluza-Klein (KK) tower of gravitons in the Randall-Sundrum (RS) localized gravity model{\\cite {RS}} are viable candidates for particle $X$.  The RS model is based on a truncated five-dimensional Anti-deSitter($AdS_5$) spacetime, with two 4-$d$ Minkowski boundaries.  Our visible 4-$d$ universe and all fields associated with the SM are assumed to be confined on one of these boundaries, referred to as the TeV brane, with the other `Planck' brane boundary separated from us by a fixed distance $r_c \\sim 10 \\, \\mpl^{-1}$, the compactification scale along the $5^{th}$ dimension; $\\mpl$ is the reduced Planck mass.   The RS geometry is such that the induced metric on the visible TeV brane generates the weak scale from a 5-$d$ scale $M_5 \\sim \\mpl$, without fine-tuning, through an exponentiation.  The interested reader is referred to Refs.  \\cite{RS,dhrprl} for the details of the RS model and its numerous phenomenological implications.  However, here we mention that a distinct feature of this model is that it predicts the existence of a tower of spin-2 KK gravitons, $G^{(n)}$ ($n = 1, 2, 3, \\ldots$), starting at the weak scale, and with weak scale mass splittings and couplings.  Phenomenological studies \\cite{dhrprl,dhrprd} suggest that the lowest lying KK graviton $G^{(1)}$ can be as light as $\\sim 400$ GeV.  The $G^{(n)}$ have couplings to all particles, due to their gravitational origin and can be produced by $\\nu {\\bar \\nu}$ annihilation, eventually decaying into $q {\\bar q}, g g, \\gamma \\gamma, \\ldots$. Thus, the $G^{(n)}$ can in principle contribute to the super-GZK spectrum in a way that is similar to the $Z$-burst contribution.  We call this graviton mediated process Gravi-Burst.  Since the $Z$ and $G^{(n)}$ have different couplings and branching fractions to the observed primary particles, we expect that experiments may be able to distinguish between $Z$-burst and gravi-burst initiated primaries. Also, depending on the behavior of the flux of neutrinos at super-GZK energies, more than one member of the KK graviton tower could contribute to gravi-burst.  In this case, the RS model predicts a characteristic multi-peaked behavior for future data at super-GZK energies and beyond. However, collider experiments may be a better place to directly search for the graviton tower, with cosmic ray data providing complementary information as we will discuss in detail later.  In the next section, we present the necessary formulae for estimating the super-GZK flux in the $Z$-burst model.  We adapt this approach to gravi-burst and give the corresponding rate estimates in this scenario.  Section 3 contains our results for a range of RS model parameters and a comparison with the $Z$-burst predictions.  We will show that if the neutrino spectrum falls sufficiently slowly with energy we can use GZK data to greatly restrict the parameter space of the RS model. Our conclusions are given in section 4. ",
        "conclusions": "In this paper we have examined the possible contribution to the spectrum of cosmic rays beyond the GZK cutoff due to new physics arising in the Randall-Sundrum model of localized gravity. Our analysis is based on the assumptions ($i$) that the events observed immediately above the GZK bound can be explained by the $Z$-Burst mechanism and ($ii$) the neutrino spectrum needed for $Z$-Bursts extends s few orders of magnitude further in neutrino energy with a reasonably slow fall-off. If these conditions hold then the existence of a series of $s$-channel Kaluza-Klein graviton resonances in the $\\nu \\bar \\nu \\to$ hadrons channel, which is predicted in the RS model, can lead to events with even higher energies, ultra-GZK, due to Gravi-Bursts. The rate for these bursts are generally at or near the present level of observability for a wide range of RS model parameters. The fact that such events are not as yet observed can be used to constrain the parameter space of the RS model once a specific form of the neutrino energy spectrum is assumed. These bounds can be more restrictive than those that can be obtained from the lack of graviton resonance production at the Tevatron during Run II ($30~fb^{-1}$) if the fall-off with energy of the UHECR neutrino flux is linear or less steep. If ultra-GZK events are not observed by future experiments such as the Auger Array, then the resulting bounds on the RS model can be complementary to those obtainable at the LHC. If such events are observed at future experiments, the RS resonance structure may be observable given both sufficient statistics and good hadronic energy resolution. In addition to hadronic modes, the RS graviton KK tower states can directly decay to photon pairs which will have more than an order of magnitude greater energies that those that can arise due to ordinary fragmentation into $\\pi^0$'s which subsequently decay into two photons. If photon and hadron induced showers can be distinguished at such energies this will provide a unique signature for the RS model as the origin of the ultra-GZK events.    \\noindent{\\Large\\bf Acknowledgements}  The authors would like to thank the Aspen Center for Physics, where this work was begun, for its hospitality. T.G.R. would like to thank T. Weiler and J.L.H. would like to thank E. Kolb for discussions related to this paper.    \\newpage  \\def\\MPL #1 #2 #3 {Mod. Phys. Lett. {\\bf#1},\\ #2 (#3)} \\def\\NPB #1 #2 #3 {Nucl. Phys. {\\bf#1},\\ #2 (#3)} \\def\\PLB #1 #2 #3 {Phys. Lett. {\\bf#1},\\ #2 (#3)} \\def\\PR #1 #2 #3 {Phys. Rep. {\\bf#1},\\ #2 (#3)} \\def\\PRD #1 #2 #3 {Phys. Rev. {\\bf#1},\\ #2 (#3)} \\def\\PRL #1 #2 #3 {Phys. Rev. Lett. {\\bf#1},\\ #2 (#3)} \\def\\RMP #1 #2 #3 {Rev. Mod. Phys. {\\bf#1},\\ #2 (#3)} \\def\\NIM #1 #2 #3 {Nuc. Inst. Meth. {\\bf#1},\\ #2 (#3)} \\def\\ZPC #1 #2 #3 {Z. Phys. {\\bf#1},\\ #2 (#3)} \\def\\EJPC #1 #2 #3 {E. Phys. J. {\\bf#1},\\ #2 (#3)} \\def\\IJMP #1 #2 #3 {Int. J. Mod. Phys. {\\bf#1},\\ #2 (#3)}"
    },
    "0010/astro-ph0010003_arXiv.txt": {
        "abstract": "The dynamical friction experienced by a body moving in a gaseous medium is different from the friction in the case of a collisionless stellar system. Here we consider the orbital evolution of a gravitational perturber inside a gaseous sphere using three-dimensional simulations, ignoring however self-gravity. The results are analysed in terms of a `local' formula with the associated Coulomb logarithm taken as a free parameter. For forced circular orbits, the asymptotic value of the component of the drag force in the direction of the velocity is a slowly varying function of the Mach number in the range $1.0$--$1.6$. The dynamical friction timescale for free decay orbits is typically only half as long as in the case of a collisionless background, which is in agreement with E.C.~Ostriker's recent analytic result.  The orbital decay rate is rather insensitive to the past history of the perturber.  It is shown that, similar to the case of stellar systems, orbits are not subject to any significant circularization. However, the dynamical friction timescales are found to increase with increasing orbital eccentricity for the Plummer model, whilst no strong dependence on the initial eccentricity is found for the isothermal sphere. ",
        "introduction": "A gravitating body moving through a background medium suffers from a drag force due to the interaction with its own induced wake. Generally speaking, the medium may be composed of collisionless matter, and/or gas. For collisionless backgrounds, the classical Chandrasekhar formula has proved useful in determining the orbital decay of galactic satellites and globular clusters around spherical systems such as elliptical galaxies (e.g. Lin \\& Tremaine 1983; Cora, Muzzio \\& Vergne 1997, and references therein), even though it was inferred for uniform media. For a uniform and infinite gaseous medium, the gravitational drag on a body moving at constant velocity on a straight-line orbit, has been estimated both by linear theory (e.g. Ruderman \\& Spiegel 1971; Rephaeli \\& Salpeter 1980; Ostriker 1999) and from numerical experiments in different settings (e.g. Shima et al.\\ 1985; Shima et al.\\ 1986; Shankar, Kley \\& Burkert 1993; Kley, Shankar \\& Burkert 1995; Ruffert 1996). All these authors consider a uniform medium because they are mainly interested in the accretion flow past the body. Here we will investigate the gaseous drag and the sinking decay of a light body orbiting on a gaseous sphere which is initially in hydrostatic equilibrium with a given gravitational potential. In fact, dynamical friction for gaseous and spherical backgrounds has its relevance to many astrophysical studies, e.g. galaxies in galaxy clusters, the relaxation of young stellar clusters, or the dynamics of massive black holes within their host galaxies. It will be useful to check in which conditions the drag formula obtained in linear theory for uniform media works for spherical backgrounds, in analogy to stellar systems.  The question of dynamical friction for gaseous and spherical backgrounds has received new interest in connection with recent studies suggesting that Galactic dark matter may be in the form of cold molecular clouds either distributed in a disc \\cite{Pfe94} or in a quasi-spherical halo (e.g. De Paolis et al.\\ 1995; Gerhard \\& Silk 1996; Walker \\& Wardle 1998; Walker 1999; De Paolis et al.\\ 1999; Sciama 2000).  For certain parameters of these clouds, the assumption of collisionless matter is no longer valid. Other authors have proposed, based on observational astrophysical grounds, that dark matter may be self-interacting with a large scattering cross-section \\cite{Spe00}. It is therefore important to find out the dynamical implications of having spherical halos of collisional matter and, in particular, its effect on the decay of satellite galaxy orbits.  Deeper physical insight into the question of how dynamical friction is affected by collisions may be provided by considering continuous gaseous media. Recently, Ostriker \\shortcite{Ost99} pointed out that for supersonically moving bodies the drag in a uniform gaseous medium is more efficient than in the case of collisionless media that are described by the standard Chandrasekhar formula; and is non-vanishing for subsonic bodies. The same feature is observed in the case of a perfectly absorbing body \\cite{Ruf96}. As a consequence, satellite galaxies may experience more rapid decay in halos made up of molecular clouds, especially in earlier epochs before most of the gas has turned into stars.  In the standard model of cosmological structure formation, forming substructures continuously exchange orbital energy with both collisionless material and with their surrounding hot medium due to dynamical friction. An understanding of the processes that can produce velocity and spatial bias where the concentration and velocity dispersion is different for different galactic populations, e.g. mass segregation in clusters of galaxies, is essential for the interpretation of observational data as well as for semi-analytical models of the formation of galaxies and galaxy clusters. Clumps of cold gas moving through a hotter medium may suffer a hydrodynamical drag that is stronger than the gravitational drag (e.g.\\ Tittley et al.\\ 1999). In this work, however, we focus on a purely gravitational perturber.  The linear response of a spherical system to an orbiting gravitational perturber was carried out by Balbus \\& Soker \\shortcite{Bal90} motivated by the problem of excitation of internal gravity waves by galaxies in the core of clusters.  In this paper we concentrate on the dynamical friction of a rigid perturber orbiting in a gaseous medium that shows a spherical density distribution with a concentration towards the centre. Initially the gas is at rest and in hydrostatic equilibrium with an external spherical gravitational potential. The gravitational perturber is assumed to be very large (typically the softening gravitational radius is taken a few times the accretion radius) and unable to accrete matter. We do not model the existence of a physical surface on the body, i.e. no inner boundary condition on the body has been imposed. A discussion on the importance of accretion is given in Section 2. Our aim is to compute the evolution and sinking time of the perturber towards the center.  The assumptions of infinity and homogeneity of the medium, that was required in Ostriker's analysis, are relaxed, and, in addition, the contribution of non-local effects on the drag force are taken into account. Also, an additional difference is that we deal with extended perturbers whose radius is a significant fraction of the size of the background system.  Since the perturbers in most cases of interest do not present circular orbits \\cite{Bos99}, and also motivated by the fact that cosmological simulations show that the majority of satellite orbits have quite large eccentricities, we consider the dependence of the sinking times on the eccentricity of the orbits.  There are some similarities with the dynamical evolution of binary star cores embedded in a common envelope (e.g. Taam, Bodenheimer \\& R\\'{o}zyczka 1994; and references therein). In those systems, however, the gravitational torques of the two cores are able to spin up the envelope to a high rotation (see also Sandquist et al.\\ 1998). The fact that the masses of the stars and the envelope are comparable is crucial for the subsequent evolution of the system. Here we explore a rather different case in which there is a single rotating body and its mass is significantly less than that in the gaseous background mass. Recall that neither wind is injected in the gaseous component nor radiation is considered in the present work.  The paper is organized as follows. In Section 2, we highlight the differences between dynamical friction in collisionless and collisional media; the importance of having mass accretion by the perturber is also addressed. The description of the model is given in Section 3. In Section 4 the orbital decay rate is computed numerically for a body orbiting around a gaseous sphere, and is compared with a `local' estimate of the drag force. Implications and conclusions are given in Section 5. ",
        "conclusions": "In this paper we have presented simulations of the decay of a rigid perturber in a gaseous sphere. As a first approximation, both the mass stripping and the barycentric motion of the primary were neglected. In addition, the softening radii of the perturbers were taken a few times larger than the accretion radius. Simulations of the evolution of very compact bodies ($R_{\\rm soft}\\ll R_{\\rm ac}$), such as black holes in a spherical background are numerically very expensive.  The Linear Approximation Prescription is able to explain successfully the evolution of the perturber provided the (i) decay is not too fast, (ii) the motion is supersonic and (iii) the body is not too close to the center to avoid that $R_{\\rm p}$ becomes comparable to $R_{\\rm soft}$. In our simulations, the angular momentum of the perturber decays almost linearly with time. Generally speaking, the associated Coulomb logarithm should depend on the distance of the perturber to the centre and on its Mach number. However, the evolution of a free perturber is relatively well described by just a linear dependence of the maximum impact parameter of the Coulomb logarithm on radius, except very close to the centre where the time averaged Mach number is less than unity.  In Chandrasekhar's formalism the Coulomb logarithm is often approximated by the logarithm of the ratio of the maximum to minimum impact parameters of the perturber. In a singular isothermal sphere of collisionless matter, the time for an approximately circular orbit to reach the center is \\begin{equation} \\tau_{\\rm s}=1.2 \\frac{r_{\\rm cir}^{2} V_{\\rm cir}}{G M_{\\rm p} \\ln \\frac{b_{\\rm max}}{b_{\\rm min}}}, \\end{equation} where $b_{\\rm max}$ is roughly the virial radius of the primary and $b_{\\rm min}$ is of the order of the softening radius of the perturber \\cite{Bin87}. We have confirmed numerically Ostriker's suggestion that the Coulomb logarithm should be replaced by \\begin{equation} \\ln \\frac{b_{\\rm max}}{b_{\\rm min}}\\rightarrow \\frac{1}{0.428} \\ln \\left(\\eta\\frac{R_{\\rm p}(t)}{R_{\\rm soft}}\\right) \\end{equation} for the gaseous case, with $\\eta\\simeq 0.35$, and independent of the eccentricity. For very light perturbers in near-circular orbits (objects with mass fractions smaller than one per cent), the time scale for dynamical friction is a factor $2$ smaller in the case of a gaseous sphere than in corresponding stellar systems (e.g.\\ globular clusters in the Galactic halo). This factor is somewhat smaller when $R_{\\rm p}$ becomes comparable to a few $R_{\\rm soft}$. The weak dependence of the decay times on eccentricity suggests that more eccentric orbits of merging satellites should not decay more rapidly in the halo during the epoch in which matter is mainly gas. Thus, they do not touch the disc at an earlier time than less eccentric orbits.  Our simulations give support to the idea that dynamical friction does not produce changes in the distribution of orbital eccentricities.  The analysis must be modified if one wishes to account for the existence of clouds in a non-uniform medium. So far, in a variety of systems it is assumed that the perturber has condensed from fragmentation of the gas; thus, the gaseous component must be thought of as being distributed in clumps or clouds even more massive than the `perturbers'.  In certain situations, compact perturbers may not be dragged but heated (e.g.\\ Gorti \\& Bhall 1996). For this reason one may be pessimistic towards a recent suggestion by Ostriker (1999) that, because of the contribution of the gaseous dynamical friction force, young stellar clusters should appear significantly more relaxed than usually expected."
    },
    "0010/astro-ph0010529_arXiv.txt": {
        "abstract": "We present details of a prototype fiber feed for use on the Anglo-Australian Telescope (AAT) that uses a dedicated fiber-fed medium/high resolution ($R\\simeq 10000$) visible-band spectrograph to give integral field spectroscopy (IFS) of an extended object.  A focal reducer couples light from the telescope to the close-packed lenslet array and fiber feed, allowing the spectrograph be used on other telescopes with the change of a single lens.  By considering the properties of the fibers in the design of the spectrograph, an efficient design can be realised, and we present the first scientific results of a prototype spectrograph using a fiber feed with 37 spatial elements, namely the detection of Lithium confirming a brown dwarf candidate and IFS of the supernova remnant SN1987A. ",
        "introduction": "The Anglo-Australian Telescope (AAT) has pioneered many astronomical instruments over the past 25 years, such as the 2dF galaxy survey instrument (\\cite{Tay97}) and including one of the earliest reported fiber feeds in the literature (\\cite{Gra82}).  These instruments all use optical fibers to reformat the two dimensional distribution of light from the focal plane of the telescope into a one dimensional fiber slit suitable for a spectrograph, allowing multiple object (in the case of 2dF) and spatially resolved spectroscopy of extended objects.  Although there are many other methods used to perform integral field spectroscopy (IFS) the extremely versatile properties of optical fibers naturally lend themselves to optical reformatting (see \\cite{Ken98} for a review of IFS methods).  SPIRAL (Segmented Pupil/Image Reformatting Array of Lenslets) Phase A is an IFS design with 37 spatially resolved elements (see Figure \\ref{fig1}) that has been built to prototype the new methods needed in building large integral field units (IFUs). Here we present a description and the first light results of SPIRAL Phase A.  \\placefigure{fig1}  The design of SPIRAL considers the optical fibers as part of the spectrograph instead of an auxillary unit to a slit-based spectrograph.  In section \\ref{ffs} we analyse a simple fiber-fed spectrograph. The optical design can be simplified by optimising for dedicated fiber use.  Section \\ref{lensletarray} then presents the design of the lens array, comparing the macrolens design with other methods of close packing fibers in the focal plane of the telescope.  By using fore-optics to match the plate scale of the telescope with the lenslet array, the spectrograph can then be used with other telescopes by changing a single lens.  The layout of SPIRAL Phase A is given in Section \\ref{layout} along with the optical design for the fiber-fed spectrograph. Section \\ref{fiberposition} briefly covers the performance and accuracy of the fiber positioning technique that is used in the construction of the lens array and focal plane unit for SPIRAL Phase A. The performance of the spectrograph is given in Section \\ref{char} and preliminary scientific results are then presented in Section \\ref{results}. ",
        "conclusions": "}  We have manufactured a lenslet array from individual lenslets and demonstrated that fibers can be aligned with these lenslets to $3\\micron$ RMS error, providing an alternative technique to the microlens array method. The use of fore-optics makes the fiber feed portable and the use of fibers leads to a simplified and more efficient spectrograph design.  Integral field observations of a supernova remnant allow maps of spatial variations in spectral line properties to be generated, and high spectral resolutions to be achieved. The success of this prototype led to the construction of SPIRAL Phase B and its comissioning in May 2000."
    },
    "0010/astro-ph0010235_arXiv.txt": {
        "abstract": "Angular correlations of ultra-high energy cosmic rays with cosmologically distant sources may provide clues to these mysterious events.  We compare cosmic ray tracks with energies above $ 10^{ 20}eV$ to a compilation of radio-loud compact QSO positions.  The statistical method emphasizes invariant quantities and a test of {\\it statistical independence} of track and source distributions. Statistical independence is ruled out by several independent statistics at confidence levels of less that $10^{-3}$ ($99.9\\% $.), indicating that track directions and QSO source positions are correlated at a highly significant level. ",
        "introduction": "The origin and propagation of the highest energy cosmic rays \\cite{linsley,agasa1999,fliesI,HP,winn} present a major challenge to current understanding of astrophysics.  The problem has developed over many decades.  In 1966 Greissen, Kuzmin and Zatsepin ($GZK$)\\cite{GZK} discovered a theoretical upper limit on cosmic ray energies of about $4 \\times 10^{ 19 }eV$.  The limit is due to laboratory-established pair and photo-nuclear production processes occuring on the cosmic background radiation.  Due to these processes, protons of such energies are unable to propagate distance greater than about 50 Mpc.  Other nuclei are even more severely restricted \\cite{puget}. It is commonly believed \\cite {Hillas,blandford,nagano} that the only possible astrophysical sources of particles with such high energies are the active galactic nuclei ($AGN$).  The vast majority of $AGN$ are cosmologically distant.  There are insufficient alternative sources within 50 Mpc of Earth to produce the observed events.  Hence the origin of numerous events above $4 \\times 10^{19 }eV$ is very puzzling.  The angular distribution of so-called ``$GZK$-violating'' events may contain important clues.  In searching for a possible source for the Fly's Eye event FE320 ($320 EeV$), Elbert and Sommers \\cite {ElbertandSommers} noticed that its arrival direction was very close to the remarkable quasar 3C147.  Farrar and Biermann ($FB$)\\cite{Farrar98} pointed out that 3C147 is a compact quasar with jets about one-tenth the size of a full-sized quasar with radio lobes. The spectrum of 3C147 is cut off at low radio frequencies, providing another characteristic of its compactness.   Here we re-examine possible correlation of compact radio-loud quasars and cosmic-ray track directions.  $FB$ claimed a correlation between track directions and compact quasars, defined by the following three criteria: \\begin{itemize} \\item[1] The quasar should be listed in the NASA/IPAC extragalactic database (NED). \\item[2] The object should be radio loud. In practice $FB$ required that the object appear in the  K\\\"{u}hr catalog \\cite{kuehr}. This catalog contains a total 1835 radio sources including all those whose flux density at 5 GHz is $\\ge$ 1 Jy. \\item[3] The object should have flat or falling spectrum at low radio frequencies. \\end{itemize}  We organize our study somewhat differently from previous work.  First, we pay extra attention to the baseline of the statistic, namely the definition of ``no signal'', employing several independent methods. This makes the study more reliable and more conservative than previous ones.  Second, we employ invariant quantitites, both for the purposes of conceptual correctness and also to correct substantial relative systematic errors.  Third and most importantly, we make no attempt to validate any proposed correlation.  In fact, correlations can exist in such myriad forms that any statistical procedure accepting one form over another {\\it a posteriori} can be suspect.  Of pivotal importance is that any proposed relation of $GZK$-violating cosmic rays and cosmological sources should test a crisp hypothesis.  The default for cosmologically-distant sources is a relation of {\\it statistical independence} of the arrival directions relative to the direction of the sources.  Besides the $GZK$ attenuation, scrambling of directions by intergalactic fields also tends to wash out any correlation of source-track directions. Independence is a well-defined hypothesis which can be tested without introducing model-dependence or extraneous postulates for these puzzling events. ",
        "conclusions": "For the data set of cosmic rays with energies above $10^ { 20 } eV$, the hypothesis of statistical independence of track directions and radio-loud $QSO$'s from the $K\\ddot{u}hr$ catlogue is not consistent with the data.  Several independent statistics yield results that are consistent with one another and not consistent with independence.  There exists highly significant correlations between the track and source directions.     \\begin{table} \\begin{tabular}{|c|c|c|c|c|} \\hline & & &Compact QSO & \\\\ Cosmic ray event & RA (degrees) & Dec. (degrees)& RA,Dec. (B1950) &$\\delta\\chi^2$ \\\\ & & & & \\\\ \\hline \\hline HP101 &  201    &\t  \t71    & 13 39 29.90, 69 38 30.4 &0.5   \\\\ HP115&  322     & \t  \t47    &   & \\\\ HP116 &  353      &\t \t19    & 23 37 58.3, 26 25 17 & 7.0  \\\\ HP126 & 179     & \t \t27    & 12 04 54.8, 28 11 34 & 0.8  \\\\ HP158  &  128     & \t \t67    & 08 59 23.02, 68 09 15.7 & 1.1  \\\\ HP159    &199     &    \t \t44    & 13 25 10.53, 43 42 00.3 & 0.5  \\\\ YU122 &    74.6     &  \t \t45.4  &    &\\\\ AG101 & 124.25  &\t\t16.8  & 07 59 55.6, 18 18 35& 4.1 \\\\ AG213 &  18.75   &\t\t21.1  & 01 09 23.7, 22 28 43.0 & 0.97 \\\\ AG106 &  281.25  &\t\t48.3  & 18 51 08.9, 48 52 06.0 & 0.6   \\\\ AG144 &  241.5   &\t\t23.0  & 15 38 30.18, 14 57 22.1 & 31.7  \\\\ AG105 & 298.5   &\t\t18.7  &    &  \\\\ AG150 & 294.5   &\t\t-5.8  & 20 08 25.90, $-$15 55 37.6 &51.3   \\\\ AG120 & 349.0  \t&\t\t12.3  & 23 28 08.6, 10 43 46  & 4.5  \\\\ AG104 & 345.75\t&\t\t33.9  & 23 27 45.8, 33 31 58 & 9.9  \\\\ VR135 & 306.7   &\t\t46.8  &   &  \\\\ SU197 &   187.1   &\t\t32    & 12 19 01.1, 28 30 57 & 1.8   \\\\ SU155 & 333.1   &\t\t-56   & 22 04 26.2, $-$54 01 14 & 0.6   \\\\ SU147 & 354.6   &\t\t-74   & 23 53 24.6, $-$68 37 42 & 3.8   \\\\ SU132 & 117.4   &\t\t-2    & 07 43 20.8, $-$00 37 00 & 0.3   \\\\ SU126 &  231.3   &\t\t-30   & 15 19 37.6, $-$27 19 29.6& 0.7 \\\\ SU116 & 356.4   &\t\t-56   & 23 53 24.6, $-$68 37 42 & 17.1\t \\\\ SU106 & 146.6   &\t\t-43   &  & \\\\ SU106 & 129.9   &\t\t-26   & & \\\\ FE320 & $85.2\\pm 0.5$\t\t&\t$48.0^{+5.2}_{-6.3}$& &\t \\\\ \\hline \\end{tabular} \\caption{Cosmic ray events with energy $E>10^{20}$ eV. For the AGASA events the error is given by the angular cone radius $\\sigma_r=1.8^o$. The corresponding $\\sigma_r$ for the rest of the events excluding the Fly's Eye event is $3^o$ \\cite{cluster}. The blank spaces under the compact QSO column correspond to cases for which the event was within $\\pm 10^o$ of the galactic plane.} \\end{table}  \\bigskip \\noindent {\\bf Acknowledgements:} We thank P. Biermann for providing us with the $K\\ddot{u}hr$ catalog. This work was supported in part under the Department of Energy grant number DE-FGO2-98ER41079, by the University of Kansas General Research Fund, and the {\\it Kansas Institute for Theoretical and Computational Science/ K*STAR} program."
    },
    "0010/gr-qc0010102_arXiv.txt": {
        "abstract": "s#1#2#3{{ \\centering{\\begin{minipage}{4.5in}\\footnotesize\\baselineskip=10pt \\parindent=0pt #1\\par \\parindent=15pt #2\\par \\parindent=15pt #3 \\end{minipage}}\\par}}  \\def\\keywords#1{{ \\centering{\\begin{minipage}{4.5in}\\footnotesize\\baselineskip=10pt {\\footnotesize\\it Keywords}\\/: #1 \\end{minipage}}\\par}}  \\newcommand{\\bibit}{\\nineit} \\newcommand{\\bibbf}{\\ninebf} \\renewenvironment{thebibliography}[1] {\\frenchspacing \\ninerm\\baselineskip=11pt \\begin{list}{\\arabic{enumi}.} {\\usecounter{enumi}\\setlength{\\parsep}{0pt} \\setlength{\\leftmargin 12.7pt}{\\rightmargin 0pt}% \\setlength{\\itemsep}{0pt} \\settowidth {\\labelwidth}{#1.}\\sloppy}}{\\end{list}}  \\newcounter{itemlistc} \\newcounter{romanlistc} \\newcounter{alphlistc} \\newcounter{arabiclistc} \\newenvironment{itemlist} {\\setcounter{itemlistc}{0} \\begin{list}{$\\bullet$} {\\usecounter{itemlistc} \\setlength{\\parsep}{0pt} \\setlength{\\itemsep}{0pt}}}{\\end{list}}  \\newenvironment{romanlist} {\\setcounter{romanlistc}{0} \\begin{list}{$($\\roman{romanlistc}$)$} {\\usecounter{romanlistc} \\setlength{\\parsep}{0pt} \\setlength{\\itemsep}{0pt}}}{\\end{list}}  \\newenvironment{alphlist} {\\setcounter{alphlistc}{0} \\begin{list}{$($\\alph{alphlistc}$)$} {\\usecounter{alphlistc} \\setlength{\\parsep}{0pt} \\setlength{\\itemsep}{0pt}}}{\\end{list}}  \\newenvironment{arabiclist} {\\setcounter{arabiclistc}{0} \\begin{list}{\\arabic{arabiclistc}} {\\usecounter{arabiclistc} \\setlength{\\parsep}{0pt} \\setlength{\\itemsep}{0pt}}}{\\end{list}}  \\newcommand{\\fcaption}[1]{ \\refstepcounter{figure} \\setbox\\@tempboxa = \\hbox{\\footnotesize Fig.~\\thefigure. #1} \\ifdim \\wd\\@tempboxa > 5in {\\begin{center} \\parbox{5in}{\\footnotesize\\smalllineskip Fig.~\\thefigure. #1} \\end{center}} \\else {\\begin{center} {\\footnotesize Fig.~\\thefigure. #1} \\end{center}} \\fi}  \\newcommand{\\tcaption}[1]{ \\refstepcounter{table} \\setbox\\@tempboxa = \\hbox{\\footnotesize Table~\\thetable. #1} \\ifdim \\wd\\@tempboxa > 5in {\\begin{center} \\parbox{5in}{\\footnotesize\\smalllineskip Table~\\thetable. #1} \\end{center}} \\else {\\begin{center} {\\footnotesize Table~\\thetable. #1} \\end{center}} \\fi}  \\def\\@citex[#1]#2{\\if@filesw\\immediate\\write\\@auxout {\\string\\citation{#2}}\\fi \\def\\@citea{}\\@cite{\\@for\\@citeb:=#2\\do {\\@citea\\def\\@citea{,}\\@ifundefined {b@\\@citeb}{{\\bf ?}\\@warning {Citation `\\@citeb' on page \\thepage \\space undefined}} {\\csname b@\\@citeb\\endcsname}}}{#1}}  \\newif\\if@cghi \\def\\cite{\\@cghitrue\\@ifnextchar [{\\@tempswatrue \\@citex}{\\@tempswafalse\\@citex[]}} \\def\\citelow{\\@cghifalse\\@ifnextchar [{\\@tempswatrue \\@citex}{\\@tempswafalse\\@citex[]}} \\def\\@cite#1#2{{$\\null^{#1}$\\if@tempswa\\typeout {IJCGA warning: optional citation argument ignored: `#2'} \\fi}} \\newcommand{\\citeup}{\\cite}  \\def\\pmb#1{\\setbox0=\\hbox{#1} \\kern-.025em\\copy0\\kern-\\wd0 \\kern.05em\\copy0\\kern-\\wd0  \\kern-.025em\\raise.0433em\\box0} \\def\\mbi#1{{\\pmb{\\mbox{\\scriptsize ${#1}$}}}} \\def\\mbr#1{{\\pmb{\\mbox{\\scriptsize{#1}}}}}  \\def\\fnm#1{$^{\\mbox{\\scriptsize #1}}$} \\def\\fnt#1#2{\\footnotetext{\\kern-.3em {$^{\\mbox{\\scriptsize #1}}$}{#2}}}  \\def\\fpage#1{\\begingroup \\voffset=.3in \\thispagestyle{empty}\\begin{table}[b]\\centerline{\\footnotesize #1} \\end{table}\\endgroup}  \\def\\runninghead#1#2{\\pagestyle{myheadings} \\markboth{{\\protect\\footnotesize\\it{\\quad #1}}\\hfill} {\\hfill{\\protect\\footnotesize\\it{#2\\quad}}}} \\headsep=15pt  \\font\\tenrm=cmr10 \\font\\tenit=cmti10 \\font\\tenbf=cmbx10 \\font\\bfit=cmbxti10 at 10pt \\font\\ninerm=cmr9 \\font\\nineit=cmti9 \\font\\ninebf=cmbx9 \\font\\eightrm=cmr8 \\font\\eightit=cmti8 \\font\\eightbf=cmbx8 \\font\\sevenrm=cmr7 \\font\\fiverm=cmr5  \\newtheorem{theorem}{\\indent Theorem}  \\newtheorem{lemma}{Lemma}  \\newtheorem{definition}{Definition} \\newtheorem{corollary}{Corollary}  \\newcommand{\\proof}[1]{{\\bf Proof.} #1 $\\Box$.}    \\newcommand{\\be}{\\begin{equation}} \\newcommand{\\ee}{\\end{equation}}  \\begin{document} \\setlength{\\textheight}{7.7truein}    %  \\runninghead{The $r$-mode instability in rotating neutron stars} {Andersson \\& Kokkotas}  \\normalsize\\textlineskip \\thispagestyle{empty} \\setcounter{page}{1}  \\copyrightheading{}\t\t%  \\vspace*{0.88truein}  \\fpage{1} \\centerline{\\bf THE $R$-MODE INSTABILITY IN ROTATING NEUTRON STARS} \\vspace*{0.37truein}  \\centerline{\\footnotesize NILS ANDERSSON} \\vspace*{0.015truein} \\centerline{ \\footnotesize \\it Department of Mathematics, University of Southampton, Southampton SO17 1BJ, United Kingdom} \\vspace*{10pt} \\centerline{\\footnotesize KOSTAS D. KOKKOTAS} \\vspace*{0.015truein} \\centerline{ \\footnotesize \\it Department of Physics, Aristotle University of Thessaloniki, Thessaloniki 54006, Greece} \\vspace*{0.225truein} \\publisher{(received date)}{(revised date)}  \\vspace*{0.21truein} \\abstracts{In this review we summarize the current understanding of the gravitational-wave driven instability associated with the so-called $r$-modes in rotating neutron stars. We discuss the nature of the $r$-modes, the detailed mechanics of the instability and its potential astrophysical significance. In particular we discuss results regarding the spin-evolution of nascent neutron stars, the detectability of $r$-mode gravitational waves and mechanisms limiting the spin-rate of accreting neutron stars in binary systems.}{}{}  \\vspace*{1pt}\\textlineskip ",
        "introduction": "\\vspace*{-0.5pt}  \\noindent The first decade of this new millenium holds great promise for gravitational physics. The hope is that forty years of development will come to fruition as the new generation of gravitational-wave interferometers (LIGO, VIRGO, GEO600 and TAMA300) come online and reach their projected sensitivities\\cite{gwreview}. These detectors should (finally!) open a new window to the universe and make the long-heralded field of ``gravitational-wave astronomy'' a reality. For gravitational-wave theorists this means that decades of  modelling will finally be tested by observations. The construction of the new detectors is accompanied by attempts to model all likely gravitational-wave sources in appropriate detail. This is highly relevant since reliable theoretical templates, against which one can match the noisy data-stream of the detectors, are needed if we want to extract reliable astrophysical information from future gravitational-wave data.  In this article we survey the territory where stellar pulsation theory meets gravitational-wave astrophysics.  There has been a resurgence of interest in this research area (that dates back to seminal work by Thorne and his colleagues in the late 1960s\\cite{tc67,kst69,ct70}) and our aim  here is to provide an overview of the recent developments. We will focus our discussion on the gravitational-wave driven instability of the so-called $r$-modes. Since its serendipitous discovery a couple of years ago\\cite{a97,fm97}, the $r$-mode instability has attracted considerable attention. In this review article we aim to decribe most of the recent ideas and suggestions, be they speculative or not, and put them in the appropriate context. We will try to introduce the relevant concepts and results from basic principles, in such a way that the article should be accessible to readers with little previous knowledge of this field. We feel that this is an important task since it may help us identify the many outstanding issues that must be addressed by future work.  The theory of stellar pulsation is richly endowed with interesting phenomena, and ever improving observations suggest that most stars exhibit complicated modes of oscillation. Thus it is natural to try to match theoretical models to observed data in order to extract information  about the dynamics of distant stars. This interplay between observations and stellar pulsation theory is known as asteroseismology\\cite{unno}. It is interesting to speculate that the advent of gravitational-wave astronomy will permit similar studies for neutron stars. One would expect most neutron stars to oscillate during various phases of their life, and actual observations would provide invaluable information about the stars internal structure and the supranuclear equation of state\\cite{astero1,astero2}. Given their compactness, with a mass above $1.3M_\\odot$ concentrated inside a radius of a mere $10$~km, oscillating neutron stars could be interesting astrophysical sources of gravitational waves. But it is not clear that astrophysical mechanisms can excite the various oscillation modes to a detectable level. It seems likely that only the most violent processes, such as the actual formation of a neutron star following a supernova or a dramatic starquake following, for example, an internal phase-transition, will be of relevance. The strengthening evidence for magnetars\\cite{duncan,magnetar}, in which a starquake could release large amounts of energy, is also very interesting in this respect. One can estimate that these events must take place in our immediate neighbourhood (the Milky Way or the Local Group) in order to be observable.  This does not, however, rule out the possibility that neutron star pulsation may be detected by the new generation of gravitational-wave detectors. First of all, one should not discard the possibility that unique events in the life of a neutron star may excite the various modes to a relevant level. More likely to be observationally important is the long recognized possibility that various pulsation modes of rapidly rotating neutron stars may be unstable due to the emission of gravitational radiation\\cite{cfsreview,jlfrev,nslr}. Should such an instability operate in a young neutron star it may lead to the emission of copious amounts of gravitational waves\\cite{o98}. Of particular recent interest is the instability of the so-called $r$-modes, for which the gravitational waves have been estimated to be detectable for sources in the Virgo cluster (at 15-20~Mpc). If we suppose that most newly born neutron stars pass through a phase where this kind of instability is active, several such events  should be observed per year once the advanced interferometers come into operation. This is a very exciting prospect, indeed. ",
        "conclusions": "In this review we have discussed the $r$-modes in rotating neutron stars and the associated gravitational-wave driven instability. This is a research area that has  attracted considerable interest following the prediction that hot young neutron stars may be spun down to a level comparable to that extrapolated from observations and the suggestion that the associated gravitational waves may be detectable with the new generation of interferometers.  Our aim in writing this review was to summarize most of the ideas and suggestions that have been made regarding the $r$-mode instability and put them in the appropriate context. In doing this we obviously had to make sacrifices, but hopefully we have managed to provide a useful introduction to the many relevant issues. What should be clear is that our present understanding of instabilities in rotating neutron stars and their potential astrophysical relevance  is unsatisfactory in many respects. All current suggestions are based on simplified, often phenomenological, models. The hope must be that future investment in this field will lead to a truly quantitative description of ``realistic'' neutron stars, and that the various proposed astrophysical scenarios will be described in detail rather than ``order of magnitude''. We have far to go before we reach this goal, and it may well be that actual observations will beat us theorists to the answers of the many challenging questions that we are just beginning to formulate.  \\nonumsection{Acknowledgements} \\noindent We would like to thank many of our colleagues for stimulating discussions of the issues described in this review. In particular,  we a grateful to John Friedman, Ian Jones, Luciano Rezzolla, Bernard Schutz and Nick Stergioulas for insightful comments. This work was partly supported by PPARC grant PPA/G/1998/00606 to NA.  \\nonumsection{References} \\noindent  \\def\\prl#1#2#3{{ Phys. Rev. Lett.\\ }, {\\bf #1}, #2 (#3)} \\def\\prd#1#2#3{{ Phys. Rev. D}, {\\bf #1}, #2 (#3)} \\def\\plb#1#2#3{{ Phys. Lett. B}, {\\bf #1}, #2 (#3)} \\def\\prep#1#2#3{{ Phys. Reports}, {\\bf #1}, #2 (#3)} \\def\\phys#1#2#3{{ Physica}, {\\bf #1}, #2 (#3)} \\def\\jcp#1#2#3{{ J. Comput. Phys.}, {\\bf #1}, #2 (#3)} \\def\\jmp#1#2#3{{ J. Math. Phys.}, {\\bf #1}, #2 (#3)} \\def\\cpr#1#2#3{{ Computer Phys. Rept.}, {\\bf #1}, #2 (#3)} \\def\\cqg#1#2#3{{ Class. Quantum Grav.}, {\\bf #1}, #2 (#3)} \\def\\cma#1#2#3{{ Computers Math. Applic.}, {\\bf #1}, #2 (#3)} \\def\\mc#1#2#3{{ Math. Compt.}, {\\bf #1}, #2 (#3)} \\def\\apj#1#2#3{{ Astrophys. J.}, {\\bf #1}, #2 (#3)} \\def\\apjl#1#2#3{{ Astrophys. J. Lett.}, {\\bf #1}, #2 (#3)}  \\def\\apjs#1#2#3{{ Astrophys. J. Suppl.}, {\\bf #1}, #2 (#3)} \\def\\acta#1#2#3{{ Acta Astronomica}, {\\bf #1}, #2 (#3)} \\def\\sa#1#2#3{{ Sov. Astro.}, {\\bf #1}, #2 (#3)} \\def\\sia#1#2#3{{ SIAM J. Sci. Statist. Comput.}, {\\bf #1}, #2 (#3)} \\def\\aa#1#2#3{{ Astron. Astrophys.}, {\\bf #1}, #2 (#3)} \\def\\apss#1#2#3{{Astrop. Sp. Sci.}, {\\bf #1}, #2 (#3)} \\def\\mnras#1#2#3{{ Mon. Not. R. Astr. Soc.}, {\\bf #1}, #2 (#3)} \\def\\prsla#1#2#3{{ Proc. R. Soc. London, Ser. A}, {\\bf #1}, #2 (#3)} \\def\\ijmpc#1#2#3{{ I.J.M.P.} C {\\bf #1}, #2 (#3)}"
    },
    "0010/astro-ph0010145_arXiv.txt": {
        "abstract": "I compare gas-dynamical and stellar-dynamical models of collisions. These two models have distinctly different physics; for example, shocks introduce irreversibility in gas systems, while stellar systems evolve in a completely reversible fashion.  Nonetheless, both models yield broadly similar results, suggesting that analogies between gas and stellar dynamics have some heuristic validity even applied to collisions. ",
        "introduction": "Stars and galaxies are prototypical examples of self-gravitating systems.  While both are held together by gravity, they differ in size, makeup, and structure.  From a dynamical point of view, the basic difference is that stars are made up of particles which {\\it often\\/} undergo collisions, while galaxies are made up of particles which almost {\\it never\\/} collide (this conference notwithstanding).  Analogies between stellar and gaseous systems run deep.  For example, there is a close relationship between spherical stellar systems with isotropic distribution functions $f = f(E)$ and gaseous systems with barotropic equations of state $P = P(\\rho)$.  The stability properties of spherical stellar and gaseous systems provide further analogies; an isotropic stellar system with $df/dE < 0$ is stable if the barotropic gas-sphere with the same density profile is stable (Antonov 1962; Lynden-Bell 1962).  This is a sufficient but not necessary condition; on the whole, spherical stellar systems seem to be {\\it more\\/} stable than their gaseous counterparts.  For flattened systems the situation is more complex.  Infinite, uniformly rotating sheets of gas and stars have very similar stability criteria (Toomre 1964).  On the other hand, certain finite, uniformly-rotating stellar disks (Kalnajs 1972) are much {\\it less\\/} stable than gaseous disks with the same density profile.  \\begin{figure} \\begin{center} \\epsfig{figure=fig01a.ps,width=\\textwidth} \\end{center} \\caption{\\small Gas (top) and grit (bottom) versions of the same collision.  Initially the two bodies approach on a parabolic relative orbit, passing each other in a counter-clockwise direction.  The mass ratio is $M_1/M_2 = 2$ and the targeted pericentric separation $r_{\\rm p} = 0.5$ is half the radius of the larger body.} \\label{fig01} \\end{figure}  Questions of existence and stability exploit analogies between stellar and gaseous systems at or near equilibrium.  In this paper I use numerical simulations to examine how such analogies work in situations which are far from equilibrium: collisions.  Colliding stars and colliding spherical galaxies evolve in superficially similar ways, as seen in Fig.~\\ref{fig01}.  Here the top row of frames shows a collision of two gas-spheres, while the bottom row shows the analogous collision of two spherical systems of collisionless particles, hereafter labeled ``grit''\\footnote{In my original presentation I used ``dust'' instead of ``grit'', borrowing this term from cosmology. However, S.~Shapro pointed out that ``dust'' is only applicable to a cold, pressureless medium.  Casting about for a suitable term, I chose ``grit'' since this conveys some resilience to pressure.}.  A more detailed comparison of the two collisions in Fig.~\\ref{fig01} reveals significant differences.  For example, it's evident that orbit decay doesn't work in quite the same way; contrasting the third pair of frames, the two bodies are much closer in the gas model than they are in the grit model.  Moreover, grit evolves {\\it reversibly\\/}, while gas evolves {\\it irreversibly\\/}; by integrating the merged grit model backward in time, one can unscramble the wreckage and recover the initial conditions (van Albada \\& van Gorkom 1977), but the presence of shocks in the gas model precludes such reversals.  \\subsection{Gas model}  Technically, the interior of a normal star is a partly-ionized plasma in local equilibrium with a black-body radiation field; support against gravity is provided by some combination of thermal and radiation pressure.  But for low-mass stars the latter is insignificant, and to a good approximation the pressure is given by the equation of state for an ideal gas.  Gas dynamics unfolds in a space of three dimensions; the dynamical variables are functions of position ${\\bf r}$ and time $t$.  Two such variables are the density $\\rho = \\rho({\\bf r}, t)$ and the velocity field ${\\bf v} = {\\bf v}({\\bf r}, t)$.  One more variable is needed to represent the thermodynamic state of the gas; in these calculations I use the entropy function $a = a(S)$, which enters directly in the equation of state: \\begin{equation} P = a(S) \\rho^\\gamma \\, , \\label{equation-of-state} \\end{equation} where $P$ is the pressure, and $\\gamma$ is the ratio of specific heats, here assigned the value $\\gamma = 5/3$ appropriate for a monatomic gas.  I adopt the following dynamical equations: \\begin{equation} \\frac{\\partial \\rho}{\\partial t} + \\frac{\\partial}{\\partial {\\bf r}} \\cdot (\\rho {\\bf v}) = 0 \\, , \\label{continuity-equation} \\end{equation} \\begin{equation} \\frac{\\partial {\\bf v}}{\\partial t} + \\left( {\\bf v} \\cdot \\frac{\\partial}{\\partial {\\bf r}} \\right) {\\bf v} = - \\frac{1}{\\rho} \\frac{\\partial P}{\\partial {\\bf r}} - \\frac{\\partial \\Phi}{\\partial {\\bf r}} \\, , \\label{euler-equation} \\end{equation} \\begin{equation} \\frac{\\partial a}{\\partial t} + \\left( {\\bf v} \\cdot \\frac{\\partial}{\\partial {\\bf r}} \\right) a = (\\gamma - 1) \\rho^{1 - \\gamma} \\, \\dot{u}_{\\rm v} \\, . \\label{entropy-equation} \\end{equation} Here eq.~\\ref{continuity-equation} represents conservation of mass and eq.~\\ref{euler-equation} represents conservation of momentum; the gravitational potential $\\Phi$ is calculated from Poisson's equation. Eq.~\\ref{entropy-equation} describes the evolution of the entropy function; the flow is adiabatic, conserving the entropy of each fluid element, except where the viscous heating function $\\dot{u}_{\\rm v} > 0$.  A standard form of artificial viscosity (Monaghan \\& Gingold 1983) is used to implement the increase in gas entropy due to shocks.  I solve these dynamical equations using Smoothed Particle Hydrodynamics or SPH (eg.~Monaghan 1992).  The code is similar to ``TREESPH'' (Hernquist \\& Katz 1989); it includes a hierarchical algorithm to compute gravitational forces, individual smoothing radii $h_i$ set so that that each particle $i$ interacts with a fixed number of neighbors, and individual time-steps adjusted to satisfy a Courant condition.  The code doesn't include terms proportional to $\\nabla h_i$ which arise when the SPH equations are derived from a Hamiltonian (Nelson \\& Papaloizou 1993, 1994).  Since the code integrates Eq.~\\ref{entropy-equation} instead of the equivalent energy equation, the neglect of these $\\nabla h_i$ terms leads to imperfect energy conservation (Hernquist 1993).  Most of the calculations presented below conserve energy to $\\sim 1$\\% or better, so the neglect of these terms is probably not critical in this application.  \\subsection{Grit model}  The dominant mass components in typical galaxies are stars and dark matter.  In the absence of conflicting evidence the latter is often assumed to be composed of particles with masses much less than $10^6 M_\\odot$ (cf.~Lacey \\& Ostriker 1985).  The gravitational potential of a galaxy is quite smooth and individual stars or dark matter particles follow smooth trajectories, undisturbed by collisions, close encounters, or any effect due to the discrete nature of the mass.  Stellar dynamics unfolds in a space of six dimensions; the phase-space distribution $f = f({\\bf r}, {\\bf v}, t)$ is a function of position ${\\bf r}$, velocity ${\\bf v}$, and time $t$.  In the limit appropriate for galaxies, $f$ evolves according to the Collisionless Boltzmann Equation or CBE: \\begin{equation} \\frac{\\partial f}{\\partial t} + {\\bf v} \\cdot \\frac{\\partial f}{\\partial {\\bf r}} - \\frac{\\partial \\Phi}{\\partial {\\bf r}} \\cdot \\frac{\\partial f}{\\partial {\\bf v}} = 0 \\label{vlasov-equation} \\end{equation} Like eq.~\\ref{continuity-equation}, eq.~\\ref{vlasov-equation} is a continuity equation, conserving the mass of the system.  But eq.~\\ref{vlasov-equation} describes an {\\it incompressible\\/} flow in six dimensions; dynamics moves elements of phase-fluid around but conserves the value of $f$ associated with each.  I solve the CBE using standard N-body techniques (eg.~Barnes 1998). The code uses a hierarchical algorithm to compute gravitational forces and a simple leap-frog integrator to advance particle coordinates.  \\subsection{Polytropes}  I chose to collide simple polytropes, instead of using stellar models. While accurate simulations of merging stars require realistic stellar models (Sills \\& Lombardi 1997), polytropes are still appropriate for comparing the basic physics of stellar and galactic collisions. Polytropes are easily constructed, and their properties are well-understood (eg.~Chandrasekhar 1939).  Moreover, grit models of polytropes have simple distribution functions (Eddington 1916) and the stability of these systems has been studied using N-body simulations (H\\'enon 1973; Barnes, Goodman, \\& Hut 1986).  Both gas and grit polytropes obey the following relationship between mass density $\\rho$ and gravitational potential $\\Phi$: \\begin{equation} \\rho = \\rho(\\Phi) = \\rho_1 (1 - \\Phi / \\Phi_1)^n \\, . \\label{polytrope-density} \\end{equation} Here $\\Phi_1$ is the value of the potential on the surface of the polytrope and $\\rho_1$ is a constant with units of density.  By construction, gas polytropes are marginally stable to convection, and this implies that the index $n$ is related to the ratio of specific heats by $n = 1 / (\\gamma - 1)$.  For grit polytropes, $n$ is a free parameter, and the distribution function takes the form \\begin{equation} f = f(E) = \\left\\{ \\begin{array}{ll} f_1 (1 - E / \\Phi_1)^{n - 3/2} & \\mbox{if $E<\\Phi_1$,} \\\\ 0                              & \\mbox{otherwise,} \\end{array} \\right. \\label{polytrope-distribution} \\end{equation} where $f_1$ is a constant with units of phase-space density.  For a monatomic gas, $\\gamma = 5/3$, and the appropriate index $n = 3/2$.  The corresponding grit polytrope has an unusually simple structure; $f$ has the constant value $f_1$ within the six-dimensional volume defined by $E = \\frac{1}{2}v^2 + \\Phi(r) < \\Phi_1$, and vanishes everywhere else.  To construct realizations of polytropes I first tabulated the density profile $\\rho(r)$, potential $\\Phi(r)$, and pressure $P(r)$.  Grit models were generated by chosing initial positions ${\\bf r}_i$ and velocities ${\\bf v}_i$ for each particle according to the distribution function $f({\\bf r}, {\\bf v}) = f(\\frac{1}{2}v^2 + \\Phi(r))$ (H\\'enon 1973).  Likewise, gas models were generated by chosing initial positions ${\\bf r}_i$ according to the density profile $\\rho(r)$, setting initial velocities ${\\bf v}_i = 0$, and assigning initial entropy function values $a_i = P(r_i) / \\rho^\\gamma(r_i)$.  In both cases all particles had equal masses.  Since particle coordinates were assigned independently, these initial configuration have Poissonian fluctuations.  Such fluctuations are inevitable in N-body simulations, but they are much larger than the fluctuations normally present in SPH calculations.  Therefore, the gas realizations were relaxed by evolving them with the SPH code using a velocity-damping term.  This has the effect of ``ironing out'' the initial density fluctuations without disturbing the overall density profile. ",
        "conclusions": "Comparison of gas-dynamical and stellar-dynamical models of colliding systems reveals many specific differences and hints at some underlying similarities.  While stellar systems interact via gravity alone, gas systems interact via a complex mixture of gravity, pressure forces, and shocks.  This mix greatly increases the range of dynamical behavior in stellar collisions.  Even when interactions between gas systems are limited to tides, the resulting deformations are stronger than those seen in equivalent encounters of stellar systems.  Remnants of gas-dynamical and stellar-dynamical mergers also exhibit significant differences.  Stellar systems have anisotropic velocity distributions, supporting a wide range of remnant morphologies, while gas systems settle into a small range of equilibria.  But at the same time, the stellar and gaseous remnants studied here have very similar density profiles, and often similar shapes as well.  This seems remarkable, since the physics behind these density profiles is rather different.  The extended envelopes of gaseous remnants contain high-entropy material produced by shocks, which are violent, well-localized events resulting from physical encounters.  On the other hand, the envelopes of stellar remnants contain material extracted from the participants by tides; this material retains the high fine-grain phase-space density it starts with, and only when coarse-grained does it approximate the atomic velocity distribution of a gaseous envelope.  Two puzzles remain: first, why do such disparate mechanisms -- shocks and tides -- produce such similar remnant envelopes?  And second, how is coarse-graining -- an operation applied to a stellar system by an observer -- analogous to the irreversible physical processes which occur in gaseous systems?"
    },
    "0010/astro-ph0010659_arXiv.txt": {
        "abstract": "{ We present initial results from observations of the low-mass X-ray binary \\object{EXO~0748$-$67} with the Reflection Grating Spectrometer on board the {\\it XMM-Newton} Observatory.  The spectra exhibit discrete structure due to absorption and emission from ionized neon, oxygen, and nitrogen.  We use the quantitative constraints imposed by the spectral features to develop an empirical model of the circumsource material.  This consists of a thickened accretion disk with emission and absorption in the plasma orbiting high above the binary plane.  This model presents challenges to current theories of accretion in X-ray binary systems. ",
        "introduction": "X-ray spectroscopic observations should provide a useful probe of the accretion processes in low-mass X-ray binaries (LMXB's).  The continuum emission from the bright central source is reprocessed as the photons are absorbed and re-emitted in the surrounding material.  The details of the resulting X-ray spectra are extremely sensitive to the physical conditions in the plasma and therefore provide an excellent way to constrain the circumstellar environment and the mass flow that fuels the X-ray emission.  Previous observatories have not had the spectral resolution necessary to resolve the discrete structure in the X-ray spectrum.  The Reflection Grating Spectrometer (RGS) on the recently launched {\\it XMM-Newton} Observatory provides both a high resolving power and a large effective area making it uniquely sensitive to the diagnostic spectral features, particularly in the soft X-ray band that contains transitions from highly ionized charge states of the most abundant elements.  In this letter we present the results of RGS observations of \\object{EXO~0748$-$67}, a highly variable LMXB that was first discovered with {\\it EXOSAT} (Parmar et al. \\cite{Parmar86}).  Analysis of the {\\it EXOSAT} light curves, which show deep eclipses with a 3.82 hour orbital period and a complex dipping structure, led to a derived inclination angle of $75\\degr-82\\degr$ (Parmar et al. \\cite{Parmar86}).  The detection of type I X-ray bursts confirmed that the compact object is a neutron star (Gottwald et al. \\cite{Gottwald}).  Observations with both {\\it ASCA} (Thomas et al. \\cite{Thomas}) and {\\it ROSAT} (Schulz \\cite{Schulz}) revealed structure in the soft X-ray spectrum, but the limited spectral resolution of these instruments made it impossible to distinguish whether this was due to absorption or emission features.  In the RGS observations that we present below, we find a spectrum that is rich in discrete structure including absorption and emission from ionized oxygen, nitrogen, and neon.  We use the available spectral diagnostics to construct a quantitative empirical model of the circumsource flow. ",
        "conclusions": "From the observed spectral behavior and the derived source parameters we can construct an empirical model of the circumstellar material. The large equivalent widths of the emission lines require that the emission regions subtend a fairly large solid angle in order to be sufficiently illuminated by the central source.  These regions must also extend high above the plane of the binary or else we would have observed eclipses or other modulations correlated with the orbital phase.  We can estimate the geometry of this material by comparing the observed absorption and emission in the hydrogen-like and helium-like oxygen ions.  For a plasma in photoionization equilibrium the ionization rate is equal to the recombination rate everywhere in the gas.  This means that the number of photons that are absorbed by a particular ion can be directly compared to the number of photons emitted by that ion.  Given the efficiency of cascade through a particular line transition, $\\eta_{\\rm line}$, this can be expressed as \\begin{equation} N_{\\rm line}=\\eta_{\\rm line} N_{\\rm recombined} = \\eta_{\\rm line} N_{\\rm absorbed} \\,, \\end{equation} where $N$ is the photon flux.  The ratio of observed absorption to observed emission is then determined by the geometry of the system.  For example, for a disk oriented with its axis along the line of sight, photons will be absorbed in the surrounding material but then re-emitted isotropically, so that the observed line emission would be higher than expected from the observed absorption.   For \\object{EXO~0748$-$67} we have compared the number of photons absorbed at the \\ion{O}{viii} and \\ion{O}{vii} edges to the number of photons emitted in the \\ion{O}{viii} Ly$\\alpha$ and \\ion{O}{vii} He-like complexes.  We used cascade efficiencies ($44\\%$ for \\ion{O}{viii} and $75\\%$ for \\ion{O}{vii}) calculated with the atomic model for a recombining plasma used in Sako et al. (\\cite{Sako}).  We see approximately three times more absorption than expected in each of the ions.  The absorbing material must therefore be flattened along our line of sight with a solid angle of roughly $\\frac{1}{3} 4\\pi$.  Since we know that the system is highly inclined, this material must be aligned with the plane of the binary and therefore with the accretion disk.  The fact that the emission lines show large velocity broadening with no systemic Doppler-shifts suggests that the observed widths are not associated with net outflow or inflow but rather with orbital motion around the central source.  Using the ionization parameters of formation calculated as described in Sako et al. (\\cite{Sako}) (see Table ~\\ref{TabValues} for these values) we find a correlation with velocity that is consistent with an orbital velocity structure; \\ion{Ne}{x}, which has the largest ionization parameter and is preferentially emitted closest to the central source, is observed to be moving with the largest velocity while \\ion{N}{vii}, which has the lowest ionization parameter and should be emitted farthest from the source, is observed to be moving with the lowest velocity.  The \\object{EXO~0748$-$67} system must therefore contain a thickened accretion disk where the orbiting material extends high above the binary plane.  The plasma in the uppermost regions probably produces both the emission and absorption features in the observed spectra.  The continuum intensity is independent of orbital phase so the central source region must be extended; a compact X-ray source would show orbital modulation in such a highly inclined system.  Since the flux in the emission lines remains constant, the source luminosity must be constant; variations in luminosity would cause changes in the ionization structure and therefore changes in both the line flux and the line ratios.  The optical depths at the absorption edges are also constant so we must always be looking through the same absorbing and emitting material.  The observed variations in continuum intensity are best explained by local obscurations of the central source region. Interestingly, the velocity parameters derived for the optically-emitting material ( $\\sim 2000 \\, {\\rm km\\, s^{-1}}$ line broadening with $210 \\pm 92\\, {\\rm km\\, s^{-1}}$ amplitude Doppler-shifts (Crampton et al. \\cite{Crampton})) are very similar to those measured here.  Although it must be in a very different region of ionization, it is possible that the optically-emitting material is also orbiting in this thickened disk.   As a simple test of this model we can compare the emission measure for an orbiting plasma to the empirical emission measure.  Assuming a velocity profile with $v=(\\frac{GM}{r})^{1/2}$ and a mass of $1.4 M_{\\sun}$ for the neutron star, we can calculate the radial distances for each ion from the measured velocities.  Then, using the ionization parameters of formation ($\\xi=\\frac{L}{n_{\\rm e} r^2}$) and the radial distances we can calculate the densities in these regions assuming a source luminosity of $1 \\times 10^{36} {\\rm ergs \\, s^{-1}}$.  Defining the emission measure as \\begin{equation} {\\rm EM}=\\int dVn_{\\rm e}^2 \\simeq \\Delta \\Omega r^{2} \\Delta r n_{\\rm e}^2 \\,, \\end{equation} we take $\\Delta r = \\frac{1}{2} r$, and $\\Delta \\Omega = \\frac{1}{2} \\frac{1}{3} 4\\pi$ to account for the estimated geometry.  The resulting emission measures as well as the components of the calculation are given in Table ~\\ref{TabValues}.  We find emission measures that range from $\\sim 22 \\times 10^{57} {\\rm cm}^{-3}$ for \\ion{Ne}{x} Ly$\\alpha$ to $\\sim 10 \\times 10^{57} {\\rm cm}^{-3}$ for \\ion{N}{vii} Ly$\\alpha$.  To calculate the empirical emission measure we divided the line luminosity by the line power evaluated at the ionization parameter of formation.  The line power for each ion was calculated using the HULLAC atomic codes (Bar-Shalom et al. \\cite{BarShalom}) following the procedure described in Sako et al. (\\cite{Sako}). We adjusted our measured line fluxes to account for interstellar absorption using the cross sections of Morrison \\& McCammon (\\cite{Morisson}) with a column density of $1.1 \\times 10^{21} {\\rm cm^{-3}}$.  We assumed a distance to the source of $D=10 \\, {\\rm kpc}$ (Gottwald et al. \\cite{Gottwald}).  The resulting emission measure for each ion, assuming solar abundances (Anders $\\&$ Grevesse \\cite{Anders}), is included in Table ~\\ref{TabValues}.  For \\ion{Ne}{x} and \\ion{N}{vii} we find emission measures of $2.1$ and $2.6 \\times 10^{57} {\\rm cm^{-3}}$.  For \\ion{O}{viii} we find an emission measure of $0.6\\times 10^{57} {\\rm cm^{-3}}$.  This suggests an underabundance of oxygen that may have interesting implications for the evolutionary state of the companion star.  Adopting the factor of two uncertainty in the empirical emission measure that is estimated by Sako et al. (\\cite{Sako}), we find that the predicted emission measures are roughly a factor of five higher than the empirical emission measure except at \\ion{O}{viii}.  Considering that the predicted emission measures are highly dependent on our geometric assumptions and on a velocity profile that is only strictly valid for a thin Keplerian disk, the agreement with the empirical emission measures is remarkably good.  This empirical model of a thickened disk is difficult to understand theoretically.  For a disk in hydrostatic equilibrium, the vertical height at a particular radius should be roughly given by the product of the radius and the ratio of the sound speed of the gas to its local orbital velocity.  Here, we can directly infer the sound speed from the ionization parameter and the orbital velocity from the widths of the lines, yet we infer a disk height that grossly violates this condition.  However, despite the theoretical challenge that this presents, we find no alternative model that satisfies the empirical constraints.  The large observed line equivalent widths and the clear evidence for intrinsic absorption essentially guarantee that this material is well out of hydrostatic equilibrium.  Hopefully, as high-resolution spectroscopic observations of other LMXB's accumulate, we may gain a clearer understanding of how such accretion flows can be formed and maintained."
    },
    "0010/nucl-ex0010014_arXiv.txt": {
        "abstract": "The \\be7pg reaction is one of the major source of uncertainties in estimating the \\b8 solar neutrino flux and is critical for undertsanding the Solar Neutrino Problem and neutrinos. The main source of uncertainty is the existence of conflicting data with different absolute normalization. Attempts to measure this reaction rate with \\xbe7 beams are under way by the UConn-LLN collaboration, and we discuss a newly emerging method to extract this cross section from the Coulomb dissociation of the radioactive beam of \\b8. We discuss some of the issues relevant for this study including the question of the E2 contribution to the Coulomb dissociation process which was measured to be small. The Coulomb dissociation appears to provide a viable alternative method for measuring the \\be7pg reaction rate, with a weighted average of the RIKEN1, RIKEN2 and GSI1 published results of $ S_{17}(0) = 19.4 \\pm 1.3 $ eV-b. ",
        "introduction": "The solar neutrino problem \\cite{Book,Phs96} may allow for new break through in our understanding of the standard model~\\cite{Wo78,Mi85}. The precise knowledge of the astrophysical $S_{17}$-factor of the $^7$Be(p,$\\gamma$)$^8$B reaction at the Gamow peak, about 20 keV, is of crucial importance for interpreting terrestrial measurements of the solar neutrino flux~\\cite{Ade98}. This is particularly true for the interpretation of results from the Homestake, Kamiokande, SuperKamiokande and SNO experiments ~\\cite{Phs96} which measured high energy solar neutrinos mainly or solely from $^8$B decay.  Eight direct measurements were reported for the $^7$Be(p,$\\gamma$)$^8$B reaction ~\\cite{Kavanagh60,Parker68,Kavanagh69,Vaughn70,Wiezorek77,Filippone83,Hammache98, Ha99}. Since the cross section at $E_{\\rm cm} \\approx$~20~keV is too small to be measured, the $S_{17}$-factors at low energies have to be extrapolated from the experimental data with the help of theoretical models (see e.g. Johnson {\\it et al.}~\\cite{Johnson92}, Jennings {\\it et al.}~\\cite{Jennings98}). But also in the energy range above $E_{\\rm cm} \\approx$~100~keV, where measurements can be performed, experimental difficulties, mainly connected with the determination of the effective target thickness of the radioactive $^7$Be target, hamper the derivation of accurate results. These difficulties are reflected in the fact that the six precise measurements~\\cite{Parker68,Kavanagh69,Vaughn70,Filippone83,Hammache98, Ha99} can be grouped into two distinct data sets which agree in their energy dependence but disagree in their absolute normalization by about 30\\%. Since this discrepancy is larger than the error of adopted $S_{17}$ in the standard solar model, experimental studies with different methods are highly desirable for improving the reliability of the input to the standard solar model~\\cite{Bahcall92}.  In Fig. 1 we show the world data including our new GSI measurement of $S_{17}$ \\cite{Iw99}.  It is clear from this figure that if we are to quote $S_{17}(0)$ with an accuracy of $\\pm 5\\%$, many more experiments using different methods are required.  \\centerline{\\psfig{figure=fig1.eps,height=4in}}  \\hspace{0.5in} \\underline{Fig. 1:}  World data on $S_{17}$ as reported by our GSI collabortion \\cite{Iw99}.  \\subsection{Extrapolation Procedures}  The discrepancy in the measured absolute value of the cross section of the \\be7pg reaction is quite possibly best addressed with a $^7Be$ radioactive beam and a hydrogen target, allowing for a direct measurement of the beam-target luminosity as we propose to do at LLN. However, additional uncertainty exists in the theoretical extrapolation of the measured cross section to solar energies (approximately 20 keV). A few theoretical studies suggest an extrapolation procedure that is accurate to approximately $\\pm 1\\%$ \\cite{Jennings98}.  Without discussing these rather strong statements we consider a similar situation that haunted Nuclear Astrophysics a few years back-- the S-factor of the $d(d,\\gamma)^4He$ reaction. It was assumed that in this case d-waves dominate and no nuclear structure effects should play a role at very low energy, as low as 100 keV. Much in the same way, it is stated today that s-waves dominate the \\be7pg reaction and we do not expect nuclear structure effects to play a role at low energies in the \\be7pg reaction. In Fig. 2 we show Fowler's extrapolated d-wave S-factor that is a mere factor of 32 smaller than measured, due to a small non d-wave component in the d + d interaction \\cite{Ba87}. A small nuclear structure effect, namely the d-wave component of the ground state of $^4He$, gives rise to a change by a factor of 32 in the predicted astrophysical S-factor. Similarly we may ask whether a small non s-wave component in the low energy interaction of p + \\xbe7 could alter the extrapolated $S_{17}(0)$ value by considerably more than one percent. A measurement of $S_{17}(0)$ with an accuracy of $\\pm 5\\%$ mandates that the cross section be measured at low energies, as low as possible, so as to also test the extrapolation procedures.  \\centerline{\\psfig{figure=ddg.eps,height=4in}}  \\hspace{0.5in} \\underline{Fig. 2:}  Extrapolation of d-wave S-factor of the $d(d,\\gamma)^4He$ reaction\\cite{Ba87}. Note the presence of small non d-wave components that yield a discrepancy from Fowler's extracted S-factor by a factor of 32. ",
        "conclusions": ""
    },
    "0010/astro-ph0010367_arXiv.txt": {
        "abstract": "We present \\xmm\\ observations of the high redshift ($z=3.104$), radio-loud quasar PKS~0537$-$286. The EPIC CCD cameras provide the highest signal-to-noise spectrum of a high-z quasar to date. The EPIC observations show that PKS~0537$-$286 is extremely X-ray luminous (L$_{X}=2\\times10^{47}$~erg~s$^{-1}$), with an unusually hard X-ray spectrum ($\\Gamma=1.27\\pm0.02$). The flat power-law emission extends over the whole observed energy range (0.4 to 40 keV in the quasar rest frame); there is no evidence of intrinsic absorption, which has been claimed in PKS~0537$-$286 and other high $z$ quasars.  However, there is evidence for weak Compton reflection. A {\\it redshifted} iron K line, observed at 1.5~keV - corresponding to $\\sim$6.15~keV in the quasar rest frame - is detected at 95\\% confidence. If confirmed, this is the most distant iron K line known. The line equivalent width is small (33eV), consistent with the `X-ray Baldwin effect' observed in other luminous quasars. The reflected continuum is also weak (R~$\\la$0.25). We find the overall spectral energy distribution of PKS~0537$-$286 is dominated by the X-ray emission, which, together with the flat power-law and weak reflection features, suggests that the X-radiation from PKS~0537$-$286 is dominated by inverse Compton emission associated with a face-on relativistic jet.  \\begin{keywords} galaxies: active -- quasars: individual: PKS~0537$-$286 -- X-rays: galaxies \\end{keywords} ",
        "introduction": "At the present time there is little known about the X-ray spectra of high redshift quasars, and thus a lack of understanding of the nature of the quasar central engine at the highest redshifts. However the situation is improving; the \\rosat\\ mission substantially increased the number of high $z$ quasars with detected X-ray emission (Brinkmann \\et 1997), whilst \\asca\\ obtained the first broad-band spectra (e.g. Cappi \\et 1997, Reeves \\et 1997, Vignali \\et 1999), albeit with rather limited photon statistics and sensitivity. The high throughput and bandpass of \\xmm\\ is about to revolutionize studies of high redshift quasars. Many high redshift quasars should be found routinely through follow-up surveys of XMM fields (Watson \\et 2001), detailed spectral studies can be performed on the brightest objects and the high sensitivity limit of \\xmm\\ should enable the detection of the most distant known quasars at $z>5$ (Brandt \\et 2000).  At $z=3.104$, PKS~0537$-$286 is one of the brightest known high $z$ quasars, first recognised as a 1 Jy Parkes southern radio source (e.g. Bolton \\et 1975) and later identified as a QSO (Wright \\et 1978). The optical (rest frame UV) spectrum shows two strong absorption systems associated with the quasar, one of which produces a cut-off in the quasar continuum at the Lyman-limit. The redshift of PKS~0537$-$286 has been confirmed at $z=3.104\\pm0.001$ (Osmer \\et 1994). PKS~0537$-$286 was first detected as an X-ray source by the Einstein observatory (Zamorani \\et 1981), with a position within 1\\arcm\\ of the radio and optical source. The identification was later confirmed by \\rosat\\ (B\\\"{u}hler \\et 1995), and with \\asca\\ there were enough counts to obtain a crude X-ray spectrum (Siebert \\et 1996, Cappi \\et 1997, Reeves \\et 1997). These observations revealed a luminous X-ray source, with a flat X-ray spectrum perhaps indicative of a soft X-ray cut-off due to absorption. We now present the \\xmm\\ observations of PKS~0537$-$286. Values of $ H_0 = 50 $~km\\,s$^{-1}$\\,Mpc$^{-1}$ and $ q_0 = 0.5 $ have been assumed and all fit parameters are given in the quasar rest-frame. ",
        "conclusions": "The \\xmm\\ observations have showed that PKS~0537$-$286 is an extremely luminous quasar, with a particularly flat X-ray spectrum and weak iron K emission and reflection features. Indeed the total power output of PKS~0537$-$286 is dominated by the X-ray emission. A likely interpretation is that the X-rays mainly arise through the inverse Compton emission associated with a relativistic jet. In this scenario relativistic electrons (or positrons) in the jet Compton upscatter either the low energy synchrotron photons produced in the jet (Synchrotron Self Compton - Jones, O'Dell \\& Stein 1974, Ghisellini, Maraschi \\& Treves 1985) or alternatively the thermal optical/UV photons from the accretion disk (External Inverse Compton - Ghisellini \\& Madau 1996). This can naturally account for the hard spectrum of PKS~0537$-$286, as in the X-ray band we are observing the onset of this Inverse Compton component, presumably extending out to $\\gamma$-ray energies.  \\begin{figure} \\resizebox{\\hsize}{!}{\\rotatebox{-90}{\\includegraphics{figure5.ps}}} \\caption{The radio to X-ray spectral energy distribution of PKS~0537$-$286. The total power output of the quasar is dominated by the X-ray emission. The SED is similar to other high luminosity blazars at high redshift (see Fabian \\et 1999).} \\end{figure}  As it appears that we can rule out a scenario whereby all of the primary X-ray emission originates above the accretion disk, we instead tried to fit the data with two continuum emission components; a `Seyfert-like' steep power-law (with $\\Gamma$ fixed at 1.9) originating from near the accretion disk, and a hard, bright jet component with a flat X-ray slope (i.e. $\\Gamma=1.2-1.3$ as before). We modified the Seyfert-like power-law with a continuum reflection component (assuming R=1) and also added a narrow iron K line of equivalent width 150~eV. Although poorly constrained, we obtained a good fit to the data when the Seyfert-like component was typically 4-5 times weaker than the flat jet component. Thus in PKS~0537$-$286 we may be observing two X-ray components; the steep Seyfert-like emission from near the disk, responsible for the iron K line, and the hard component from the jet, which considerably dilutes the disk reflection and line features.  This provides a possible explanation for the apparent weakness of the iron K line and Compton reflection hump, when compared to other observations of Seyfert galaxies and quasars. If the relativistic jet is orientated close to the line-of-sight, as expected in unification schemes for quasars (Antonucci 1993), then any disk reflection features can be diluted due to Doppler boosting of the featureless X-ray continuum, through the bulk motion of the jet. This is consistent with other observations of X-ray luminous, predominantly radio-loud quasars, where the strength of the iron line is diminished (the so-called `X-ray Baldwin effect' - Iwasawa \\& Taniguchi 1993). For PKS~0537$-$286, at a redshift of $z=3.104$, any iron K emission features are redshifted into a more favourable part of the EPIC bandpass, where the effective area is near its maximum. {\\it This has enabled us to detect a weak iron K line in PKS~0537$-$286; indeed at $z=3.104$ this is the most distant iron K emission line detected to date.} The limited statistics do not allow the line width or shape to be constrained. However, the median energy of 6.15~keV implies reprocessing in `cold' matter close to the central black hole or from infalling material at $v\\sim13000$~km~s$^{-1}$. Clearly, more detailed studies with \\xmm\\ of other high redshift quasars, both radio-loud and radio-quiet, are required to explore this.  Past studies with \\rosat\\ and \\asca\\ have showed that several high redshift radio-loud quasars appear to show intrinsic X-ray absorption, above that from our own Galaxy (see Bechtold \\et 1994, Elvis \\et 1994b, Reeves \\et 1997, Cappi \\et 1997, Fiore \\et 1998). Correlations involving samples of quasars (Reeves \\et 1997, Reeves \\& Turner 2000) have showed that this apparent excess absorption is correlated with object redshift, with many high z quasars requiring substantial absorption columns (up to $10^{22}$~cm$^{-2}$), whilst low z quasars remain unabsorbed. Recently even, evidence has been presented for substantial soft X-ray absorption towards one blazar-like object (GB~1428+4217) at $z=4.71$ (Boller \\et 2000). At present the nature of this absorption is unclear, but it could be related to the dense Galactic-scale environments proposed in the early evolutionary stages of quasars (see Fabian 1999). Alternatively in some objects the X-ray absorption may originate from intervening damped Lyman-$\\alpha$ systems, which are thought to be associated with the disks of young protogalaxies (e.g. Wolfe \\et 1994).  From the published optical (rest frame UV) spectrum, it is known that PKS~0537$-$286 has substantial UV absorption, with at least two different Lyman-limit systems observed close to the known redshift of the quasar (Wright \\et 1978, Osmer \\et 1994). The velocities of these systems are $\\Delta\\nu=4200$ and 40000~km~s$^{-1}$, {\\it blueshifted} with respect to the quasar. The latter system has a very deep Lyman-continuum cut-off at 912\\.{A} (rest frame), where the quasar continuum flux drops close to zero. So it perhaps seems surprising at first that in this \\xmm\\ observation we detect no intrinsic X-ray absorption towards PKS~0537$-$286. However in order to measure appreciable line-of-sight X-ray absorption (of the order $N_{H}\\sim10^{22}$~cm$^{-2}$), one would have to encounter a damped Lyman-$\\alpha$ system. We can now rule this scenario out in PKS~0537$-$286, and note the chance coincidence of such a system out to $z=3$ is $\\sim$20-30\\% (e.g. Steidel 1992).  The \\xmm\\ observation is in contrast to some earlier \\asca\\ and \\rosat\\ observations of PKS~0537$-$286, where a low energy X-ray cut-off had been claimed (Reeves \\& Turner 1997, Cappi \\et 1997). One possibility is that the absorption is variable; we note the `warm' absorbers observed in low z AGN are known to vary (e.g. MCG-6-30-15, Otani \\et 1996; NGC 3227, George \\et 1998). It is also possible that if the absorption is associated with a high velocity system, the absorbing material may move out of the line-of-sight. Finally, the photon statistics on the earlier \\asca\\ and \\rosat\\ observations were quite poor, so it is possible that the earlier detection was spurious. We conclude that it will be particularly important to utilise future \\xmm\\ observations to estimate the frequency of these `absorbed quasars', as a probe of some of the most distant matter in the Universe, and to determine the environments of some of the earliest quasars."
    },
    "0010/astro-ph0010151_arXiv.txt": {
        "abstract": "}[2]{{\\footnotesize\\begin{center}ABSTRACT\\end{center} \\vspace{1mm}\\par#1\\par \\noindent {~}{\\it #2}}}  \\newcommand{\\TabCap}[2]{\\begin{center}\\parbox[t]{#1}{\\begin{center} \\small {\\spaceskip 2pt plus 1pt minus 1pt T a b l e} \\refstepcounter{table}\\thetable \\\\[2mm] \\footnotesize #2 \\end{center}}\\end{center}}  \\newcommand{\\TableSep}[2]{\\begin{table}[p]\\vspace{#1} \\TabCap{#2}\\end{table}}  \\newcommand{\\FigCap}[1]{\\footnotesize\\par\\noindent Fig.\\  % \\refstepcounter{figure}\\thefigure. #1\\par}  \\newcommand{\\TableFont}{\\footnotesize} \\newcommand{\\TableFontIt}{\\ttit} \\newcommand{\\SetTableFont}[1]{\\renewcommand{\\TableFont}{#1}}  \\newcommand{\\MakeTable}[4]{\\begin{table}[htb]\\TabCap{#2}{#3} \\begin{center} \\TableFont \\begin{tabular}{#1} #4 \\end{tabular}\\end{center}\\end{table}}  \\newcommand{\\MakeTableSep}[4]{\\begin{table}[p]\\TabCap{#2}{#3} \\begin{center} \\TableFont \\begin{tabular}{#1} #4 \\end{tabular}\\end{center}\\end{table}}  \\newenvironment{references}% { \\footnotesize \\frenchspacing \\renewcommand{\\thesection}{} \\renewcommand{\\in}{{\\rm in }} \\renewcommand{\\AA}{Astron.\\ Astrophys.} \\newcommand{\\AAS}{Astron.~Astrophys.~Suppl.~Ser.} \\newcommand{\\ApJ}{Astrophys.\\ J.} \\newcommand{\\ApJS}{Astrophys.\\ J.~Suppl.~Ser.} \\newcommand{\\ApJL}{Astrophys.\\ J.~Letters} \\newcommand{\\AJ}{Astron.\\ J.} \\newcommand{\\IBVS}{IBVS} \\newcommand{\\PASP}{P.A.S.P.} \\newcommand{\\Acta}{Acta Astron.} \\newcommand{\\MNRAS}{MNRAS} \\renewcommand{\\and}{{\\rm and }} {{\\it BVI} photometry of the Magellanic Clouds collected during the OGLE-II microlensing experiment makes it possible to study in detail photometric properties of the \"major\" stellar distance indicators in the Magellanic Clouds. In addition to Cepheids, RR~Lyr and red clump stars, which photometry was presented in the earlier OGLE papers, we present the so far most accurate determination of brightness of the tip of the red giant branch in the LMC and SMC.  We analyze the ratios of brightness of all four distance indicators in the LMC and SMC. Additionally, we include in our analysis, when possible, photometric data of the distance indicators in the metal poor Carina dwarf galaxy for which photometry was also collected during the OGLE-II experiment. The analysis is largely differential, free from zero point and extinction uncertainties.  The main conclusion is that the distance scales resulting from all four stellar distance indicators are fully consistent with each other. Therefore the distance scale problem is not a problem resulting from different distances yielded by distance indicators, but rather a problem of the proper zero point of this common distance scale. All four stellar distance indicators have to be treated as an {\\it ensemble}, and any determination of the zero point for one of them must predict reasonable luminosities of the others. This puts strong constraints on many proposed calibrations of distance indicators. For example, very bright calibrations of Cepheids or RR~Lyr can practically be ruled out.  At present, the most likely calibration of the zero point of the common distance scale, which would be consistent with observations of all four distance indicators, is that resulting from faint calibration of RR~Lyr stars or calibration of the red clump stars. With the OGLE-II photometry it leads to the distance moduli of ${(m-M)_{\\rm LMC}=18.24}$~mag, ${(m-M)_{\\rm SMC}= 18.75}$ mag and ${(m-M)_{\\rm CAR}=19.94}$~mag for the LMC, SMC and Carina dwarf galaxy, respectively. The systematic uncertainty is of the order of 0.07~mag, while the standard deviation of four determinations of only 0.02~mag.  We also analyze the difference of distance moduli between the SMC and LMC. The average value is ${\\Delta(m-M)_{\\rm SMC-LMC}=0.50\\pm0.03}$~mag from four independent measurements. This very good agreement allows us to draw conclusions on the interstellar extinction in the Magellanic Clouds. Our photometric data also provide constraints on the properties of red clump stars.  Finally, the differential comparison of brightness of Cepheids with brightness of other distance indicators in the Magellanic Clouds, and preliminary observations of Cepheids in the IC1613 galaxy indicate no dependence of the zero point of Cepheid Period--Luminosity relation on metallicity.}{Magellanic Clouds -- Galaxies: distances and redshifts -- distance scale} \\vspace*{12pt} ",
        "introduction": "The distance to the Large Magellanic Cloud is one of the most important distances in astrophysics because it sets the zero point of the extragalactic distance scale. Both Magellanic Clouds harbor large populations of \"major\" stellar distance indicators, namely Cepheid and RR~Lyr variable stars, red clump giants and red giants reaching the tip of the red giant branch (TRGB). Therefore they are an ideal place for testing their properties and calibrating their brightness. Unfortunately, the distance to the LMC has been a subject of controversy for a long time. After release of the Hipparcos catalog the situation became even more unclear. Large number of papers appeared in literature each claiming determination of the distance to the LMC with high precision and accuracy (see for example the review of Gibson 2000). The range of determined distance moduli, ${(m-M)_{\\rm LMC}}$, increased to 18.1--18.7~mag. What worse, even determinations for the same distance indicator considerably differed \\eg for Cepheids from  18.3~mag (Luri \\etal 1998) to 18.7~mag (Feast and Catchpole 1997). However, the general feeling is that the Cepheid variables and TRGB stars yield longer distance moduli of ${(m-M)_{\\rm LMC}\\approx 18.5}$~mag or more (\"long\" distance scale) while RR~Lyr and the best calibrated red clump stars give ${(m-M)_{\\rm LMC}\\approx18.3}$~mag (\"short\" distance scale). Nevertheless, the opposite conclusions can also be found in literature. The distance modulus to the LMC seemed to diverge rather than converge to one well established value.  Part of this ambiguity could be attributed to the poor photometric coverage of the Magellanic Clouds in the past. Fortunately, the situation in this field considerably changed when microlensing survey programs began regular photometric monitoring of these galaxies. For example the OGLE-II microlensing project covered photometrically large areas of both Magellanic Clouds in the standard {\\it BVI}-bands (Udalski \\etal 1998b, Udalski \\etal 2000) very suitable for studying stellar population properties in these galaxies.  In this paper we attempt to clear up the problem of the distance scale resulting from stellar distance indicators. We summarize photometric properties of four \"major\" stellar standard candles observed in the Magellanic Clouds and additionally in the Carina dwarf galaxy. The basic properties of Cepheids, RR~Lyr and red clump stars were already presented in the earlier OGLE papers. Here we complete the sample of \"major\" distance indicators by presenting determination of brightness of TRGB stars.  Extensive and homogeneous photometric material allows us to study the photometric properties of all four distance indicators simultaneously. We present a differential analysis --  comparison of ratios of brightness of all combinations of standard candles in each object. It is free from extinction and other systematic uncertainties. Similar approach for selected standard candles was already presented by Udalski (1998a), reanalyzed later by Popowski (2000), and Bersier (2000). Our study clearly indicates that the distance scales resulting from all four standard candles are remarkably consistent. We constrain the possible zero point of this common distance scale and derive the most likely calibrations.  We also perform additional test by calculating the difference of distance moduli between the SMC and LMC. We find that all determinations are in a very good agreement with the standard deviation as low as ${\\pm0.03}$~mag from four independent measurements, confirming again consistency of distance scales of all four distance indicators. Based on this result we may further constrain some photometric properties of the stellar standard candles.  \\vspace*{12pt} ",
        "conclusions": "Photometry of the Magellanic Clouds collected during the OGLE-II microlensing project provides a unique opportunity to study all four \"major\" stellar distance indicators in different environments and to analyze their properties as standard candles with the same homogeneous observational dataset (Table~1). So far the vast majority of comparisons of different distance indicators were based on the final distance determinations. However, such comparisons are usually meaningless because the observations were usually collected in different regions of the Magellanic Clouds, by different observers  and different interstellar reddening corrections were used. The number of possible systematic errors is large and difficult to estimate. With the huge OGLE-II photometric dataset we are in position to analyze for the first time all four distance indicators simultaneously. \\MakeTable{cccc}{12.5cm}{Ratio of brightness of \"major\" stellar standard candles} {\\hline \\noalign{\\vskip4pt} & CARINA & SMC & LMC\\\\ \\hline \\noalign{\\vskip4pt} RR~LYR -- CEPHEIDS & & & \\\\ \\noalign{\\vskip3pt} $ \\langle V_0^{\\rm RR}\\rangle_{{\\rm [Fe/H]}_{\\rm LMC}}-\\langle V_0^{\\rm C}\\rangle$ & -- & $4.58\\pm0.04$ & $4.62\\pm0.03$ \\\\ \\noalign{\\vskip2pt} $ \\langle V_0^{\\rm RR}\\rangle_{{\\rm [Fe/H]}_{\\rm LMC}}-\\langle W_I^{\\rm C}\\rangle$ & -- & $6.36\\pm0.04$ & $6.34\\pm0.03$ \\\\ \\noalign{\\vskip4pt} \\hline \\noalign{\\vskip4pt} TRGB -- CEPHEIDS & & & \\\\ \\noalign{\\vskip3pt} $ \\langle I_0^{\\rm TRGB}\\rangle-\\langle I_0^{\\rm C}\\rangle$ & -- & $0.66\\pm0.04$ & $0.71\\pm0.03$ \\\\ \\noalign{\\vskip2pt} $ \\langle I_0^{\\rm TRGB}\\rangle-\\langle W_I^{\\rm C}\\rangle$ & -- & $1.75\\pm0.04$ & $1.76\\pm0.04$ \\\\ \\noalign{\\vskip4pt} \\hline \\noalign{\\vskip4pt} RED CLUMP -- CEPHEIDS & & & \\\\ \\noalign{\\vskip3pt} $ \\langle I_0^{\\rm RC}\\rangle_{{\\rm [Fe/H]}_{\\rm LMC}}-\\langle I_0^{\\rm C}\\rangle$ & -- & $4.25\\pm0.04$ & $4.35\\pm0.03$ \\\\ \\noalign{\\vskip3pt} $ \\langle I_0^{\\rm RC}\\rangle_{{\\rm [Fe/H]}_{\\rm LMC}}-\\langle W_I^{\\rm C}\\rangle$ & -- & $5.34\\pm0.05$ & $5.40\\pm0.04$ \\\\ \\noalign{\\vskip4pt} \\hline \\noalign{\\vskip4pt} RED CLUMP -- TRGB & & & \\\\ \\noalign{\\vskip4pt} $ \\langle I_0^{\\rm RC}\\rangle_{{\\rm [Fe/H]}_{\\rm LMC}}-\\langle I_0^{\\rm TRGB}\\rangle$ & $3.64\\pm0.08$ & $3.59\\pm0.04$ & $3.64\\pm0.03$ \\\\ \\noalign{\\vskip4pt} \\hline \\noalign{\\vskip4pt} RR~LYR -- TRGB & & & \\\\ \\noalign{\\vskip3pt} $ \\langle V_0^{\\rm RR}\\rangle_{{\\rm [Fe/H]}_{\\rm LMC}}-\\langle I_0^{\\rm TRGB}\\rangle$ & $4.58\\pm0.07$ & $4.61\\pm0.04$ & $4.58\\pm0.04$ \\\\ \\noalign{\\vskip4pt} \\hline \\noalign{\\vskip4pt} RR~LYR -- RED CLUMP & & & \\\\ \\noalign{\\vskip3pt} $ \\langle V_0^{\\rm RR}\\rangle_{{\\rm [Fe/H]}_{\\rm LMC}}-\\langle I_0^{\\rm RC}\\rangle_{{\\rm [Fe/H]}_{\\rm LMC}}$ & $0.94\\pm0.07$ & $1.02\\pm0.05$ & $0.94\\pm0.03$ \\\\ \\noalign{\\vskip4pt} \\hline}  \\Subsection{Ratio of Brightness of Distance Indicators} To compare brightness of \"major\" distance indicators in the Magellanic Clouds and Carina dwarf galaxy we calculated the ratios of brightness of all possible pairs of distance indicators. It should be stressed that our analysis, as long as the brightness in the same passbands is compared, is fully differential as it is performed on the same dataset. Therefore all systematic errors, like uncertainty of the zero point of photometry, uncertainty of interstellar extinction (as long as different standard candles are equally reddened -- see Section~4.5) etc., cancel.  In Table~2 we list the ratios of brightness (differences of magnitudes) of all possible pairs of four \"major\" stellar standard candles. For RR~Lyr and red clump stars we used the magnitudes corrected for metallicity effects, \\ie lumino\\-sities which the stars would have if they were of the LMC metallicity. For Cepheid comparisons we list the ratios for {\\it VI}-bands and additionally for extinction insensitive index $W_I$ (Udalski \\etal 1999a). In the latter case and in all inter-color comparisons the errors are somewhat larger because these comparisons are not fully differential and the zero point uncertainties must be taken into account.  Figs.~7 and 8 present visualization of the data contained in Table~2. Fig.~7 shows the differences of magnitudes between the mean brightness of Cepheids and other standard candles in the LMC and SMC (unfortunately the Carina dwarf galaxy does not host Population~I Cepheids) while Fig.~8 -- for the remaining combinations of standard candles. The data for the Carina dwarf galaxy are included in Fig.~8. The abscissa in Fig.~8 represents schematically metallicity of the environment. For all populations of stars in these three galaxies the order of increasing metallicities is Carina $\\rightarrow$ SMC $\\rightarrow$ LMC and the difference of metallicity between the Carina dwarf galaxy and SMC is about three times larger than between the SMC and LMC. \\begin{figure}[htb] \\centerline{\\includegraphics[bb=120 315 510 710, width=10.5cm]{fig7.ps}} \\vskip3pt \\FigCap{Differences of magnitudes of Cepheids and other distance indicators.} \\end{figure} \\begin{figure}[htb] \\centerline{\\includegraphics[bb=125 370 505 710, width=10.5cm]{fig8.ps}} \\vskip3pt \\FigCap{Differences of magnitudes of RR Lyr, red clump and TRGB stars.} \\end{figure}  If the standard candles were indeed good brightness references then one would expect that their brightness differences should be constant in all considered environments, \\ie their magnitude differences should be parallel to the abscissa axes in Figs.~7 and 8. Figs.~7 and 8 clearly indicate that this is indeed the case (dashed line). All combinations of distance indicators present basically the same picture, \\ie their differences of magnitude in different objects are constant to ${\\pm0.05}$~mag in the worst case.  Looking carefully at Fig.~7 or the Cepheid data in Table~2 one can notice a small systematic trend -- the differences of magnitude between the distance indicators and Cepheids seem to be systematically slightly larger in the LMC than in the SMC. This effect could result from a small dependence of Cepheid brightness on metallicity. Therefore to check if the metallicity effect can be real we observed Cepheids in the metal poor galaxy IC1613 (metallicity of IC1613 is smaller than that of the SMC; metallicity of Cepheids in the LMC and SMC is equal to ${\\rm [Fe/H]=-0.3}$~dex and ${-0.7}$~dex respectively, Luck \\etal 1998). Although this project is still in progress, preliminary analysis of the data collected so far indicates that the difference of the mean {\\it I}-band brightness of Cepheids with 10~day period and TRGB stars in the IC1613 galaxy is equal to ${\\langle M^{\\rm TRGB}_I\\rangle-\\langle M^{\\rm C}_I \\rangle=0.75\\pm0.06}$~mag. Asterisk in the appropriate panel of Fig.~7 marks that value and the horizontal location of points corresponds to the mean metallicity of objects. This result confirms that the brightness of Cepheids does not depend on metallicity to a few hundredths of magnitude in the wide metallicity range of ${\\approx-1<{\\rm[Fe/H]}<-0.3}$~dex. More detailed analysis of the IC1613 data will be presented when this sub-project of the OGLE search is finished.  The conclusion from our exercise is very straightforward. All four \"major\" distance indicators, namely Cepheids, RR~Lyr, red clump and TRGB stars are equally good standard candles. It is worth noting that these standard candles represent very different populations of stars from the oldest RR~Lyr stars, through somewhat younger TRGB population, intermediate age red clump stars and finally young population of classical Cepheids. The distance scales from all four distance indicators are fully compatible with each other, and all four distance indicators can be equally well used for distance determinations within the domain they are considered to be a standard candle. Fig.~8 strongly suggests that the small corrections for difference of metallicity of RR~Lyr and red clump stars (Eqs.~7 and 8) we applied are correct. Otherwise systematic deviations would be observed. It is worth noticing that the consistency of distances determined from the red clump and TRGB stars was already noted by Bersier (2000).  \\vskip15pt \\Subsection{Constraints on the Zero Point} \\vskip6pt  As we showed in Section~4.1 all four distance indicators are fully compatible with each other and, therefore, the distance scales determined from them are fully consistent. Thus, the only important problem related to the distance determination is a proper calibration of the zero point of this common distance scale. Because the brightness ratios of all four indicators are well constrained based on our differential analysis, all four indicators must be treated as an {\\it ensemble}. The calibration of any of them sets the calibrations of all the remaining indicators. In other words any proposed calibration of any of the \"major\" stellar distance indicators must predict reasonable calibrations for the remaining ones. In this way we can put constraints on the possible calibration of the zero point of the common distance scale.  First, let us briefly discuss proposed calibrations of stellar distance indicators. The Cepheid calibrations fall into three categories: bright, medium and faint. Bright calibrations predict ${M_V^{\\rm C}\\approx-4.2}$~mag for a Cepheid with the period of ${P=10}$ days (Feast and Catchpole 1997, Lanoix, Paturel and Garnier 1999, Groenewegen and Oudmaijer 2000) and are based on analyses of Hipparcos direct parallaxes of these stars. One should, however, remember that the parallaxes of Cepheids measured by Hipparcos are very small and have very large errors (only a few Cepheids has the relative error smaller than 30\\%). Medium brightness calibrations give ${M_V^{\\rm C}\\approx-4.05}$~mag and are mostly based on Barnes--Evans method (Gieren \\etal 1997) and pre-Hipparcos Galactic calibrations (Laney and Stobie 1994). Faint calibrations predict ${M_V^{\\rm C}\\approx-3.85}$~mag based on a kind of statistical parallaxes method (Luri \\etal 1998). Indirect calibration {\\it via} Cepheids in the NGC4258 galaxy also indicates faint brightness of Cepheids (Maoz \\etal 2000).  Calibrations of RR~Lyr stars in the post Hipparcos era were reviewed and summarized by Popowski and Gould (1998). This review concludes that the most likely absolute magnitude of RR~Lyr stars is ${M_V^{\\rm RR}\\approx0.70}$~mag at metallicity of ${\\rm [Fe/H]=-1.6}$~dex. However, although the majority of empirical determinations seem to point to fainter calibrations of RR~Lyr, subdwarf fitting technique results in much higher luminosity of RR~Lyr stars, namely ${M_V^{\\rm RR}=0.45}$~mag at ${\\rm [Fe/H]=-1.6}$~dex (Carretta \\etal 2000). The drawbacks of this method when applied to RR~Lyr calibration are described in Popowski and Gould (1998). Nevertheless, this calibration is still a possible alternative to the faint RR~Lyr calibration.  Two kinds of calibrations are used for distance determination with TRGB stars. The empirical calibration based on Galactic globular clusters data (Lee \\etal 1993a) gives ${M_I^{\\rm TRGB}=-4.0\\pm0.1}$~mag for metallicities ${\\rm [Fe/H]<- 0.7}$~dex. It is, however, based on the old RR~Lyr stars calibration predicting ${M_V^{\\rm RR}=0.55}$~mag at ${\\rm [Fe/H]=-1.6}$~dex. This is by about 0.1--0.15~mag brighter than the present, most likely calibration of RR~Lyr stars (Popowski and Gould 1998). Therefore, the correction of this order should be taken into account when deriving the distance with TRGB what has never been done. The second calibration is based on the theoretical modeling of TRGB stars and predicts TRGB magnitude brighter by about 0.1~mag than the empirical one (Salaris and Cassisi 1997). However, because the uncertainties related to the modeling are still rather large (Dominguez \\etal 1999, Castellani \\etal 2000b) we do not consider this calibration as a very strong alternative to the empirical one.  Finally, the calibration of red clump stars is superior to calibration of any other standard candle what is one of the main advantages of this distance indicator. Because these stars are very numerous in the solar neighborhood, the Hipparcos sample of red clump stars with good accuracy parallaxes (relative error less than 10\\%) and precise photometry is very large (several hundred objects, Paczy{\\'n}ski and Stanek 1998). This made it possible to determine the absolute magnitude of red clump stars with high accuracy: ${M_I^{\\rm RC}=- 0.23\\pm0.03}$~mag for metallicity of ${\\rm [Fe/H]=0.0}$~dex (Stanek and Garnavich 1999, Udalski 2000). Unfortunately, the local red clump stars only partially overlap in metallicity with the LMC red clump stars. However, the dependence on metallicity (Eq.~8) is rather modest so the mean {\\it I}-band magnitude of the red clump stars in the LMC is ${M_I^{\\rm RC}=-0.31\\pm 0.04}$~mag. No significant dependence of the red clump magnitude on age was assumed here what, among others, is clearly supported by Fig.~8 (Section~4.6). On the other hand, theoretical modeling of red clump stars indicates that its magnitude depends on both metallicity and age. Girardi and Salaris (2000) claim that the combined effect may reach as much as 0.20~mag in the LMC case. If true, the LMC red clump magnitude would be of ${M_I^{\\rm RC}=-0.43}$~mag.  Having in mind all these calibrations, less and more likely, we may constrain the possible zero point of the common distance scale. First, we may almost immediately exclude the bright calibration of Cepheids. For example, ${M_V^{\\rm C}=-4.21}$~mag and ${M_I^{\\rm C}=-4.94}$~mag (Groenewegen and Oudmaijer 2000) for the Galactic Ce\\-pheids would indicate the following magnitudes of the remaining distance indicators: $M_V^{\\rm RR}=0.39$~mag, ${M_I^{\\rm TRGB}=-4.23}$~mag and ${M_I^{\\rm RC}=-0.64}$ mag. The mean magnitude of RR~Lyr stars would have to be brighter than the brightest calibration from subdwarf fitting. The TRGB magnitude would be much brighter than indicated by Lee \\etal (1993a) calibration but this could be marginally consistent when correction to the TRGB calibration resulting from the appropriate RR~Lyr calibration (${\\approx0.15}$~mag brighter) was applied. However, the predicted absolute magnitude of the red clump stars practically rules out the bright Cepheid calibration. The mean magnitude of the red clump stars of the LMC metallicity would have to be by about 0.35~mag brighter than the empirical calibration or 0.2~mag brighter than the theoretical predictions. This large discrepancy could only vanish when the brightness of Cepheids would very strongly depend on metallicity (about 1.0~mag/dex in the {\\it I}-band) making the Galactic Cepheids much brighter. As we have shown in Section~4.1 our data do not indicate any significant dependence of Cepheid brightness on metallicity. Therefore this calibration can be ruled out with high confidence.  Similar arguments can exclude the bright calibration of RR~Lyr stars. $M_V^{\\rm RR}=0.45$~mag would indicate ${M_V^{\\rm C}=-4.15}$~mag, ${M_I^{\\rm TRGB}=-4.14}$~mag and $M_I^{\\rm RC}=-0.52$~mag. While Cepheids and TRGB calibrations could be consistent with the bright RR~Lyr calibration, the inferred magnitude of red clump stars is again too bright by at least 0.1~mag from theoretical predictions or more than 0.2~mag from the empirical calibration. Because we have already corrected RR~Lyr stars magnitude for metallicity it would be extremely difficult to find additional 0.1--0.2~mag that could explain this inconsistency. Therefore the bright calibration of RR~Lyr also seems to be unlikely.  On the other hand two of the already  discussed calibrations of the distance indicators presented here, namely Hipparcos calibration of red clump stars and faint calibration of RR~Lyr, are consistent with all four standard candles. The latter calibration (Popowski and Gould 1998), ${M_V^{\\rm RR}=0.70}$~mag, would indicate: $M_V^{\\rm C}=-3.90$~mag, ${M_I^{\\rm TRGB}=-3.89}$~mag and ${M_I^{\\rm RC}=-0.27}$~mag. All these predictions are in good agreement with the faint calibration of Cepheids (statistical parallaxes and Cepheids in NGC4258), Lee \\etal (1993a) calibration of TRGB (corrected down by 0.15 mag due to fainter RR~Lyr brightness scale) and the empirical (Hipparcos) calibration of red clump stars (Udalski 2000), respectively. The empirical calibration of red clump stars of the LMC metallicity, ${M_I^{\\rm RC}=-0.31}$~mag (Udalski 2000), leads to very similar predictions: ${M_V^{\\rm RR}=0.66}$~mag, ${M_I^{\\rm C}=- 4.61}$~mag and $M_I^{\\rm TRGB}=-3.93$~mag.  Very consistent results obtained with calibrations of RR~Lyr and red clump stars strongly suggest that in the entire range of possible calibrations those two are the most likely. If we adopt the mean results based on the RR~Lyr and red clump empirical calibrations and our differential analysis of \"major\" stellar standard candles then the following formulae for individual standard candles can be written:\\\\ Cepheids: \\setcounter{equation}{8} \\begin{eqnarray} M_V=(-2.775\\pm0.031)\\times(\\log P-1) - 3.92\\pm0.05\\\\ M_I=(-2.977\\pm0.021)\\times(\\log P-1) - 4.61\\pm0.05\\\\ M_{W_I}=(-3.300\\pm0.011)\\times(\\log P-1) - 5.67\\pm0.05 \\end{eqnarray} TRGB: $$M_I^{\\rm TRGB}=-3.91\\pm0.05~{\\rm mag}\\qquad({\\rm [Fe/H]}<-0.7~{\\rm dex})\\eqno{(12)}$$ RR~Lyr: $$M_V^{\\rm RR}=(0.18\\pm0.04)\\times({\\rm [Fe/H]}+1.6)+0.68\\pm0.05 \\eqno{(13)}$$ Red clump stars: $$M_I^{\\rm RC}=(0.14\\pm0.04)\\times({\\rm [Fe/H]}+0.5)-0.29\\pm0.05 \\eqno{(14)}$$ where all zero points refer to the LMC metallicity and their errors are estimated from uncertainty of the Hipparcos calibration of red clump stars ($\\pm0.04$~mag) and the consistency of both calibrations. It should be, however, remembered that our differential test cannot exclude somewhat brighter calibrations, \\ie up to $M_V^{\\rm RR}\\approx0.55$~mag for RR Lyr and ${M_I^{\\rm C}\\approx-4.7}$~mag for Cepheids. They could be consistent at least with theoretical predictions for red clump brightness and thus cannot be fully ruled out. These limits go by about 0.1~mag fainter (\\ie to Eqs.~9--14) if the empirical calibration of the red clump is considered.  \\Subsection{Distance to the LMC, SMC and Carina Dwarf Galaxy} Eqs.~(9)--(14) can be directly applied to the dereddened magnitudes of standard candles provided in Table~1 to determine the distances to the analyzed galaxies. In this case, however, the analysis is no longer differential and any systematic errors can affect the results. While the zero points of the OGLE photometry were tested many times (\\eg Udalski \\etal 1998ab, Udalski \\etal 2000) and it is unlikely that the systematic uncertainty from this factor is larger than 0.02~mag, the main source of uncertainty remains the interstellar extinction applied to the photometry to deredden the data. Accuracy of the zero points of extinction maps used by OGLE was discussed in Udalski \\etal (1999bc). Also as we will show in Section~4.5 there is no indication that the interstellar reddening is different for younger and older populations as suggested by Zaritsky (1999). The mean reddening of ${E(B-V)=0.143}$~mag and 0.087~mag in the bar of the LMC and SMC, respectively, seems to be consistent with all determinations in these regions with hot young or older stars (including the recent determination from RR~Lyr instability strip colors by Clementini \\etal 2000, when correct boundaries of the strip, \\ie bluer by about 0.05~mag, \\eg from Walker 1998, are used). Therefore, it is rather unlikely that the uncertainty of $E(B-V)$ reddening is larger than 0.02~mag.  Distances to the LMC, SMC and Carina dwarf galaxy calculated with Eqs.~(9) --(14) are listed in Table~3. As it can be expected all distances are very similar. This is not surprising as we already showed that all four distance scales are fully consistent. The mean values of distance moduli from four \"major\" stellar standard candles are equal to ${(m-M)_{\\rm LMC}=18.24}$~mag, ${(m-M)_{\\rm SMC}=18.75}$~mag and ${(m-M)_{\\rm CAR}=19.94}$~mag. The systematic error is of the order of 0.07~mag and results mostly from the reddening and calibration (Eqs.~9--14) uncertainties while the standard deviation of these measurements is only of 0.02~mag. \\renewcommand{\\arraystretch}{0.9} \\MakeTable{cccc}{12.5cm}{Distance moduli to the Carina dwarf galaxy, SMC and LMC} {\\hline \\noalign{\\vskip3pt} STANDARD CANDLE & CARINA & SMC & LMC\\\\ \\hline \\noalign{\\vskip5pt} CEPHEIDS (mean {\\it V,I},$W_I$) & -- & $18.77\\pm0.07$ & $18.23\\pm0.07$ \\\\ \\noalign{\\vskip3pt} RR~LYR ({\\it V})& $19.93\\pm0.08$ & $18.76\\pm0.07$ & $18.23\\pm0.07$ \\\\ \\noalign{\\vskip3pt} TRGB ({\\it I})& $19.94\\pm0.08$ & $18.74\\pm0.08$ & $18.24\\pm0.07$ \\\\ \\noalign{\\vskip3pt} RED CLUMP ({\\it I}) & $19.96\\pm0.08$ & $18.71\\pm0.07$ & $18.26\\pm0.07$ \\\\ \\noalign{\\vskip3pt} \\hline}  The resulting distance to the LMC is in the range of \"short\" distances, \\ie it is by about 0.25~mag smaller than the classical distance modulus to the LMC of ${(m-M)_{\\rm LMC}=18.5}$~mag. We can summarize that the OGLE photometric data of four \"major\" stellar distance indicators clearly indicate that the most likely zero point of the common distance scale which is consistent with all four standard candles is that which gives the \"short\" distance scale. Although we are not able to completely exclude calibrations up to 0.15~mag brighter, all the remaining ones seem to be inconsistent with at least one of the four distance indicators. It is worth noting at this point that the recent determinations of the distance to the LMC with independent, largely geometric standard candle, -- eclipsing binaries -- also provide strong support for the \"short\" distance scale to the LMC (${(m-M)_{\\rm LMC}\\approx18.3}$~mag, Guinan \\etal 1998, Fitzpatrick \\etal 2000). This is completely independent confirmation that the calibrations presented in this paper are the most likely. Also, as we already mentioned, the \"short\" distance scale to the LMC results from geometric determination of the distance to NGC4258 galaxy (Herrnstein \\etal 1999) and photometry of Cepheids in this object (Maoz \\etal 1999). Nevertheless, the final solution of the zero point problem will probably have to wait until the next space astrometric missions (FAME, SIM, GAIA) are launched and finished and direct good quality parallaxes of large samples of Cepheids, RR~Lyr and TRGB stars are determined.  \\vspace*{-9pt} \\Subsection{Difference of Distance Moduli SMC--LMC} \\vspace*{-5pt} The differences of distance moduli between the SMC and LMC resulting from Cepheids, RR~Lyr, TRGB and red clump stars are listed in Table~4. They were derived using the extinction free luminosities listed in Table~1. \\renewcommand{\\arraystretch}{1} \\MakeTable{cc}{12.5cm}{Difference of distance moduli SMC--LMC} {\\hline \\noalign{\\vskip3pt} STANDARD CANDLE & $(m-M)_{\\rm SMC}-(m-M)_{\\rm LMC}$\\\\ \\hline \\noalign{\\vskip5pt} CEPHEIDS ($W_I$) & $0.51\\pm0.05$ \\\\ \\noalign{\\vskip3pt} RR~LYR ({\\it V})& $0.53\\pm0.07$ \\\\ \\noalign{\\vskip3pt} TRGB ({\\it I}) & $0.50\\pm0.05$ \\\\ \\noalign{\\vskip3pt} RED CLUMP ({\\it I}) & $0.45\\pm0.05$\\\\ \\noalign{\\vskip3pt} \\hline}  It is striking that all independent determinations of ${\\Delta (m-M)_{\\rm SMC- LMC}}$ are very similar. The most reliable one comes from the $W_I$ index for Cepheids, because those magnitudes are in principle extinction free. Therefore by comparison with $W_I$ Cepheid distance we may assess reliability of other distance indicators as well as the reddening scale used for distance determinations.  The numbers from Table~4 can be affected by systematic error resulting from the uncertainty of extinction in both galaxies. It should, however, affect very similarly all determinations, except for $W_I$ index which is extinction independent. Very good agreement of the values from different distance indicators as compared to $W_I$ value indicates, however, that this error is probably very small (${<0.02}$~mag) and the extinction scales in both Magellanic Clouds are correct. The mean difference of distance moduli between the SMC and LMC from all four independent determinations (Cepheids, RR~Lyr, TRGB and red clump) is equal to 0.50~mag with very small standard deviation of only 0.03~mag.  \\Subsection{Interstellar Extinction in the LMC} Very good agreement of $\\Delta (m-M)_{\\rm SMC-LMC}$ from all distance indicators provides strong constraints on many observables. First, we may practically rule out the possibility that the interstellar reddening is significantly different for different populations of stars in the LMC as suggested by Zaritsky (1999). Based on a new method he determined interstellar extinction in the {\\it V}-band equal to ${A_V\\approx0.4}$ mag for hot young stars and only ${A_V\\approx0.1}$~mag for old cool population of stars in a field in the LMC overlapping with our NW field (Section~3.4). The simplest test of reality of Zaritsky (1999) hypothesis is to compare the most sound difference of distance moduli between the SMC and LMC, namely resulting from Cepheid extinction free index $W_I$ and equal to 0.51~mag, with that obtained from old populations under the assumption of very small extinction for these groups of stars. Even if we assume the extinction toward the SMC equal to zero (what is an obvious underestimate), $A_V=0.1$~mag in the LMC and the  mean magnitudes of RR~Lyr, TRGB and red clump stars  indicate that ${\\Delta(m-M)_{\\rm SMC-LMC}}$ from the old population would already be by about 0.1~mag smaller than the extinction free value resulting from Cepheid $W_I$ index. When non-zero reddening in the SMC is applied the gap widens. ${\\Delta(m-M)_{\\rm SMC-LMC}}$ from the old population converges to that from young Cepheids only when the interstellar extinction is the same for both populations and its value is close to that of the OGLE extinction maps.  Another test of the Zaritsky (1999) hypothesis is provided by Fig.~7. Good agreement of differences of magnitudes of Cepheids and old population standard candles in the LMC and SMC under the assumption that the interstellar extinction is the same for all stars in both Magellanic Clouds practically rules out that hypothesis. Even if the extinction for young population were larger than for the old one, as Zaritsky suggests, it would have to be larger by exactly the same amount in the LMC and SMC. That would imply exactly the same properties and distribution of dust in both these galaxies what is extremely unlikely. Moreover, the interstellar extinction in the SMC is much smaller than in the LMC and simply there is no room for so large extinction difference as Zaritsky (1999) claims in the LMC. Only very unlikely conspiracy could mask the differences of magnitudes of Cepheids and old standard candles so they were exactly the same in both Magellanic Clouds. Both tests rather unambiguously indicate that the interstellar reddening, ${E(B-V)}$ in the Magellanic Clouds is to 0.01--0.02~mag similar for young and old populations and results of Zaritsky (1999) are probably an effect of his method of extinction determination for cool stars (because the derived interstellar extinction for hot stars, \\ie those with simple and well known spectra, seems to be reasonable: ${E(B-V)\\approx0.12}$~mag \\vs 0.11~mag from the OGLE maps).  \\vspace*{-9pt} \\Subsection{Properties of the Red Clump Stars} \\vspace*{-5pt} The mean extinction free {\\it I}-band magnitude of red clump stars in the LMC is sometimes a subject of controversy. For example Romaniello \\etal (1999) claim that it is by about 0.15~mag dimmer than the value presented in this paper. They determined ${\\langle I_0^{RC}\\rangle=18.12}$~mag in a field of very high and non-uniform extinction around SN1987A, based on HST photometry. Such a value can, however, be easily ruled out by comparison of differences of distance moduli between the SMC and LMC. Only if the extinction in the SMC were equal to zero, the difference of distance moduli inferred from red clump stars with Romaniello \\etal (1999) value of $\\langle I_0^{RC}\\rangle$ could be consistent with the Cepheid extinction free $W_I$ index value (Table~4). With any non-zero extinction in the SMC the difference of distance moduli of red clump stars would be smaller -- up to 0.2~mag if the SMC extinction presented in this paper were used. This inconsistency makes the Romaniello \\etal (1999) value highly unlikely.  Observational data presented in this paper can also be used for constraining the possible dependence of the mean brightness of the red clump stars on age. Fig.~8, where comparison of the red clump brightness with other distance indicators is presented, clearly shows that the metallicity correction given by Eq.~(8) very well corrects the excess of red clump brightness. There are no systematic trends which could suggest other slope in Eq.~(8) and the residuals from a straight line, parallel to the abscissa axis, are of the order of ${\\pm0.04}$~mag only. If the excess of brightness of red clump stars in lower metallicity objects is indeed caused by metallicity differences then the very small residuals indicate that the red clump brightness is practically independent on age. It should be noted that on average the red clump in the Carina dwarf galaxy is much older than in the Magellanic Clouds.  On the other hand theoretical modeling of red clump stars by Girardi and Salaris (2000) indicates that its magnitude is dependent on both metallicity and age. Girardi and Salaris (2000) constructed synthetic CMDs using their evolutionary models and assumed star formation rate (SFR) and age-metallicity relations (AMR) for the Magellanic Clouds and Carina dwarf galaxy to analyze the mean properties of red clump stars in these objects. They claim that the most likely assumptions produce, due to combined metallicity and age dependence, the red clump brighter by 0.20, 0.29 and 0.29~mag in the LMC, SMC and Carina dwarf galaxy, respectively, than the local red clump stars measured by Hipparcos. It is, however, extremely difficult to assess reality of these results. Qualitative comparison of the best synthetic CMDs with the empirical ones indicates that the simulations are rather far from reality -- in particular the synthetic SMC red clump (Fig.~17, Girardi and Salaris 2000) does not resemble at all very round and featureless red clump observed in this galaxy (Fig.~10; Paczy{\\'n}ski \\etal 1999). Also the LMC bar and Carina dwarf galaxy synthetic red clumps hardly resemble the empirical ones (Udalski \\etal 2000, Udalski 1998a).  The Carina dwarf galaxy simulations are the most clear indication that the modeling is very far from being accurate. Looking at Fig.~1 of Girardi and Salaris (2000) one can find that it is practically impossible to obtain the red clump in Carina dwarf galaxy brighter than 0.1~mag than in the LMC when its age is larger than 3~Gyr, independently of SFR and AMR assumptions. Indeed, Girardi and Salaris (2000) obtain the red clump of only 0.09~mag brighter than the LMC red clump or even fainter if other SFR is assumed. Observations, however, indicate that it is by about 0.2~mag brighter (Fig.~8, Table~1). This is a large discrepancy and it probably gives a realistic information on accuracy of the models and synthetic CMD simulations. Modeling results while reasonably well reproducing qualitative features of stellar populations are not accurate to better than 0.1--0.2~mag (see also Castellani \\etal 2000b and discussion in Paczy{\\'n}ski \\etal 1999).  Another empirical confirmation of very weak and shallow dependence of the mean {\\it I}-band magnitude of the red clump on age within the age range of 2--10~Gyr provides comparison of red clump and TRGB magnitudes. Bersier (2000) presented such a comparison for the Magellanic Clouds, Carina and Fornax dwarf galaxies and found that the magnitude correction for metallicity (Eq.~8) makes these two distance indicators very consistent. His results are fully confirmed in this paper (Fig.~8). As we mentioned, small residuals indicate that the age dependence must be weak. Because of large brightness, the TRGB magnitude is relatively easy to determine in many even quite far located objects (similar brightness Cepheids require at least 20--30 epochs to determine their mean magnitudes). Therefore, we attempted to extend the comparison of the mean red clump and TRGB magnitudes to the Local Group galaxies with populations suitable for such comparisons (low metallicity, older stars for TRGB and intermediate age stars for red clump). Beside the Magellanic Clouds and Carina dwarf galaxy data presented in this paper we found in literature data for nine additional objects whose CMDs include both TRGB and red clump stars. We limited ourselves to objects in which the red clump is reasonably above the limit of photometry so that its mean magnitude is not affected. The data for all objects are listed in Table~5 in order of decreasing metallicity of the object, as well as ${\\langle M^{\\rm RC}_I\\rangle-\\langle M^{\\rm TRGB}_I\\rangle}$ calculated with the red clump magnitude converted to the LMC metallicity (\\ie with correction, LMC$^{\\rm RC}_{\\rm COR}$, listed in the fifth column in Table~5 and obtained from Eq.~8). Unfortunately, in all but one cases (NGC147) we did not have the original data, so that we had to rely on authors' determinations or in a few cases to determine the mean magnitudes from CMD plots. Therefore, the presented exercise is not fully homogeneous. Nevertheless, the magnitudes presented in Table~5 should be accurate to about 0.05~mag. We usually used the same photometry to determine the difference of red clump and TRGB magnitudes to avoid systematic errors. In a few cases, however, we had to use additional source for TRGB magnitude -- the HST fields where photometry reached red clump stars were too small for reliable TRGB magnitude determination. In these cases we first verified whether both photometric data sets were on the same scale.  Differences ${\\langle M^{\\rm RC}_I\\rangle-\\langle M^{\\rm TRGB}_I\\rangle}$, listed in the sixth column of Table~5, indicate that except for two of twelve objects they all fall in the narrow range of $3.62\\pm0.08$~mag. This sample includes objects with younger intermediate age population like LMC and older objects -- Fornax, M32 (${\\approx8.5}$~Gyr) or Leo~II ($\\approx 9$~Gyr). One should note that the Girardi and Salaris (2000) modeling (their Fig.~1) predicts red clump fainter by about by 0.25~mag in the latter cases. Clearly, the dependence on age is very weak.  Two deviating cases in Table~5 can easily be explained. In the case of Leo~A the vast majority of red clump population are the young stars of 0.9--1.5~Gyr age (Tolstoy \\etal 1998). It is not surprising that they are brighter making the difference of red clump and TRGB magnitudes smaller. NGC147 is on the opposite end. The vast majority of its population is older than 10~Gyr, so that it is not surprising that the red clump magnitude drops, making the difference much larger. The NGC147 case is very similar to the NGC121 cluster case in the SMC (Udalski 1998b), also possessing old (${\\approx12}$~Gyr) population and the red clump by about 0.4~mag fainter than that of younger clusters. Results of comparison of red clump and TRGB magnitudes in the nearby Local Group galaxies are in fact very consistent with results of the analysis of clusters in the LMC and SMC (Udalski 1998b) also indicating that in the 2--10~Gyr range the dependence of the red clump brightness on age is very weak.  Concluding, presently we do not find any convincing arguments on reality of significant dependence of the red clump brightness on age, and on the contrary the empirical data presented in this paper rather ~unambiguously~ seem to indicate that such a dependence is marginal for ages within 2--10~Gyr. This confirms our earlier results from analysis of the Magellanic Cloud star clusters (Udalski 1998b). That analysis, in particular of the SMC clusters, was criticized by Girardi and Salaris (2000) because of applying geometric corrections resulting from the spatial extent of the SMC (these corrections result, however, from our current picture of the SMC geometry). Nevertheless, we would like to point attention to the fact that at least two clusters of very different age from the Udalski's (1998b) sample are located very close to each other on the sky (Kron~3 -- 7.5~Gyr and L11 -- 3.5~Gyr) so their photometry can be compared differentially. The mean magnitude of red clump in these objects is similar to $\\pm0.02$~mag after correcting for a small metallicity difference. Girardi and Salaris (2000, Fig.~1) models predict the red clump brighter by about 0.15~mag in younger L11. The brightness is also very similar to the mean magnitude of red clump of field stars indicating that the clusters are located at the same distance as the bulk of SMC stars in that direction. This is the only pair of clusters of different age allowing direct comparison of red clump magnitude. Other comparisons, for example those of Galactic clusters (Girardi and Salaris 2000), suffer from uncertainties of distance much larger than the geometric correction uncertainties in the SMC and very small number statistics of red clump stars.  \\begin{landscape} \\MakeTableSep{ l@{\\hspace{15pt}}c@{\\hspace{15pt}}c@{\\hspace{9pt}}c@{\\hspace{9pt}} r@{\\hspace{9pt}}l@{\\hspace{9pt}}c@{\\hspace{9pt}}c@{\\hspace{9pt}} }{12.5cm}{TRGB and red clump stars in nearby galaxies} {\\hline \\noalign{\\vskip3pt} \\multicolumn{1}{c}{GALAXY}& TRGB & RED CLUMP & [Fe/H] & \\multicolumn{1}{c}{LMC$^{\\rm RC}_{\\rm COR}$}& \\multicolumn{1}{c}{RC--TRGB} & Ref. & Remarks\\\\ & [mag]& [mag] & [dex] & \\multicolumn{1}{c}{[mag]}&\\multicolumn{1}{c}{[mag]} & & \\\\ \\hline \\noalign{\\vskip5pt} M32 & $\\langle I\\rangle=20.75$ & $\\langle I\\rangle=24.35$ & $-0.3$ &$-0.03$~~~~ & ~~~~3.55 & (1) & $\\approx 8.5$~Gyr \\\\ LMC & $\\langle I_0\\rangle=14.32$ & $\\langle I_0\\rangle=17.95$ & $-0.5$ &$0.00$~~~~ & ~~~~3.64 & this paper& \\\\ M31 (G302) Field & $\\langle I\\rangle=20.65$ & $\\langle I\\rangle=24.25$ & $-0.6$ &$0.01$~~~~ & ~~~~3.6 & (2) & \\\\ SMC & $\\langle I_0\\rangle=14.80$ & $\\langle I_0\\rangle=18.34$ & $-1.0$ &$0.07$~~~~ & ~~~~3.59 & this paper& \\\\ Fornax & $\\langle I\\rangle=16.65$ & $\\langle I\\rangle=20.24$ & $-1.0$ &$0.07$~~~~ & ~~~~3.66 & (3) & \\\\ Pegasus& $\\langle I\\rangle=20.95$ & $\\langle I\\rangle=24.50$ & $-1.1$ &$0.08$~~~~ & ~~~~3.65 & (4), (5) & \\\\ Phoenix& $\\langle I\\rangle=19.00$ & $\\langle I\\rangle=22.50$ & $-1.4$ &$0.13$~~~~ & ~~~~3.65 & (6) & \\\\ Leo~II & $\\langle I\\rangle=17.70$ & $\\langle I\\rangle=21.25$ & $-1.6$ &$0.15$~~~~ & ~~~~3.7 & (7), (8) & $\\approx 9$~Gyr\\\\ Carina & $\\langle I_0\\rangle=16.03$ & $\\langle I_0\\rangle=19.46$ & $-2.0$ &$0.21$~~~~ & ~~~~3.64 & this paper, (9)\\\\ Leo~I & $\\langle I\\rangle=18.25$ & $\\langle I\\rangle=21.60$ & $-2.0$ &$0.21$~~~~ & ~~~~3.55 & (10) & \\\\ \\noalign{\\vskip3pt} \\hline \\noalign{\\vskip3pt} Leo~A & $\\langle I\\rangle=20.50$ & $\\langle I\\rangle=23.77$ & $-1.7$ &$0.17$~~~~ & ~~~~3.44 & (11) & $0.9-1.5$~Gyr\\\\ NGC147 Inner Field& $\\langle I\\rangle=20.31$ & $\\langle I\\rangle=24.02$ & $-0.9$ &$0.06$~~~~ & ~~~~3.77 & (12) & $>10$~Gyr \\\\ NGC147 Outer Field& $\\langle I\\rangle=20.31$ & $\\langle I\\rangle=24.06$ & $-1.0$ &$0.07$~~~~ & ~~~~3.82 & (12) & $>10$~Gyr \\\\ \\noalign{\\vskip3pt} \\hline \\noalign{\\vskip3pt} References:&&&&&&&\\\\ \\multicolumn{8}{l}{(1) Grillmair \\etal (1996), (2) Holland, Fahlman and Richer (1996), (3) Bersier (2000),}\\\\ \\multicolumn{8}{l}{(4) Aparicio (1994), (5) Gallagher \\etal (1998), (6) Mart{\\'{\\i}}nez-Delgado \\etal (1998),}\\\\ \\multicolumn{8}{l}{(7) Lee (1995), (8) Mighell and Rich (1996), (9) Smecker-Hane \\etal (1994),}\\\\ \\multicolumn{8}{l}{(10) Lee \\etal (1993b), (11) Tolstoy \\etal (1998), (12) Han \\etal (1997).}\\\\ } \\end{landscape}   \\Acknow{We would like to thank Drs.\\ B.\\ Paczy\\'nski, K.Z.\\ Stanek, and M.~Szyma{\\'n}ski for comments and remarks on the paper. The paper was partly supported by the Polish KBN grant 2P03D00814 to A.\\ Udalski and 2P03D00916 to M.\\ Szyma{\\'n}ski. Partial support for the OGLE project was provided with the NSF grant AST-9820314 to B.~Paczy\\'nski.}"
    },
    "0010/astro-ph0010198_arXiv.txt": {
        "abstract": "This paper predicts self--consistent faint galaxy counts from the UV to the submm wavelength range. The {\\sc stardust} spectral energy distributions described in Devriendt et al. \\cite*{DGS99} (Paper I) are embedded within the explicit cosmological framework of a simple semi--analytic model of galaxy formation and evolution. We begin with a description of the non--dissipative and dissipative collapses of primordial perturbations, and plug in standard recipes for star formation, stellar evolution and feedback. We also model the absorption of starlight by dust and its re--processing in the IR and submm. We then build a class of models which capture the luminosity budget of the universe through faint galaxy counts and redshift distributions in the whole wavelength range spanned by our spectra. In contrast with a rather stable behaviour in the optical and even in the far--IR, the submm counts are dramatically sensitive to variations in the cosmological parameters and changes in the star formation history. Faint submm counts are more easily accommodated within an open universe with a low value of $\\Omega_0$, or a flat universe with a non--zero cosmological constant. We confirm the suggestion of Guiderdoni et al. \\cite*{GHBM98} that matching the current multi--wavelength data requires a population of heavily--extinguished, massive galaxies with large star formation rates ($\\sim 500~M_\\odot$ yr$^{-1}$) at intermediate and high redshift ($z \\geq 1.5$). Such a population of objects probably is the consequence of an increase of interaction and merging activity at high redshift, but a realistic quantitative description can only be obtained through more detailed modelling of such processes. This study illustrates the implementation of multi-wavelength spectra into a semi--analytic model. In spite of its simplicity, it already provides fair fits of the current data of faint counts, and a physically motivated way of interpolating and extrapolating these data to other wavelengths and fainter flux levels. ",
        "introduction": "In the dense gas clouds that harbour starbursts, the ultraviolet (UV) light of young stars is absorbed by dust grains which, in turn, release their thermal energy at infrared (IR) and submillimetre (submm) wavelengths. Thus, understanding the star formation history of galaxies clearly requires a correct assessment of the UV to submm luminosity budget. The most straightforward and simple observational probe of such a luminosity budget is the analysis of the faint galaxy counts obtained at various wavelengths. In this purview, this paper proposes self--consistent theoretical predictions of faint galaxy counts at optical, IR and submm wavelengths that can be directly compared with the current host of data, and used to prepare observational strategies with forthcoming instruments.  In the local universe, only 30 \\% of the bolometric luminosity is released in the IR/submm wavelength range \\cite{SN91}, and the effect of extinction can thus be considered as a mere correction that does not change the main evolutionary trends elaborated from optical studies. However, there is now a growing amount of evidence that this fraction was much higher in the past. Indeed, the discovery of the Cosmic Infrared Background (CIRB) at a level ten times higher than the no--evolution predictions based on the {\\sc iras} local IR luminosity function, and twice as high as the Cosmic Optical Background obtained from optical counts, showed that dust extinction and emission play a major role in defining the luminosity budget of high--redshift galaxies (Puget et al. \\cite*{P+al96}, Guiderdoni et al. \\cite*{GBPLH97}, Schlegel et al. \\cite*{SFD98}, Fixsen et al. \\cite*{FDMBS98}, Hauser et al. \\cite*{H+al98}). Since this major breakthrough, deep surveys with the {\\sc iso} satellite at 15 $\\mu$m (Oliver et al. \\cite*{O+al97}, Aussel et al. \\cite*{ACES99}, Elbaz et al. \\cite*{EC+al99}) and 175 $\\mu$m (Kawara et al. \\cite*{K+al98}, Puget et al. \\cite*{P+al99}), and with the SCUBA instrument at 850 $\\mu$m (Smail et al. \\cite*{SIB97}, Barger et al. \\cite*{BCSFTSKO98}, Hughes et al. \\cite*{HS+al98}, Eales et al. \\cite*{ELGDBHLC99}, Barger et al. \\cite*{BCS99}) have begun to break the CIRB into its brightest contributors. Although identification and spectroscopic follow--up of the submm sources are not easy, the preliminary results of such studies seem to show that part of these sources are the high--redshift counterparts of the local luminous and ultraluminous IR galaxies (LIRGs and ULIRGs) discovered by {\\sc iras} (Smail et al. \\cite*{SIBK98}, Lilly et al. \\cite*{L+al99}, Barger et al. \\cite*{BCSIBK99}). In parallel to this pioneering exploration of the ``optically--dark'' and ``infrared--bright'' side of the universe, a more careful examination of the Canada--France Redshift Survey (CFRS) galaxies at $z\\sim 1$, and Lyman Break Galaxies at $z \\sim 3$ and 4 do show a significant amount of extinction (Flores et al. \\cite*{FH+al99}, Steidel et al. \\cite*{SAGDP99}, Meurer et al. \\cite*{MHC99}). The previous estimates of the UV fluxes, and consequently of the star formation rates, in these objects have to be respectively multiplied by factors 3 and 5 to take into account the effect of extinction. Dust seems to be present at still higher redshifts. For instance, it is seen in a lensed galaxy at $z=4.92$ \\cite{SNFMI98} and even in a Lyman $\\alpha$ galaxy at $z=6.68$ \\cite{CLP99}.  The synthetic spectra of stellar populations in galaxies are easily computed from spectrophotometric models of galaxy evolution. Unfortunately, most of these models neglect the influence of dust on the spectral appearance of galaxies. Guiderdoni \\& Rocca--Volmerange \\cite*{GRV87} proposed a first modelling of the effect of dust extinction. Later, Mazzei et al. \\cite*{MXZ92} basically used the same recipe for extinction, but they also computed dust emission to get spectral energy distributions (SEDs) from the UV to the far--IR. Complete sets of synthetic spectra are now available from the {\\sc grasil} \\cite{SGBD98} and {\\sc stardust} models (Devriendt et al. \\cite*{DGS99}, hereafter Paper I) of spectrophotometric evolution. These models share the same spirit, but they differ by a number of details. The {\\sc grasil} SEDs in the IR are computed from a more sophisticated model of transfer, that is more explicit physically, but involves several free parameters, whereas {\\sc stardust} SEDs in the IR are computed with a minimal number of free parameters, by weighing various dust components to reproduce the observed {\\sc iras} colour--luminosity relations.  These spectra can be used in phenomenological models of faint galaxy counts, that extrapolate the evolution of the local galaxies backwards under the assumption of pure luminosity evolution. For instance, the predictions of faint galaxy counts at optical wavelengths by Guiderdoni \\& Rocca--Volmerange \\cite*{GRV90} used their optical spectra with extinction, whereas Franceschini et al. \\cite*{FZTMD91}, \\cite*{FMZD94} used the Mazzei et al. spectra with dust extinction and emission to produce the first set of counts at optical {\\it and} FIR wavelengths. However, semi--analytic models of galaxy formation (hereafter SAMs) are a much more powerful approach to describe the physical processes that rule galaxy formation and evolution within an explicit cosmological context (White \\& Frenk \\cite*{WF91}, Lacey \\& Silk \\cite*{LS91}, Kauffmann et al. \\cite*{KWG93}, Cole et al. \\cite*{CAFNZ94}, Somerville \\& Primack \\cite*{SP00}). White \\& Frenk \\cite*{WF91}, Lacey et al. \\cite*{LGRS93}, Kauffmann et al. \\cite*{KGW94}, and Cole et al. \\cite*{CAFNZ94} proposed predictions of faint galaxy counts at optical wavelengths (basically the $B$ and $K$ bands) from their models. However, predictions at IR/submm wavelengths from a SAM were produced much later by Guiderdoni et al. (\\cite*{GBPLH97}, \\cite*{GHBM98}, hereafter GHBM). But this first study did not give the corresponding predictions at optical wavelengths, and was restricted to the $\\Omega_0=1$ standard Cold Dark Matter (CDM) model.  In this paper, we implement the {\\sc stardust} spectra into a SAM to make predictions of faint galaxy counts and redshift distributions from the UV to the submm wavelength range, very much in the spirit of GHBM. We also extend the SAM to other cosmologies, and study the sensitivity of the results to the cosmological parameters and star formation history. Although our approach has a number of shortcomings which are due to the simplicity of our model, this paper primarily intends to show that (i) the implementation of {\\sc stardust} SEDs into SAMs is straightforward because of its small number of free parameters;~ (ii) the model gives fits that are already very satisfactory in spite of the simplicity of the approach;~ and (iii) the outputs are a physically motivated tool to interpolate or extrapolate the current observations of faint counts to other wavelengths and/or flux levels. This is needed to prepare the observational strategies with the forthcoming IR/submm satellites {\\sc sirtf}, {\\sc first} and {\\sc planck}, as well as with the Atacama Large Millimetre Array.  In section 2, we briefly describe how we connect the {\\sc stardust} spectra with the various physical processes that are relevant to galaxy formation, within our SAM. We point out the differences with GHBM. Section 3 discusses the values of the free parameters that define the so--called ``quiescent'' mode of star formation. Section 4 focuses on the sensitivity of the faint galaxy counts to a change in the cosmological parameters. Section 5 studies the sensitivity of the faint galaxy counts to galaxy evolution, and more specifically to the presence of a heavily--extinguished ``starburst'' mode of star formation similar to the one in ULIRGs. Finally, a fiducial model is proposed. We discuss our results in Section 6. ",
        "conclusions": "In this paper, we have proposed a first implementation of the set of {\\sc stardust} synthetic spectra into a SAM of galaxy formation. Although our model is quite simple, and cannot properly handle the merging history trees of halos and galaxies, we have shown that the implementation of {\\sc stardust} is quite straightforward. This implementation can be easily achieved in more sophisticated SAMs.  We have not explored a large set of statistical properties in this paper, but have only illustrated the ability of this approach to reproduce the optical/IR/submm luminosity budget by producing predictions of faint galaxy counts at UV, visible, NIR, mid--IR, FIR, and submm wavelengths. As in GHBM, we have defined a quiescent mode of star formation which corresponds to the ``natural'' values of these astrophysical parameters as they are suggested by local observations of disks (for $\\beta$ and $f_c$), or by numerical simulations (for $\\epsilon_{SN}$).  We have then studied the influence of the cosmological parameters $\\Omega_0$ and $\\Lambda$ on the faint counts. In SAMs, the cosmological parameters influence the counts at various stages of the computation, through halo collapse as well as through the relationship of the cosmic times, luminosity distances and volume elements to redshifts. These quantities intervene in the computation of the counts in different ways, and their effect is dimmed or enhanced by the $k$--corrections. The net result on the optical and NIR counts, with the influence of a positive $k$--correction, is a rather weak sensitivity to the values of the cosmological parameters. In contrast, because of the negative $k$--correction that enhances what happens at high redshift, the submm counts show a strong sensitivity to cosmology.  We know that some of the SCUBA sources detected at 850 $\\mu$m are the high--redshift counterparts of local LIRGs and ULIRGs. These objects harbour heavily--extinguished starbursts due to gas inflows triggered by interaction/merging. Our SAM is unable to address this process, and, as in GHBM, we chose to implement a heavily--extinguished starburst mode of star formation, by increasing the number fraction of massive objects that undergo this stage, as the squared density. Of course, the LIRGs and ULIRGs can also be powered by a top--heavy IMF. We have shown the strong sensitivity of the submm counts to the details of the ULIRG scenario, because of the negative $k$--correction, contrasting it with the weak influence on the optical counts.  We have also produced redshift distributions at various wavelengths and flux cuts. Here again, the sensitivity of the results to cosmology and evolution is dramatic at submm wavelengths. We got a fair fit of the CFRS redshift distribution in the $I$ band, and the redshift distribution predicted for the NEPR survey looks much better than in GHBM. At submm wavelengths, we predict a double--peaked distribution with nearby (mostly quiescent) sources and distant (mostly starburst) sources. The relative weight of the two broad peaks is sensitive to cosmology and evolution. Sufficient statistics in an observational sample could in principle help disentangling these effects. Unfortunately, the identification process of the submm sources is difficult, either in ISOPHOT or in SCUBA samples, and we might have to wait for multi--wavelength observations with forthcoming satellites such as {\\sc sirtf} and {\\sc first} to solve this issue. In this context, our predicted counts are also a useful ingredient to analyse current data, and to prepare observational strategies with these satellites. In this purview, we have proposed a new ``fiducial model'' in a $\\Lambda$CDM cosmology which can take the place of the model ``E'' in GHBM, and which is available upon request.  Merging--triggered starbursts and subsequent bulge formation are key--processes in the paradigm of hierarchical galaxy formation. It turns out that these processes are particularly apparent at IR/submm wavelengths, and almost invisible at optical wavelengths. So multi--wavelength observations are required to constrain the history of galaxy formation in the quiescent and starburst modes. The complete merging history of galaxies has to be followed in detail, especially during the short periods of interaction and merging which produce IR luminous sources. A hybrid method of galaxy formation using high--resolution N--body simulations to plant semi--analytic galaxies is being developed and will be used to quantify the importance of merging processes (Hatton et al. \\cite*{HNDBGSV00} and following papers). In this context, the reader is invited to notice the following point: any change in the luminosities and number densities of the ULIRGs results into spectacular changes in the submm counts, because of their super--Euclidean regime. This sort of ``instabilities'' in the behaviour of the faint counts is going to be very useful to constrain the luminosity and duration of the starbursts in more refined models.  Throughout this paper, we have assumed that dust heating is powered by star formation. The other possible engine is the presence of heavily--extinguished AGNs located at the centers of ULIRGs. In local samples, heating is dominated by starbursts, except in the most luminous galaxies (Genzel et al. \\cite*{G+al98}, Lutz et al. \\cite*{LSRMG98}). So far, we do not know how this situation evolves with redshift. The redshift distribution of the SCUBA sources seems to peak around redshift 2.5 where the observed redshift distribution of quasar activity may also peak \\cite{P95}. Perhaps a sophisticated combination of AGNs and starbursts with a top--heavy IMF is responsible for the evolution seen by SCUBA, and we signal that attempts to include both types of ingredients are already discussed in the literature (e.g. Blain et al. \\cite*{BJSLKI99}). However, properly disentangling all these components definitely requires a more sophisticated model than the one presented here. We therefore defer this issue to a later study, but remark that AGNs can be implemented self--consistently within the framework of the hybrid method \\cite{KH00}, so that one could hope to set quantitative limits on their respective contributions to the submm fluxes.  The situation is indeed as complex from the theoretical as well as the observational points of view. That is why this simple modelling should be considered as an exploratory step to probe the complex issues involved in multi--wavelength counts, and to design a more satisfying approach of the problem."
    },
    "0010/astro-ph0010221_arXiv.txt": {
        "abstract": "We report the detection ($3\\sigma$ significance level) of a strong iron emission line in the X-ray spectrum of the afterglow of GRB 000214 (``Valentine's Day Burst'') observed by BeppoSAX. An emission line feature was observed with a centroid energy of $4.7\\pm0.2$ keV which, if interpreted as K$\\alpha$ emission from hydrogen-like iron, corresponds to a redshift of z=0.47. The intensity (EW$\\sim 2$~keV) and duration (tens of hours) of the line give information on the distance, from the burst region, of the emitting material ($R\\geq3\\times10^{15}$ cm) and its mass ($M\\geq1.4M_{\\odot}$). These results are not easily reconciled with the binary merger and hypernova models for gamma ray bursts, because they require large amounts of mass (about $1 M_\\odot$), at large distances ($\\approx 10^{16}\\: cm$), and at newtonian speeds. ",
        "introduction": "Three years after the first identification of the X-ray afterglow from a Gamma Ray Burst (GRB) (Costa et al., 1997), the nature of Gamma Ray Bursts' progenitors remains uncertain. The most widely discussed theoretical models for GRBs consider vastly different scenarios both in terms of the progenitors and environment. If the GRBs originate from neutron star-neutron star or neutron star-black hole mergers (Narayan, Paczynski and Piran 1992) then, for about a half of all GRBs, the explosion should be located far from the place where the progenitor binary system was formed and most probably in a low-density interstellar medium. On the other hand, if GRBs originate from Hypernovae (Paczynski, 1998; Woosley, 1993) or SupraNovae (Vietri \\& Stella, 1998), the immediate progenitor is a massive, rotating star, and thus the explosion should take place in a high density medium, probably a star-forming region. The discovery of emission lines in the X-ray spectrum of a GRB afterglow represents a major step toward the understanding of the nature of GRB progenitors. In fact, the measurement of X-ray lines emitted by a GRB or its afterglow provides a direct measurement of the redshift and a powerful diagnostic of both the nature of the central engine and the environment in which GRBs go off (Perna \\& Loeb, 1998; M\\`esz\\`aros \\& Rees, 1998; B\\\"ottcher et al., 1999, Ghisellini et al., 1999; Paerels et al., 2000). Other methods, such as collecting offsets between GRBs afterglow and their host galaxies (Bloom et al., 1999) or searching for absorbing material between the GRB and the observer (Wijers \\& Galama, 1999; Owens et al., 1999), are comparatively less direct.  Presently, there are only two marginal detections of an X-ray emission line in GRB afterglows,  one in GRB 970508 (chance probability for a statistical fluctuation of 0.7\\%, Piro et al. 1999) and one in GRB 970828 (probability of 1.7\\%, Yoshida et al. 1999).  The presence of a strong iron line in the X-ray spectrum of a GRB afterglow implies a rich environment located very close to the GRB region and may be an important clue in favor of collapsar models (M\\`esz\\`aros \\& Rees, 1998; Lazzati, Campana, \\& Ghisellini, 1999; Weth et al., 2000; Vietri et al., 1999). In fact, in the case of the Hypernova scenario an iron rich circumburst environment may be produced by the stellar wind before the explosion of the Hypernova (M\\`esz\\`aros \\& Rees, 1998). A similarly favorable situation occurs in the SupraNova scenario (Vietri \\& Stella, 1998) where a supernova explosion precedes by a few months the GRB event. GRB 000214 (``Valentine's Day Burst'') has no optical nor near-infrared afterglow (Rhoads et al., 2000; Wijers et al., in preparation) as is the case for about half of all GRBs with follow--up observations (e.g., Kulkarni et al., 2000).  The lack of optical and IR counterparts to a GRB may result from extinction in a dense surrounding medium; therefore, searching for emission lines in the X-ray afterglows of these GRBs looks especially promising.  Based on a BeppoSAX pointed observation of GRB 000214, we report on the detection of an Iron K$\\alpha$ line from its X-ray afterglow. ",
        "conclusions": "The only cosmologically abundant element that can produce lines beyond 4 keV is iron, provided one precludes Doppler shifts to shorter wavelengths.  Iron emits a K$\\alpha$ line photons with an energy between 6.4 (for the lower ionization stages of iron) and 6.95 keV (for hydrogen-like iron). The observed feature may be also interpreted as the Iron recombination K--edge at 9.28 keV (corresponding to a redshift of $z=0.9$) but, in such a case, according to Weth et al. (2000), we should expect both a Fe K--edge and a K$\\alpha$ line having about the same intensity. The Fe-K$\\alpha$ line should be redshifted at an energy between 3.2 and 3.5 keV but no line is detected in the MECS spectrum of GRB 000214 afterglow within this range.  If we identify the emission feature in GRB 000214 as Fe K$\\alpha$, the corresponding redshift is between $z=0.37$ and 0.47. We will assume $z=0.47$ for simplicity, but none of the following discussion hinges upon this exact value.  The case of GRB 000214 differs from that of GRB 970508 in two ways. First, the iron line is more significant in GRB 000214: we find that the probability of a chance fluctuation is $0.27\\%$, as compared to $0.7\\%$. This difference becomes even more remarkable when one considers that there is no independent determination of the redshift for GRB 000214, while GRB 970508 has $z = 0.835$ from optical observations (Metzger et al., 1997). Second, the afterglow observation of GRB 000214 did not allow detection of a possible reburst, contrary to GRB 970508. In fact, GRB 000214 was observed just once after the burst proper, beginning $12$ h after the main event, and for a total of $\\approx 10^5$ s.  GRB 970508, instead, was observed on four different occasions after the burst, beginning $6$ h after the main event, with the last set of observations more than $6$ days after the burst (Piro et al., 1999). Thus, for GRB 000214, we have no time coverage to exclude a reburst before or after the time of our observations: all we can say is that, during this limited time, the flux decreased by about a factor of $2$ (see Section 2.1).  This is consistent with GRB 970508, where the line was bright as the flux was decreasing (Fig. 2a of Piro et al., 1999); thus the model of Vietri et al. (1999), connecting the appearance of the iron line to the reburst, is still viable.  For $z = 0.47$, $H_\\circ = 70\\;$ km s$^{-1}$ Mpc$^{-1}$, and $\\Omega = 1$, we find a total energy release, in the line only, of $E_l = 3\\times 10^{48}\\;$ erg, i.e. $N_l = 3\\times 10^{56}$ photons emitted within the first $t_d \\approx 10^5$ s after the burst; the emission is isotropic because the line is sub--relativistic. If each iron atom were to emit a single line photon, then $M_{Fe}\\approx 12 M_\\odot$, way too large. Thus each iron atom must produce several line photons (Lazzati \\etal, 1999), and, consequently, the time--scale $t_{Fe}$ over which every iron atom produces a photon satisfies $t_{Fe} \\ll t_{d} \\approx 10^5\\;$ s, and does not seem to begin fading. This implies that the time--scale $t_d$ is set by the geometrical factor: the line--emitting material must be located at $R \\ga 3\\times 10^{15}\\;$ cm, and we are seeing some sizeable fraction of it.  The minimum iron mass is determined by Lazzati \\etal (1999) as \\begin{equation} M_{Fe} \\geq 5\\times 10^{-3} M_\\odot \\frac{F_{-13}t_5 R_{16}^2}{q E_{50}} \\end{equation} where $F_{-13} \\approx 0.5$ is the line luminosity in units of $10^{-13}$ \\cgs, $t_5 \\approx 1$ is the line duration in units of $10^5\\;$ s, $R_{16}\\approx 0.3$ the material distance in units of $10^{16}\\;$ cm, $q$ is the fraction of the absorbed ionizing fluence and reprocessed into the line ($q\\la 0.1$ as derived by Ghisellini et al., 1999) and $E_{50}$ is the total X--ray burst luminosity in units of $10^{50}$ erg; for GRB 000214 we find $ E_{50} \\approx 8$ if the burst emission were isotropic. With these values \\begin{equation} \\label{iron} M_{Fe} \\ga 2.4\\times 10^{-3} M_\\odot \\;. \\end{equation}  This argues against a possible origin of this GRB in a neutron star - neutron star (or neutron star - black hole) binary merger. The problem is not one of total iron content of the model: for realistic parameters, this scenario can easily satisfy the above limit. The problem lies instead in the capacity of the model to deposit this much iron at these large distances, and at Newtonian speeds.  In fact, under our assumption that the line is not blueshifted (at least, its Lorenz factor is $\\approx 1$), the iron cannot have been released in the burst event itself: there is not time enough for matter ejected in the explosion to rush ahead of photons, slow down, and then be hit by the burst flash. The material generating the line must have been released well before the burst. Given their lack of atmospheres and winds, the only mechanism capable of releasing matter from inspiralling neutron stars is tidal interaction. A mass loss estimate for this process by M\\`esz\\`aros and Rees (1992) has a lower limit of $M_{th} = 4\\times 10^{30}\\; g$, enough to account (barely!) for Eq. \\ref{iron}, but this optimistically assumes that the typical viscosity damping time-scale equals the light travel time across the star (Bildsten and Cutler 1992). Though the exact amount of viscosity in a neutron star is not well--known, theoretical arguments fall short of the amount required for the correctness of the estimate by M\\`esz\\`aros and Rees (1992), by about a factor $2000$ (Kochanek 1992, Eq. 4.7, Bildsten and Cutler 1992, see the discussions after Eq. 9 and 16).  In hypernova models, an intense wind should characterize the massive progenitor before the collapse. Calling $A$ the iron mass fraction relative to the solar abundance (Lang 1991, p. 84), we expect from Eq. 2  a minimum mass of \\begin{equation} M \\geq 1.4 M_\\odot \\frac{1}{A E_{50}} \\;. \\end{equation} The radial mass distribution from a massive stellar wind is $d\\!M = \\dot{M} d\\!r/v_\\infty$.  The most extreme ratios of stellar wind mass loss rate $\\dot{M}$ to asymptotic speed $v_\\infty$ occur in Luminous Blue Variables and Red SuperGiant Variables ($\\dot{M} = 10^{-4} M_\\odot\\;$ yr$^{-1}$, $v_\\infty = 10-100$ km s$^{-1}$, Lamers, 1995, van Loon et al., 1999). For these winds, at $R\\approx3\\times 10^{15}$~cm there can only be $M\\approx 10^{-2}-10^{-3} M_\\odot$, missing the above lower limit by $2-3$ orders of magnitude. The densest circumstellar envelopes around the progenitors of some peculiar core collapse supernovae ({\\it e.g.}  $n\\sim10^8-10^9$~cm$^{-3}$ at $R\\sim 3\\times10^{15}$~cm in SN1997ab, Salamanca et al., 1998) imply a mass of $M\\sim10^{-1}-10^{-2}~M_{\\odot}$, substantially lower than required by Eq. 3.  A model immune from these inconsistencies is the SupraNova, where a core collapse supernova explosion precedes the GRB event by months to years (Vietri and Stella 1998).  On the other hand, the lack of large amounts of material along the line of sight, as testified by the absence of photoelectric soft X-ray absorption on top of the Galactic value\\footnote{The upper limit derived above, $N_H = 2.7\\times 10^{20}$ cm$^{-2}$, for the non--Galactic column density, corresponds to a total mass, within $12$ light hours, of $5\\times 10^{-6} M_\\odot$, which could indicate a large amount of mass only for implausibly large ($\\approx 1-10^{-6}$) ionization fractions.}, and the development of a normal X--ray afterglow, clearly suggest the presence of outlying material located in the progenitor's equatorial plane, at large angles from the line of sight. This material may be either left over from the progenitor's formation process (as in a primordial gas cloud) or spewed out by the progenitor. Key future observations capable of discriminating between these possibilities will be the measurement of the line width, and a search of atomic species produced in the interiors of evolved stars. Velocities of $\\ga 1000\\; km\\; s^{-1}$ can rule out any primordial cloud and/or hypernovae winds; the detection of cobalt or nickel may establish the recent origin of the material from a SN explosion.  \\medskip  This research is supported by the Italian Space Agency (ASI) and Consiglio Nazionale Ricerche of Italy. Beppo-SAX is a joint program of ASI and the Netherlands Agency for Aerospace Programs (NIVR).  We thank all members of the Beppo-SAX SDC, SOC and OCC. We also thank L. Burderi, G.L. Israel, F. Fiore, G.Matt and G. Bono and an anonymous referee for helpful comments."
    },
    "0010/astro-ph0010017_arXiv.txt": {
        "abstract": " ",
        "introduction": " ",
        "conclusions": "The IRAS color of 19312+1950,  $(C_{12}, C_{23})=(0.50, 0.78)$, suggests a uniqueness of this object.  In a two-color diagram (van der Veen, Habing 1988), this object falls into region VIII,  which is occupied by a number of very cool stars. About 30\\% of objects with the IRAS LRS spectra in region VIII indicate planetary nebulae. The LRS spectra of this object (Volk et al. 1991) exhibits weak silicate absorption on a red continuum. On the other hand, this object mimics the color of young stars found in star--forming regions (MacLeod et al. 1998a). Based on the IRAS colors, it is hard to classify IRAS 19312+1950 as a young planetary nebula or a young star in a star--forming region. It is not surprising that a negative result of the 6.7 GHz methanol maser search was reported toward this object (MacLeod et al. 1998b).  Figure 2 shows the 2MASS near-infrared image (Skrutskie et al. 2000) \\footnote{Atlas Image obtained as part of the Two Micron All Sky Survey (2MASS),  a joint project of the University of Massachusetts and the Infrared Processing  and Analysis Center/California Institute of Technology, funded by the National Aeronautics  and Space Administration and the National Science Foundation, U.S.A.} of the region around IRAS 19312+1950. We can see a nebulosity with two horns extended towards to south west. These horns can clearly be seen on this image, though the IRAS source is point-like in the $J$ band. From this image and the detections of SiO masers in this source, we conclude that this is a post-AGB star surrounded by a bipolar-type nebulosity (like the Rotten Egg Nebula, OH 231.8+4.2; Morris et al. 1987). There is no nearby infrared star which is bright in the 2MASS $K$-band image. We conclude that the extended nebulosity at the center of figure 2 is a part of this object.  A more observational investigation is necessary to clarify the nature of this object. For instance, the radial velocities of the maser and thermal lines toward this object are puzzling. From the velocities of the maser lines (SiO and H$_{2}$O), we infer a stellar velocity of approximately $V_{\\rm lsr}=20$ km s$^{-1}$ [it is well known that the SiO radial velocity coincides with the stellar velocity (Jewell et al. 1991), while H$_2$O spectra often exhibit single or double peaks  centered at the stellar velocity (Engels, Lewis 1996)]. However, the $J=1-0, v=2$ spectrum exhibits another component at  $V_{\\rm lsr}=52$ km s$^{-1}$. It is possible that this interloper component comes from a nearby source which is located by chance in the same telescope beam of about 40$''$ toward this object; such examples are already known for two cases among the densely populated disk IRAS sources (Deguchi et al. 1999). On the other hand, it is possible to interpret this component as part of a doubly peaked profile which is formed in a rotating disk or expanding shell surrounding this object. In this case, the stellar velocity would be $V_{\\rm lsr} \\simeq 39$ km s$^{-1}$. If this is correct, it is also possible (but less plausible) to interpret this source as a young stellar object. In past studies, the detection of SiO masers in several star-forming regions has been reported (e.g., Hasegawa et al. 1986; Snyder, Buhl 1974). A doubly peaked profile with a separation of more than 20 km s$^{-1}$ was found in Orion IRc2, and has been explained by a rotating and expanding disk model (Plambeck et al. 1990; Barvainis 1984). On the other hand, SiO masers from proto-planetary nebula show a somewhat broad line profile in past studies; it resembles to a typical profile of SiO masers from a circumstellar envelope of the Mira-type variables (e.g., Nyman et al. 1998). Up to now, no clear doubly peaked profile of SiO masers in the bipolar nebulae has been reported.  The radial velocities of H$^{13}$CN and SO are shifted by about 12 km s$^{-1}$ from the maser lines. Because of this large separation in radial velocities and the slightly asymmetric profiles of these lines, we consider that these thermal lines come from a background molecular cloud. According to the Columbia CO survey in the galactic disk (Cohen et al. 1986), CO emission at $(l, b)= (55.0^{\\circ}, 0.25^{\\circ})$ is peaked at $V_{\\rm lsr}\\simeq$ 30 km s$^{-1}$, which coincides with the radial velocities of H$^{13}$CN and SO found in this paper. However,  we cannot completely neglect the possibility that the two broad lines of H$^{13}$CN and SO come from the outflowing envelope of this object, unless the component ($V_{\\rm lsr}=$52 km s$^{-1}$) of SiO is proved to come from a totally different source by chance, and until the CO emission in the background molecular cloud is mapped by telescopes with better spatial resolution.  It is known that very few sources are detected in SiO masers beyond $C_{12}=0.5$ (Nyman et al. 1998); only two stars, OH 231.8+4.5 and OH 15.7+0.8, are such extreme sources with $C_{12}>0.5$ accompanying SiO. However, the color $C_{23}$ of these sources are below 0.4. One of the sources with the colors compatible to 19312+1950, is IRAS 18498$-$0017 (OH 32.8$-$0.3; $C_{12}=0.46$ and $C_{25}=0.74$); H$_2$O, SiO, and OH masers have been detected in this source (Cesaroni et al. 1988; Engels, Heske 1989; te Lintel Hekkert et al. 1989). Near-infrared images of this source are not available at present in the 2MASS archive. It is quite interesting to see if this source also exhibits a bipolar-type nebulosity or not. The color, $C_{23}\\sim 0.7$, of these two objects seems to be meaningfully higher than those of OH 231.8+4.5 and OH 15.7+0.8. However, because the positions of 18498$-$0017 and 19312+1950 are very near to the galactic plane ($b=-0.04^{\\circ}$ and 0.2$^{\\circ}$), emission from the interstellar dust grains possibly contaminates the 60 $\\mu$m flux densities of these objects, as found by the Nobeyama SiO maser survey of the galactic disk IRAS sources (Izumiura et al. 1999). The long-term near-infrared  monitoring of 18498$-$0017  (Engels et al. 1983) found a variability  with a time-scale of more than 1700 d, suggesting that 18498$-$0017 is not a normal Mira variable.  It may be at a transient stage to the proto-planetary nebula. \\par \\vspace{1pc}\\par"
    },
    "0010/hep-ph0010244_arXiv.txt": {
        "abstract": "We consider neutralino-proton cross sections for halo dark matter neutralinos ($\\tilde\\chi_1^0$) within the framework of supergravity models with R-parity invariance for models with universal soft breaking (mSUGRA) and models with nonuniversal soft breaking. The analysis includes the necessary corrections to treat the large tan$\\beta$ region (i.e. L-R mixing in the squark and slepton mass matrices, loop corrections to the $b$ and $\\tau$ masses,etc) and includes all coannihilation phenomena. For mSUGRA, dark matter detectors with current sensitivity are seen to be probing the region where tan$\\beta \\stackrel{>}{\\sim}$25, $\\Omega_{\\tilde\\chi_1^0} h^2 <0.1$, $m_{\\tilde\\chi_1^0}\\stackrel{<}{\\sim}90$ GeV, and for the light Higgs, $m_h \\stackrel{<}{\\sim}120$ GeV. Nonuniversal models can have a much larger cross section, and current detectors can probe part of the parameter space where tan$\\beta\\stackrel{>}{\\sim}4$. Minimum cross sections are generally greater than $10^{-9}$ pb to $10^{-10}$ pb for $m_{1/2} <600$ GeV (and hence accessible to planned future detectors), with the exception of a region when $\\mu <0$ where for $m_{1/2}\\stackrel{>}{\\sim}450$ GeV, $4\\stackrel{<}{\\sim} \\tan\\beta \\stackrel{<}{\\sim}20$, the cross section drops to a minimum of about $1\\times10^{-12}$ pb at $m_{1/2}=600$ GeV, $\\tan \\beta \\simeq10$. In this region, the gluino and squarks lie above 1 TeV, but should still be accessible to the LHC. ",
        "introduction": "If the dark matter that exists in the Milky Way is a supersymmetric weakly interacting particle (wimp), there are several ways in which it might be detected. Annihilation of two wimps in the halo might give rise to signals of gamma rays, anti-protons or positrons. Dark matter particles caught by the gravitational fields of the Sun or Earth would be expected to sink to the center, and there annihilate leading to neutrinos that might be detected on the surface of the earth. Finally direct detection of incident wimps from their scattering by nuclear targets on the Earth is possible. Of these, the last possibility appears most promising, and there are now a large number of detectors searching for supersymmetric wimps. We consider here what signals might be available within the framework of supergravity(SUGRA) models with grand unification of the gauge coupling constants at the GUT scale $M_G\\cong2\\times10^{16}$GeV. There are three different types of SUGRA models currently being investigated, which differ by the mechanisms used to achieve supersymmetry(SUSY) breaking in the physical sector. These are gravity mediated SUGRA models, gauge mediated models, and anomaly mediated models. Of these, the gravity mediated models with R-parity invariance have the most robust candidate for particle dark matter, and we will restrict our discussion here to such models. In gravity mediated models, the dark matter particle is the lightest supersymmetric particle, the LSP, (absolutely stable due to the R-parity invariance), and this is generally the lightest neutralino, $\\tilde\\chi_1^0$. The nucleus-$\\tilde\\chi_1^0$ scattering cross section contains a spin independent part and a spin dependent part. For heavy nuclear targets, the spin independent scattering dominates, and it is possible to extract from data the $\\tilde\\chi_1^0$-proton cross section, $\\sigma_{\\tilde\\chi_1^0-p}$. There are a number of astronomical uncertainties, but making conventional assumptions, current detectors (DAMA, CDMS, UKCDM) are sensitive to halo $\\tilde\\chi_1^0$ if \\begin{equation} \\sigma_{\\tilde\\chi_1^0-p} \\stackrel{>}{\\sim} 1\\times10^{-6} {\\rm pb} \\end{equation} and future detectors (GENIUS, Cryoarray) plan to achieve sensitivities of \\begin{equation} \\sigma_{\\tilde\\chi_1^0-p} \\stackrel{>}{\\sim} (10^{-9} -10^{-10}){\\rm pb} \\end{equation}  We discuss here how these sensitivities might relate to supergravity models. In particular, we consider the minimal supergravity GUT model (mSUGRA)[1] which has universal soft breaking masses at $M_G$, and nonuniversal soft breaking models[2] which allow nonuniversal Higgs masses and nonuniversal third generation squark and slepton masses at $M_G$ (but keep the gaugino masses universal at $M_G$). While the models are physically different, they lead to qualitatively similar results: Current detectors are sensitive to a significant part of the SUSY parameter space, and future detectors should be able to cover all of the parameter space, except for special regions where there is an accidental cancelation of terms making $\\sigma_{\\tilde\\chi^0_1-p}$ anomalously small. Each of the above models contains a number of arbitrary new parameters. In spite of this they can still make relevant predictions for two main reasons: (i) Using the renormalization group equations (RGE) starting from $M_G$, they allow for radiative breaking of $SU(2)\\times U(1)$ at the electroweak scale (and thus furnish a natural explanation for the Higgs mechanism); (ii) Along with being able to calculate $\\sigma_{\\tilde\\chi^0_1-p}$, the models can also calculate the relic density of neutralinos, i.e. $\\Omega_{\\tilde\\chi_1^0} h^2$, where $\\Omega_{\\tilde\\chi_1^0} = \\rho_{\\tilde\\chi_1^0}/\\rho_c$, $\\rho_{\\tilde\\chi_1^0}$ is the relic mass density of the ${\\chi_1^0}$, and $\\rho_c=3H_0/8\\pi G_N$. Here $H_0=h(100\\mbox{km/s Mpc})$ is the Hubble constant, and $G_N$ is the Newton constant. Both of the above lead to important constraints on the SUSY parameter space. Thus one has that \\begin{equation} \\Omega_{\\tilde\\chi^0_1}h^2\\sim(\\int^{x_f}_0 dx\\langle\\sigma_{\\rm ann}v\\rangle)^{-1} \\end{equation} where $\\sigma_{\\rm ann}$ is the annihilation cross section in the early universe, v is the relative  neutralino velocity at annihilation, and $<...>$ means thermal average. The dominant Feynman diagrams for $\\sigma_{\\rm ann}$ and spin independent $\\sigma_{\\tilde\\chi^0_1-p}$ are shown in Fig.\\ref{fig1}, and roughly speaking $\\sigma_{\\rm ann}$ depends on the crossed diagrams relative to $\\sigma_{\\tilde\\chi^0_1-p}$. Thus usually, then, when $\\sigma_{\\tilde\\chi^0_1-p}$ is large, $\\sigma_{\\rm ann}$ will also be large, and hence by Eq(3), $\\Omega_{\\tilde\\chi_1^0}h^2$ will be small. Thus lower bounds on $\\Omega_{\\tilde\\chi_1^0}h^2$ will produce upper bounds on $\\sigma_{\\tilde\\chi^0_1-p}$. In the following, we will assume $h=0.70\\pm0.07$ and for matter(m) and baryonic matter(b) the values $\\Omega_m=0.3\\pm0.1$, and $\\Omega_b=0.04$. (This corresponds to a dark energy amount of $\\Omega_\\Lambda \\sim0.65$). For the dark matter then one has $\\Omega_{\\tilde\\chi_1^0}=0.26\\pm 0.10$, and if one combines errors in quadrature, one finds $\\Omega_{\\tilde\\chi_1^0} h^2 =0.13\\pm0.05$. Since there is undoubtedly a large amount of systematic error in the above estimates, in the following we will assume the approximately 2$\\sigma$ spread of \\begin{equation} 0.02 < \\Omega_{\\tilde\\chi_1^0} h^2 < 0.25 \\end{equation} The lower bound of Eq.(4) lies somewhat below other estimates. However, it also allows for the possibility that not all the dark matter in the Galaxy are neutralinos (e.g. some may be machos). In addition to the above, there are accelerator bounds that constrain the SUSY parameter space. In the following we use the LEP bounds for the light Higgs ($h$) of $m_h > 104$ GeV for tan$\\beta$ = 3, $m_h > 102$ GeV for tan$\\beta=$5 and for the light chargino $ m_{\\tilde\\chi_1^\\pm} > 102$ GeV. (For tan$\\beta > $5, the Higgs mass bounds do not restrict the parameter space significantly.) The Tevatron gives the gluino ($\\tilde g$) mass bound of $m_{\\tilde g} > 270$ GeV (for gluino and squarks nearly degenerate). In addition there is the CLEO measurment of the $b\\rightarrow s + \\gamma$ decay. We take here a 2$\\sigma$ range around the experimental central value of the $b\\rightarrow  X_s + \\gamma$ branching ratio[3]: \\begin{equation} 1.8 \\times10^{-4} < B(B\\rightarrow X_s \\gamma) < 4.5 \\times10^{-4} \\end{equation} Of the above, the most significant constraints come from the Higgs mass bounds and the $b \\rightarrow s\\gamma$ branching ratio. \\begin{figure}[htb] \\centerline{ \\DESepsf(diag.eps width 10 cm) } \\caption {\\label{fig1} Dominant diagrams for $\\sigma_{\\rm ann}$ (upperdiagrams) and spin independent part of $\\sigma_{\\tilde\\chi^0_1-p}$(lower diagrams).} \\end{figure} ",
        "conclusions": "We have examined here the predictions of several SUGRA models which possess gauge coupling constant unification at $M_G\\cong2\\times10^{16}$ GeV, to see what parts of the SUSY parameter space are accessible to current detectors obeying Eq.(1), and what will be accessible to future detectors with the sensitivity of Eq.(2). For the minimal SUGRA model with universal soft breaking parameters, mSUGRA, current detectors are scanning parts of the parameter space where tan$\\beta\\stackrel{>}{\\sim} 25$, $m_{\\tilde\\chi^0_1}\\stackrel{<}{\\sim} 90$ GeV and $\\Omega h^2\\stackrel{<}{\\sim}0.1$. In addition, the light Higgs obeys $m_h\\stackrel{<}{\\sim} 120$ GeV, and hence possibly would be accessible to RUN2 at the Tevatron. For nonuniversal models, where one allows the Higgs and third generation squark and slepton softbreaking masses to be nonuniversal, the neutralino-proton cross section can be significantly increased, by a factor of 10-100, with an appropriate choice of sign in the soft breaking deviations from universality. Thus current detectors here could scan regions of parameter space as low as tan$\\beta\\simeq4$, though in these regions $m_h$ is very light, and the the minimum allowed tan$\\beta$ may have to be raised as LEP raises the the bound on the Higgs mass. However, the possibility of large cross section here has already allowed current detectors to exclude parts of the high tan$\\beta$ region, e.g. when tan$\\beta \\stackrel{>}{\\sim}$15. How low SUGRA cross sections can lie is complicated by the existance of coannihilation effects where the R-sleptons (particularly the $\\tilde\\tau_1$) can become nearly degenerate with the neutralino. This allows the relic density constraint Eq.(4) to be satisfied in a narrow rising corridor (about 25 GeV wide in $m_0$) in the $m_0-m_{1/2}$ plane rising to relatively large $m_0$, and thus reducing the size of the ${\\tilde\\chi^0_1}-p$ cross section. For large tan$\\beta$, the effect is sensitive to the value of $A_0$, the range of $m_0$ increasing with $A_0$. Thus for $\\mu >$0, one finds for $m_{1/2}=$600 GeV, the minimum cross section at e.g. tan$\\beta=40$, $A_0=4 m_{1/2}$ is almost the same as that at tan$\\beta=3$, the increase of the cross section due to the increase in tan$\\beta$ being offset by the decrease due to the allowed large value of $m_0$. One finds however, that for $\\mu>0$, at m$_{1/2}$ = 600 GeV ($m_{\\tilde g}\\stackrel{\\sim}{=}1.5$ TeV), the minimum cross sections would still be accessible to detectors with the sensitivity of Eq.(2). The minimum cross sections for $\\mu < 0$ is complicated by the possibility of accidental cancelations in the scattering amplitudes, allowing the cross section to sink below 10$^{-10}$ pb in certain regions of the parameter space. Thus one finds for mSUGRA that $\\sigma_{\\tilde\\chi^0_1-p }< 1\\times 10^{-10}$ pb for $m_{1/2} \\stackrel{>}{\\sim}450$ GeV ($m_{\\tilde g}\\stackrel{>}{\\sim}1.1$ TeV) when $4 \\stackrel{<}{\\sim}\\tan\\beta \\stackrel{<}{\\sim}$ 20, with a minmum cross section of $1\\times 10^{-12}$ pb reached at $\\tan\\beta\\cong$10. Similar results hold for the nonuniversal models. In this domain SUGRA models imply that halo dark matter would not be accessible to detectors with sensitivities of Eq.(2), and that the gluino and squarks would be quite heavy. They would still however be observable at the LHC which can detect gluinos with mass $m_{\\tilde g}\\stackrel{<}{\\sim}$ 2.5 TeV [16]."
    },
    "0010/astro-ph0010292_arXiv.txt": {
        "abstract": "Since 1996, a hybrid experiment consisting of the emulsion chamber and burst detector array and the Tibet-II air-shower array has been operated at Yangbajing (4300~m above sea level, 606 g/cm$^2$) in Tibet. This experiment can detect air-shower cores, called as burst events, accompanied by air showers in excess of about 100~TeV. We observed about 4300 burst events accompanied by air showers during  690 days of operation and selected 820 proton-induced events with its primary energy above 200 TeV using a neural network method. Using this data set, we obtained the energy spectrum of primary protons in the energy range from 200 to 1000~TeV. The differential energy spectrum obtained in this energy region can be fitted by a power law with the index of -2.97  $\\pm$ 0.06, which is steeper than that obtained by direct measurements at lower energies. We also obtained  the energy spectrum of helium nuclei at particle energies around 1000~TeV. ",
        "introduction": "Shock acceleration at supernova blast waves gives a good explanation of the origin of the bulk of  cosmic rays.  It may be well accepted that cosmic rays below about 10 TeV are predominantly due to the  explosion of stars (supernova explosion) into the normal interstellar medium, while particle acceleration at supernova remnants (SNR's) has an upper limit of about 100~TeV \\cite{1,2}.  Also, there is an argument that the cosmic rays from near 10~TeV to several times 1000~TeV very likely  originate in the explosion of massive stars into their former stellar wind \\cite{3}. These processes  have been examined to be able to explain the cosmic ray spectra fairly well up to the highest energy where abundances are known \\cite{4}.  For energies beyond about 1000~TeV, however, there is no consensus. On the other hand, ground-based air-shower experiments observe cosmic rays with energies up to $\\sim$100~EeV (10$^{20}$~eV). Measurements so far reported \\cite{5} suggest that the  slope of the all-particle spectrum in the energy range of about 100 - 1000~TeV is somewhat flatter than that observed at lower energies, while at higher energies over several times 1000~TeV the energy spectrum  becomes steeper with the slope of about -3.0.  The break in the overall spectrum at around 1000~TeV is often referred to as the ``knee'' in the spectrum.  Clearly, the knee of the primary cosmic ray spectrum has its origin in  the acceleration and propagation of high-energy cosmic rays in our Galaxy. The acceleration model by supernova blast waves leads to the formation of a  power-law spectrum of particle energies with the index of about -2 at sources \\cite{1}, and plausible propagation models of their confinement by galactic magnetic fields and of their eventual escape from our Galaxy can explain well a steeper power-law spectrum than that at the source region \\cite{5}, suggesting a rigidity-dependent bending for different cosmic ray composition. Within the framework of this picture the average mass of primary cosmic rays before the knee should increase with increasing primary energy. In other words, the knee composition becomes heavy dominant as the proton spectrum may first bend at an energy of about 100~TeV, corresponding to a maximum energy gained by shock acceleration at SNR's \\cite{1}.  While the  origin of cosmic rays with energies beyond the knee is still in controversy, observations of cosmic rays in such a high energy region  may naturally stand in need of other acceleration mechanisms \\cite{6,7}  or new cosmic ray sources \\cite{8,9}. Among those, one of the most promising  models may be  that the cosmic rays come from extra-galactic sources such as active galactic nuclei \\cite{9}, though the evidence is far from convincing. However, such an extra-galactic source model should  predict proton-enriched primary composition around and beyond the knee.  Thus,  measurements of the primary cosmic rays around the knee are  very important and its composition is fundamentally input for understanding the particle  acceleration mechanism that pushes cosmic rays  to very high energies. Among various primary particles,  protons hold the key to the situation and its spectrum provides major constraints on the model parameters  of the origin of high-energy cosmic rays. Because of extremely low and steeply decreasing flux at high energies, however, direct measurements of primary proton spectrum on board balloons are still limited in the energy region lower than a few hundred TeV.  In a recent report by the JACEE group \\cite{10} it was concluded that the proton spectrum as well as the helium spectrum are consistent with power laws with no spectral breaks, meaning that there is no bending up to the highest energy they measured (about 800~TeV). However, this is a surmise based on statistically sparse  data, so more studies are required.  On the other hand, most studies on the cosmic ray composition around the knee have been carried out  with  ground-based instruments that can observe the various air-shower parameters. Recently, for example, measurements of muon content in each air-shower \\cite{11} or muons in the deep underground \\cite{12,13}, measurements of the lateral distribution of air shower Cherenkov lights \\cite{14} or the maximum depth of shower development using air Cherenkov telescopes \\cite{15}, and  multiparameter measurements of air showers \\cite{16} have been carried out. However, the  results obtained by these methods have been  derived by  indirect ways that may  strongly rely  on how the observed quantities depend on the composition, on the precision of the measurements,  and on the air-shower and detector simulations as well.  Therefore, the  conclusions sometimes differ with experiments considerably.  Within the ground-based experiments those which set up at higher altitudes are preferable. The reasons include, first, the observation level is close to the maximum of the shower developments induced by cosmic rays with energies around the knee, so that the energy determination is more precise and less dependent upon the unknown composition \\cite{17} ; second, the higher energy flux in the core region of air showers can be observed with emulsion chambers or burst detectors \\cite{18}. High-energy air-shower cores are sensitive to the intensity of protons in the primary cosmic rays and also to the composition around the knee.  A hybrid-experiment of emulsion chamber and air-shower array at high altitude has a great advantage for studying the composition of primary particles at the knee energy region \\cite{19,20}. In a previous paper \\cite{21} we have developed a method to study the primary cosmic ray composition with a hybrid detector of the emulsion chamber and air-shower array based on a Monte Carlo simulation. It is shown there that an artificial neural network (ANN) can be used as a classifier for the species of primary particles since high-energy air-shower cores accompanying air showers are characterized by several parameters and that such a hybrid experiment is powerful enough to select the events induced by protons in the knee energy region. We have applied a three layered feed forward neural network with a back-propagation learning algorithm  to the data  obtained with the Tibet burst detector and the air-shower array \\cite{22}.  Here, we report our study on the primary proton spectrum using the data obtained with the Tibet burst detector and air-shower array. The experiment, including the apparatus, its performance and data selection, is described  in Sec. II.  Air-shower simulations to compare with the experimental data are described in Sec. III.  The ANN used is briefly introduced in Sec. IV. Section V is devoted to the results and discussions and a brief summary is given in Sec. VI. ",
        "conclusions": "\\subsection{Behavior of burst events}  First we discuss the behavior of burst events. In Fig. \\ref{Fig7}, we present the burst size ($\\sum N_{b}$) spectrum observed in our experiment and compare it with the simulation results obtained by three different models.  This figure shows that the CORSIKA+HD and COSMOS+HD models are almost consistent with the experiment.  It is noted that two hadronic interaction models, QGSJET in CORSIKA and quasiscaling  in COSMOS, can fairly well reproduce many data obtained by accelerator and cosmic ray experiments. However, the absolute intensity by the CORSIKA+PD model  gives results about three times as high as that by the HD model. This difference can be mostly attributed to the difference of the proton flux  in both models since  most selected events are induced by protons, in other words, the observed flux of the burst events is very sensitive to the absolute intensity of primary protons.  The distribution of the number of fired burst detectors and the air-shower size spectrum are also shown in Figs. \\ref{Fig8} and \\ref{Fig9}, respectively, where the experimental results  are compared with the simulations obtained by the CORSIKA+HD and COSMOS+HD models. From these comparisons,  we can assure  that almost all behavior of the burst events observed are compatible with a heavy enriched primary composition at energies around the knee. In the previous paper \\cite{24}, we also discussed the detailed features of the burst events whose primary energies are in the knee energy region, say higher than $10^3$ TeV and reached the same conclusion. Based on these results, in the following we try to obtain the primary proton spectrum from the observed burst events using the ANN discussed above.  \\subsection{Selection of proton-induced events with ANN}  We trained and tested the ANN using the simulation events obtained from the CORSIKA+HD model, since this model can explain well the behavior of the observed burst events as discussed above.  For this, $ 2 \\times 10^4$ events by protons and $2 \\times 10^4$  events by other nuclei were used as the training data set and the same number of events as the test data set. The target value for protons was put to 0 and for other nuclei to 1. A strict middle-point condition was used to measure the classification ability of the network, that is, when the ANN output is smaller than 0.5, the event is assigned as a proton origin, while when the ANN output is larger than 0.5, the event is considered to be an origin of other nuclei. The fraction of correct classifications as a function of the number of epochs of the weight updating is shown in Fig. \\ref{Fig10}. The dashed and solid lines are for the training and the test data sets, respectively. The learning of the network becomes very stable after 300 epochs and the change of the weights is small. It is found that the network is able to correctly select 75.7 \\% of the two kinds of events we input. The wrong classifications are approximately equally distributed among those two.  As discussed in Sec. III, different primary models give different fractions of the events produced by each species of primary particles, thus we need to use different values for cutting the network output in order to reduce the wrongly classified events to the desired amount.  The ANN output distribution of the test events in the case of the HD model is presented in Fig. \\ref{Fig11}. It is seen that the proton-induced events  can be clearly separated from others with a  proper cut value  of the ANN output. Shown in Fig. \\ref{Fig12} are the ratio of $N(<y_{out})/N_{total}$ and the selection efficiency of proton events as a function of the cut $y_{out}$ in the network output, where $N(<y_{out})$ is the number of events with the cut $<y_{out}$ and $N_{total}$ is the total number of test events used. Here we examined three cases : (1) both  training and test data sets consist of HD events  ; (2) both training and test data sets consist of PD events ; and (3) training data set consists of PD events while the test data set consists of HD events. As seen in Fig. \\ref{Fig12}, it is confirmed that the ANN training is almost independent upon the primary composition and  the selection efficiency of proton-induced events is about 90 \\%  when the cut value of ANN output  $y_{out}$  is set to 0.15.  Using the ANN trained by the CORSIKA + HD events, we selected 820 candidate events induced by protons out of 4274 observed events.  \\subsection{Proton spectrum}  The primary energy of each event can be estimated from a value of $\\sum N_{b}$ observed with the burst detector array. Shown  in Fig. \\ref{Fig13} is the scatter plots between the burst size $\\sum N_{b}$ and the primary energy $E_0$ of proton-induced events which were selected from the data set of the CORSIKA+HD events by setting the ANN output value to 0.15. A  fairly good correlation  between $E_0$ and $\\sum N_{b}$, as seen in this figure, enables us to estimate the primary energies  of observed burst events with small ambiguity. A good correlation is also found between the air-shower size $N_e$ and the primary energy $E_0$ [ it can be expressed as $E_0 \\simeq 2.5 (GeV) \\times N_e$ for $N_e >10^5$ ], and it is checked that both give almost the same values on the primary energy, while the size estimation becomes worse for  air-shower events with $N_e < 10^5$ because of small number of detectors to be used for fitting. In the present analysis, then,  we used the burst sizes for the estimation of primary energies.  The systematic error  on the primary energy estimation was evaluated by the Monte Carlo simulation and was estimated to be about 30 \\% at energies around 500 TeV.  Shown in Fig. \\ref{Fig14} is the effective collecting area of the burst array calculated for primary protons incident at the top of the atmosphere isotropically within the zenith angle smaller than $45^\\circ$. The burst events satisfying the selection criteria discussed in Sec. II B and accompanying air showers with $N_e >10^{4.5}$ are selected in this calculation.  In Fig. \\ref{Fig15}, we present the primary proton spectrum obtained from the burst events, which were selected using the ANN trained by the CORSIKA+HD events. To examine whether or not the result depends on the primary composition model used, the following check was done. For this, first we trained the ANN by using the events obtained from the CORSIKA+PD model. Then we  selected the proton-induced events from the experimental data to obtain the proton spectrum.  The primary proton spectrum, thus obtained, is also shown in Fig. \\ref{Fig15} to compare with that obtained from the HD composition. Note that in spite of a big difference between the HD and PD models on the  power index and absolute flux of proton component, both results give the same  spectrum for protons, as seen in Fig. \\ref{Fig15}. Hence,  we may say that the primary proton spectrum obtained from our experiment using the ANN method is almost independent of the primary composition model used in the simulation, and it is estimated that the ANN can select the proton-induced events from others with an uncertainty of about 10 \\% under our experimental condition.  The proton spectrum obtained from this experiment  can be represented by the power-law fit as shown in Fig. \\ref{Fig15}.  The power indexes are estimated to be $-2.97\\pm0.06$ and $-2.99\\pm0.06$ for the spectra obtained using the ANN trained by the CORSIKA+HD and CORSIKA+PD events, respectively, where errors quoted  are statistical ones.  It is known that the interpretation of air shower measurements depends on the model of the shower development in the atmosphere. The largest uncertainties may originate from the hadronic interaction which is not well known at very high energies as well as small momentum transfers. Thus, using different hadronic interactions may lead to different predictions for some air-shower observables. No drastic changes, however, have been observed on the hadronic interactions at least up to p$\\bar{\\rm p}$ collider energies, corresponding to $\\sim$ 1000 TeV in the laboratory system. Also, it is noted that the air shower size observed at high altitude weakly depends on the model, while the difference becomes larger near sea level \\cite{24}.  Furthermore, we examined in the previous paper \\cite{24} that both  CORSIKA (QGSJET) and COSMOS simulation  codes give almost the same results on the behavior of the burst events observed with our detector, resulting in that the spectrum obtained here does not depend on the simulation code we used. Consequently, we estimate that the systematic errors on the proton flux are smaller than 40 \\% in this experiment.  Direct measurements of the proton spectrum in the energy region up to about  100 TeV \\cite{17,29,30}, while statistics is still scanty, may suggest a slightly flat spectrum with the slope of  -2.5 $-$ -2.7. When both results are combined, we may say that the proton spectrum changes its slope at energy around 100 TeV. This may be in favor of shock acceleration at SNRs  and when we compared this with the all-particle spectrum obtained by the Tibet air-shower array \\cite{17}, the primary composition becomes heavy dominant at energies around the knee.  \\subsection{On the helium spectrum}  Our experiment is also sensitive to the helium component. In order to estimate the primary  helium spectrum  from our experimental data, we adopted the following method. Monte Carlo events induced by protons and helium nuclei are first gathered as one  group and its ANN target output is assigned to be 0, while the events induced by other nuclei belong to another group with the ANN target output being 1.  After training the ANN with the Monte Carlo events, then the proton+helium events were selected with a proper cut of the ANN output as described in our previous paper \\cite{21}. The $\\sum N_{b} $ spectrum of the proton+helium events minus that of the proton events should  give  the pure helium spectrum.  Calculating the effective area for observing the helium-induced events  with our burst detectors and also using a relation between the  burst size $\\sum N_{b}$ and the primary helium energy  calculated by the CORSIKA+HD model, we obtained the energy spectrum of primary helium nuclei in the energy region above about 100~TeV/n, which is shown in Fig. \\ref{Fig16}. The spectrum obtained based on the CORSIKA+PD model is also shown in the same figure to compare with  each other. Our data is compatible with those extrapolated from the RUNJOB \\cite{30} and MUBEE \\cite{31} data, and the spectrum is not so hard as the JACEE data \\cite{17} at high energies."
    },
    "0010/astro-ph0010637_arXiv.txt": {
        "abstract": "Near-infrared photometry and narrow/broad-band imaging of the massive and luminous young stellar object IRAS 05361+3539 are presented. Imaging observations were made at Mt. Abu while the photometric data were taken from the 2MASS. From the color-color and color-magnitude diagrams, we identified several sources of faint Class II type and about six Class I type in the parent molecular cloud complex. The IRAS 05361+3539 itself was seen to be a Class I object and our images in Br$\\gamma$ and H$_{2}$ lines show jets/outflows from this object. The jet/outflow matches with the axis of CO outflow detected earlier. The near-infrared and the IRAS far-infrared flux distribution suggests a possible accretion disk with dust temperatures between 80 to 800K and extent of several tens to hundreds of AU. A possible FU Orionis type of source was detected in the cluster. ",
        "introduction": " ",
        "conclusions": ""
    },
    "0010/astro-ph0010401_arXiv.txt": {
        "abstract": "Hierarchical galaxy formation models make specific predictions concerning the relative assembly rates and star formation histories of spiral bulges and field ellipticals. Independently of the cosmological model and initial power spectrum of fluctuations, at all epochs the stellar populations in spiral bulges should be older and redder than those in typical ellipticals selected at the same redshift. To test this simple prediction, we analyze the internal optical colors of a complete sample of \\I $<$24 mag early-type and spiral galaxies from the Northern and Southern Hubble Deep Fields (HDF). The subset of galaxies in the Northern HDF are also investigated in the near-infrared using NICMOS photometry. We compare the central (inner 5\\% radius) colors of those spirals with clearly visible bulges with the integrated colors of ellipticals in our sample. Comparisons are possible to a redshift $z\\simeq$1 at which point well-defined bulges become difficult to locate. The reliability of determining bulge colors using central apertures is tested by considering the homogeneity of the pixel-by-pixel colors for typical cases and through comparisons based on the simulated appearance at moderate redshift of the local sample of de~Jong. We show via these tests and by selecting HDF subsets chosen according to inclination that disk contamination effects should be minimal. While spiral bulges are systematically redder in their optical colors than their associated disks at all redshifts,  we find that the majority are significantly bluer than the red locus occupied by most field ellipticals at similar redshifts. In the near-infrared, similar trends are found at redshifts $z<0.6$, but at higher redshifts some bulges as red as the reddest ellipticals are found. We conclude that a significant rejuvenation may have occurred in the inner stellar populations of many spiral galaxies, particularly those at intermediate redshifts. We examine the optical and near-infrared colors of the HDF bulges in the context of models which include the effects of secondary star formation superimposed upon pre-existing old populations and conclude the data is best fit when this secondary activity is burst-like. We discuss the consequences for models of secular evolution in disks should these bursts have been particularly prevalent at $z\\simeq$0.6 as the limited HDF data seems to imply. ",
        "introduction": "The origin of galactic bulges is an important and unsolved issue in our understanding of galaxy evolution. The hitherto established view is that galactic bulges form at high redshifts through early dissipationless collapse (Eggen, Lynden-Bell, \\& Sandage 1962, hereafter ELS). This is based principally on the evidence for old stellar populations concentrated in the bulge of our own Galaxy and contrasts with more recently-developed hierarchical galaxy formation models (Kauffmann et al 1993, Baugh et al 1998) where elliptical galaxies form from the merger of early disk systems which can, in turn, continue to accrete gas to form a two component spiral galaxy.  A robust prediction of all hierarchical models is that spiral bulges should, on average, contain older stars than their associated disks which form by subsequent accretion. Moreover, statistically at a given redshift, bulges should be older and redder than field ellipticals which predominantly form from the merger of previously created spirals. Importantly, these conclusions should remain valid regardless of the particular cosmological model or initial power spectrum which governs the rate of assembly of massive galaxies. As such, a comparison of the relative colors of ellipticals and spiral bulges offers a remarkably simple, but powerful, test of hierarchical assembly models.  A third alternative for the origin of stellar bulges proposes their manufacture via various instabilities of pre-existing disks (see the recent review by Combes 1999). Of greatest interest here is the suggestion that bulges may form through secular processes (those whose timescale are longer than the dynamical time), mostly through interactions and the evolution of galactic bars. Galactic bars can be transformed into bulges through a vertical heating of the inner disk via resonant scattering of stellar orbits by the bar potential. Numerical work also suggests that bulges may form via bar destruction through the growth and subsequent collapse of bar instabilities in cold rotating disks. Evidence that such secular activity may play an important role in the formation of bars is growing (Kormendy 1993; Kuijken \\& Merrifield 1995; Norman, Sellwood, \\& Hasan 1996) and there is rather good dynamical evidence that most bulges in very late-type galaxies are not true bulges, but rather vertically heated central disks (Kormendy 1993). This suggests that the relative importance of the three bulge formation processes may be a strong function of Hubble type.  Most investigators of bulge formation have sought to infer their origin by studying the stellar populations or dynamical properties of local galaxies. Of particular importance in this regard are extinction-free near-infrared imaging observations of the Milky Way bulge (Dwek et al 1995) and resolved studies of associated stellar populations (eg. Rich 1997). These studies and those possible for nearby galaxies have emphasized a much greater diversity in color and stellar history than had hitherto been believed (Wyse et al 1997) although the effects of dust and other factors remain unresolved (Peletier et al 1999).  Through the Hubble Deep Field (HDF) campaigns (Williams et al 1996, 1999), high resolution imaging data is now available with sufficient signal-to-noise to enable the direct study of bulge growth and evolution {\\em in situ} to redshifts $z \\sim 1$. Given the above discussion, it is clear that a detailed observational comparison of the resolved stellar populations of field spirals and ellipticals as a function of redshift will offer important insight into some of the most fundamental aspects of galaxy formation. Indeed, only via intermediate redshift data can the epoch corresponding to the formation of the bulge component be determined independently of the age of its stars.  In an earlier paper (Abraham et al. 1999a) we investigated the importance of secular processes in forming bulges by determining the proportion of barred spiral galaxies in the Hubble Deep Fields as a function of redshift, finding a tentative decline in the fraction of barred galaxies beyond redshift $z\\simeq0.5$. The physical mechanisms responsible for such a deficit of barred spirals at high redshift remain unclear. Possibilities include dynamically hotter (or increasingly dark-matter dominated) disks or an enhanced efficiency in bar destruction at high redshift. Regardless, the absence of bars at high redshift does seem to pose an observational challenge for models where bulges form via secular activity initiated by bars. Such activities may be episodic (Sellwood 1999), or confined to low-luminosity late-type spirals (Kormendy 1993) which are too faint at high redshifts to have been included in our earlier investigation.  In a precursor of this study (Abraham et al. 1999b), the dispersion in the internal colors of a sample of high-redshift HDF spirals with redshifts was used to investigate the relative colors and star-formation histories of bulges and disks at high redshifts. This work concluded that bulges are redder and older than their surrounding disks in essentially all luminous high-redshift spirals. However, while bulges appear redder than their surrounding disks out to high redshifts, no investigation of the relative colors of bulges and ellipticals as a function of redshift was undertaken at that time.  Such a comparison forms the basis of this paper. Here we continue our HDF investigations concerning the origin of galactic bulges by measuring the relative colors of spiral bulges and ellipticals as a function of photometric and spectroscopic redshift. Using automated morphological classifications, our study includes a total sample of 95 spirals and 60 early-type galaxies with \\I $<$24 mag in the Northern and Southern Hubble Deep Fields. An additional aspect of this work is the augmentation of our optical data by near-infrared NICMOS observations (Dickinson 2000) of a subset of our data, extending coverage into the $J_{F110W}$ and $H_{F160W}$ photometric bands in the Northern Hubble Deep Field. These infrared observations are neither as deep nor as well-sampled as the optical data in the HDF-N, but provide a number of interesting constraints on the conclusions derived from optical photometry alone.  A plan of this paper follows. In \\S2 we describe our overall sample selection and the techniques used to assign redshifts to these galaxies. In \\S3 we describe our technique for estimating the relative colors of ellipticals and bulges via the measurement of a central aperture \\VI color. We quantify the uncertainty in these estimates for the bulges in our sample by examining individual cases and applying a similar technique to the local spiral galaxy sample of de~Jong (1996). We discuss the bulge-elliptical comparison in the context of hierarchical models in \\S4 and summarize our main conclusions in \\S5. ",
        "conclusions": "A robust prediction of hierarchical scenarios that associates morphological transformations with mergers and disk growth is that bulges must be statistically redder and older than ellipticals. In order to test this prediction, we have collated a magnitude limited sample of spirals and ellipticals from both Hubble Deep Fields and compared the aperture-based \\VI color-redshift relation for the central bulges of spirals selected in various ways with those of field ellipticals.  In detail we find the following:  \\begin{itemize}  \\item At $z\\lesssim0.6$, where we demonstrate that contamination from underlying disk light is minimal, bulges are significantly bluer than elliptical galaxies. Very few bulges follow traditionally-accepted passive evolutionary tracks. At optical wavelengths most bulges are 0.3-1.0 mag {\\it bluer} than ellipticals at comparable redshifts, with a significant color dispersion at a given redshift, in marked contrast to the expectations of recent semi-analytic predictions. Statistically speaking, bulges remain optically bluer than ellipticals at all redshifts $z\\gtrsim 0.6$ to the limits of our data, and contamination by bluer disk light and/or aperture mismatches is unlikely to explain the bulk of this effect.  \\item Although at first sight our data suggest that bulges are recently formed, we show that secondary star formation involving bursts of 15\\% by mass superimposed upon a pre-existing old stellar population can readily reproduce the color dispersion observed. We discuss two general scenarios: short-lived bursts and extended events possibly associated with continuous infall of gas.  \\item Near-infrared NICMOS observations of a sub-sample of our dataset agree with the optical trends at $z\\lesssim 0.6$, but at higher redshifts most infrared bulges appear to be as red as most ellipticals. Model comparison with the optical trends suggests that the duty cycle for the optical blueing of high-redshift bulges is quite short. Most importantly, however, our data firmly rule out the traditional picture whereby bulges form at high redshift and evolve passively as miniature ellipticals. Although somewhat speculative at this stage, the divergence observed in the IR properties of bulges and ellipticals at $z\\sim 0.6$ may suggest that other physical processes become important in forming galactic bulges, for example secular evolution in stellar disks linked to the formation of galactic bars.  \\item Much work needs to be done at both high and low redshifts in order to establish the star-formation history of bulges in various galaxy types and their environments. It is clear that such studies are central to testing hierarchical models for the morphological evolution of galaxies. In this paper we have shown significant differences between the distribution of colors in  bulges and ellipticals which either seriously challenge our theoretical views on bulge formation or, more likely, reveal the necessity for including additional physics into these models, corresponding to periods of extended star formation in the inner regions of spiral galaxies.  \\end{itemize}"
    },
    "0010/astro-ph0010493_arXiv.txt": {
        "abstract": "Diffferentially rotating stars can support significantly more mass in equilibrium than nonrotating or uniformly rotating stars, according to general relativity. The remnant of a binary neutron star merger may give rise to such a ``hypermassive'' object. While such a star may be dynamically stable against gravitational collapse and bar formation, the radial stabilization due to differential rotation is likely to be temporary. Magnetic braking and viscosity combine to drive the star to uniform rotation, even if the seed magnetic field and the viscosity are small. This process inevitably leads to delayed collapse, which  will be accompanied by a delayed gravitational wave burst and, possibly, a gamma-ray burst.  We provide a simple, Newtonian, MHD calculation of the braking of differential rotation by magnetic fields and viscosity. The star is idealized as a differentially rotating, infinite cylinder consisting of a homogeneous, incompressible conducting gas. We solve analytically the simplest case in which the gas has no viscosity and the star resides in an exterior vacuum. We treat numerically cases in which the gas has internal viscosity and the star is embedded in an exterior, low-density, conducting medium. Our evolution calculations are presented to stimulate more realistic MHD simulations in full $3+1$ general relativity. They serve to identify some of the key physical and numerical parameters, scaling behavior and competing timescales that characterize this important process. ",
        "introduction": "Neutron stars may form with appreciable differential rotation. The merger of a binary neutron star system, for example, will form a differentially rotating remnant, an outcome considerably enhanced by the liklihood that the neutron stars are nearly irrotational before coalescence (Bildsten and Cutler 1992; Kochanek 1992; Lai, Rasio \\& Shapiro 1993). The formation of a differentially rotating remnant following binary merger was first demonstrated in Newtonian hydrodynamic simulations (see e.g., Rasio and Shapiro 1992,1994,1999) and dramatically confirmed in fully relativistic simulations (Shibata \\& Uryu 2000), where the remnant's core is found to rotate appreciably faster than its envelope. Core collapse in a supernova may also result in a differentially rotating neutron star, even if the rotation of the progenitor at the onset of collapse is only moderately rapid and almost uniform (see, eg, Zwerger \\& M\\\"{u}ller 1997; Rampp, M\\\"{u}ller \\& Ruffert 1998, and references therein). Conservation of angular momentum during collapse causes the ratio $\\beta = T/|W|$, where $T$ is the rotational kinetic energy and $W$ is the gravitational potential energy, to grow like $R^{-1}$, where $R$ is the equatorial radius. Thus $\\beta$ can increase by several orders of magnitude during the implosion. Since uniformly rotating, compressible stars can only support very small values of $\\beta$ in equilibrium without shedding mass at the equator (see, e.g., Tassoul 1978, Shapiro \\& Teukolsky 1983), collapsed cores of rapidly rotating progenitors must acquire some degree of differential rotation at birth as they settle into equilibrium. Similar arguments hold for accretion-induced collapse of white dwarfs to neutron stars.  Differentially rotating neutron stars can support significantly more rest mass than their nonrotating or uniformly rotating counterparts. Baumgarte, Shapiro \\& Shibata (2000; hereafter BSS), have recently constructed relativistic equilibrium models of differentially rotating ``hypermassive'' polytropes. These configurations have rest masses that exceed both the maximum rest mass of nonrotating spherical stars (the TOV limit) and uniformly rotating stars at the mass-shedding limit (``supramassive stars''), all with the same polytropic index. BSS performed dynamical simulations in full general relativity to demonstrate that hypermassive stars can be constructed that are dynamically stable against radial collapse and nonradial bar formation (see also Shibata, Baumgarte \\& Shapiro 2000). The fully relativistic binary coalescence calculation of Shibata \\& Uryu (2000) reveals that such a dynamically stable hypermassive star can arise following binary merger. BSS have argued that the dynamical stabilization of a hypermassive remnant by differential rotation may lead to delayed collapse and a delayed gravitational wave burst. The reason is that the stabilization due to differential rotation, although expected to last for many dynamical timescales (i.e. many milliseconds), will ultimately be destroyed by magnetic braking and/or viscosity. These processes drive the star to uniform rotation, which cannot support the excess mass, and will lead to catastrophic collapse, possibly accompanied by some mass loss.  The purpose of this paper is to provide a time-dependent calculation of the damping of differential rotation  by magnetic braking and viscosity in a rapidly spinning star. The goal is to construct an exact, albeit idealized, Newtonian model which readily reveals some of the important physical effects, together with their timescales and scaling behavior. By adopting a sufficiently simple, laminar model with a high degree of spatial symmetry, we are able to perform a tractable analysis of the evolution from first principles and without approximation. Our motivation for this simplified approach is to help identify the minimum ingredients for a more realistic, future numerical calculation in full general relativity. We wish to demonstrate explicitly why it is essential to include magnetic fields in fully relativistic simulations of the late stages of coalescing binary neutron stars and other dynamical scenarios where differential rotation naturally arises and significantly influences the structure and stability of the final configuration. The results have important implications for the detection of gravitational wave and gamma-ray bursts (BSS; Spruit 1999a).  Our discussion is by no means the first, (see, e.g., Spitzer 1978 for discussion and references on magnetic braking of angular momentum in interstellar clouds; see Spruit (1999b) and references therein for a discussion of differential rotation and the evolution of initially weak magnetic fields in stellar interiors), and surely will not be the last on this nontrivial topic, especially given the simplified nature of our model. However, the importance of magnetic braking has been pointed out only recently in the context of binary neutron star coalescence, differential rotation of the remnant and the emission of gravitational waves following delayed collapse (BSS). Presumably the most effective way to stimulate numerical relativists to include magnetic fields in $3+1$ fully relativistic hydrodynamics codes now under development is to provide a concrete calculation demonstrating their physical importance, which we do below.  In Section 2 we describe our basic model. In Section 3 we set out the fundamental equations and put them in a convenient nondimensional form. In Section 4 we summarize solutions for three distinct cases, depending on whether the stellar exterior region is a pure vacuum or filled with conducting plasma, and whether or not we include viscosity. In the first case (vacuum exterior, no viscosity) our treatment is entirely analytic, while in the other two the analyses are numerical. In Section 5 we restore dimensions and evaluate our key results for the case of a differentially rotating remnant of a binary neutron star merger. Here we summarize the implications of our calculations to future observations and numerical simulations. Appendicies A and C summarize details of our numerical calculation, while Appendix B provides a proof that the lowest energy state for our adopted rotating stellar models is the state of uniform rotation, for fixed angular momentum. ",
        "conclusions": "The presence of a seed poloidal magnetic field in a differentially rotating star may be sufficient to drive the star to a state of uniform rotation. Differential rotation generates toroidal Alfv\\'{e}n waves through the star and these waves can convert an appreciable fraction of the kinetic energy in differential motion into magnetic field energy. Even in the absence of internal dissipation or Poynting energy loss to an external plasma, the conversion of rotational to magnetic energy by these waves can remove a portion of the star's rotational support against gravity, which can upset the equilibrium balance in a stationary configuration.  In the case of a hypermassive neutron star supported against collapse by differential rotation, the generation of Alfv\\'{e}n waves can lead to catastrophic collapse to a black hole, possibly accompanied by mass loss. Such an outcome may characterize the final fate of neutron stars formed in binary mergers (BSS).  Our model calculations help to identify at least five distinct timescales that govern the evolution of a a differentially rotating, equilibrium neutron star containing a seed poloidal magnetic field. We evaluate them here for neutron star parameters consistent with the stable remnant found in the relativisitic simulations of binary mergers by Shibata and Uryu (2000). These parameters are comparable to the values of the hypermassive, differentially rotating equilibrium model constructed and evolved by BSS, who found that their configuration was also dynamically stable. All of these models employ a polytropic equation of state with polytropic index $n=1$, for which there is scale freedom to choose the physical mass. If we take the mass to be $M \\approx 2 \\times 1.5M\\odot \\approx 3 M\\odot$, a reasonable value for a binary remnant, then the equatorial radius of the star is $R \\approx 5M \\approx 20 {\\rm km}$ and the central angular velocity is $\\Omega_0 \\approx 0.3/M \\approx 2 \\times 10^4 {\\rm Hz}$, a factor of three times higher than the angular velocity at the equator.   The dynamical timescale associated with the star is given by \\begin{equation} t_{\\rm dyn} = \\left( {R^3 \\over M} \\right)^{1/2} \\approx 0.15 \\left({R \\over 20 \\mbox{km}}\\right)^{3/2} \\left({M \\over 3 M_{\\odot}}\\right)^{-1/2} \\mbox{ms}, \\label{thnine} \\end{equation} the time it takes the binary remnant to achieve equilibrium following coalescence. This is also the time that it takes the star, once it is driven out--of--equilibrium by magnetic braking, to undergo collapse. The central rotation period of the remnant is \\begin{equation} t_{\\rm rot} = {2\\pi \\over \\Omega_0} \\approx 0.3 \\left({R \\over 20 \\mbox{km}}\\right)^{3/2} \\left({M \\over 3 M_{\\odot}}\\right)^{-1/2} \\mbox{ms}, \\label{for} \\end{equation} while the period at the equator is about three times longer. The timescale for magnetic braking of differential rotation by Alfv\\'{e}n waves is given by \\begin{equation} t_ {\\rm A} = {R \\over v_{\\rm A}}\\approx 10^2 \\left({B_0 \\over 10^{12} \\mbox{G}}\\right)^{-1} \\left({R \\over 20 \\mbox{km}}\\right)^{-1/2} \\left({M \\over 3 M_{\\odot}}\\right)^{1/2} \\mbox{s}. \\label{forone} \\end{equation} On this timescale the angular velocity profile in the star is signifcantly altered, and this change can drive the star far out of equilibrium whenever the initial rotation supplies an appreciable fraction of the support against gravity. In the case of a hypermassive neutron star, which depends on rapid differential rotation for hydrostatic support, magnetic braking can lead to catastrophic collapse to a Kerr black hole, possibly accompanied by some mass and field ejection. A hypermassive star is a likely outcome of binary neutron star coalescence. The resulting delayed collapse following merger may generate a brief secondary burst of gravitational waves lasting a time $t_{\\rm dyn}$. If the time of the coalescence can be inferred from the initial wave burst signal, then the measurement of this delay should provide an estimate of the strength of the poloidal magnetic field in the interior of the merged neutron star (BSS).  Neutron stars below the supramassive limit do not require differential rotation for hydrostatic support. Such stars have masses at most $~ 20\\%$ larger than the nonrotating, spherical TOV limit (see, e.g, Butterworth \\& Ipser (1975); Friedman, Ipser \\& Parker (1986); Komatsu, Eriguchi and Hachisu (1989); Cook, Shapiro \\& Teukolsky 1992,1994; Salgado et al. 1994). \\footnote{The most recent equations of state place the maximum gravitational mass of a spherical star in the range $1.8 - 2.2 M_{\\odot}$ (Akmal, Pandharipande \\& Ravenhall 1998), but observations of SN 1987A suggest that it could even be as low as $1.5 M_{\\odot}$ (Brown \\& Bethe 1994)} If such a star is formed with differential rotation, magnetic braking will still drive Alfv\\'{e}n waves in the star and alter the rotation profile on the Alfv\\'{e}n time. While the star may undergo dynamical oscillations, it will not collapse. Instead, viscous dissipation ultimately will drive the star to a new, uniformly rotating equilibrium state on a timescale \\begin{eqnarray} t_{\\nu} &\\!\\! = \\!\\! & {R^2 \\over 8 \\nu} \\\\ & \\!\\! \\approx \\!\\! &  2\\times 10^9 \\left({R \\over 20 \\mbox{km}}\\right)^{23/4}\\left({T \\over 10^9 {\\mbox K}}\\right)^2 \\left({M \\over 3 M_{\\odot}}\\right)^{-5/4} \\mbox{s}  ,  \\nonumber \\label{fortwo} \\end{eqnarray} where $\\nu = 347 \\rho^{9/4} T^{-2} {\\rm cm^2 s^{-1}}$ (Cutler \\& Lindblom 1987).  If a differentially rotating neutron star is immersed in an ambient plasma at formation, Alfv\\'{e}n waves generated in the interior will propagate into the external medium, carrying off energy and angular momentum. According to equation (\\ref{Afour}), everytime a wave propagates to the stellar surface it will transmit a fraction $\\approx 4 (\\rho_{\\rm ex} / \\rho)^{1/2}, ~~ (\\rho_{\\rm ex} / \\rho) \\ll 1$, of its energy to the surrounding plasma, causing the star to spindown in an e-folding time \\begin{eqnarray} t_ {\\rm A,ex} & \\!\\!=\\!\\! & {\\mbox {ln 2} \\over 4} \\left({\\rho \\over \\rho_{\\rm ex}}\\right)^{1/2} t_{\\rm A}  \\\\ &  \\!\\! \\approx \\!\\! & 20 \\left({B_0 \\over 10^{12} \\mbox{G}}\\right)^{-1} \\left({R \\over 20 \\mbox{km}}\\right)^{-1/2} \\left({M \\over 3 M_{\\odot}}\\right)^{1/2} \\left({\\rho \\over \\rho_{\\rm ex}}\\right)^{1/2} \\mbox{s}.  \\nonumber \\label{forthree} \\end{eqnarray} ({\\it cf.} Spitzer 1978, eqn 13.54 and discussion therein for an alternative argument leading to an analogous result). Determining the final spin rate and fate of a newly formed neutron star may thus depend on the nature of its immediate surroundings.  Our naive model calculations do not account for convective instabilities (Pons et al. 1999), which can drive the seed magnetic field to high values greatly exceeding $10^{12}{\\rm G}$ (Duncan \\& Thompson 1992), or other possible MHD instabilities, which may also contribute to the redistribution of angular momentum (Balbus \\& Hawley 1994; Spruit 1999b). Our main goal was to show that even in the simplest laminar description, a differentially rotating neutron star is a transient phenomenon, since magnetic braking and viscosity inevitably will bring a differentially rotating star into uniform rotation (although not before many dynamical timescales have elapsed, according to our timescale estimates above). Thus all radio pulsars are likely to be uniformly rotating. However, differential rotation may characterize a nascent neutron star formed in a supernova, following fallback, or in a merged binary. The braking of this motion may have important consequences for gravitational wave signals and gamma-ray bursts.  More realistic evolutionary calculations of magnetic braking in neutron stars should clarify some of the above issues. One computational subtlety arises from the inequality $t_{\\rm dyn} \\ll t_{\\rm A}$, which holds for seed magnetic fields of small or moderate strength (see equations [\\ref{thnine}] and [\\ref {forone}]). In this regime it may prove too taxing to a relativistic MHD code to evolve a differentially rotating star for the required length of time for magnetic braking to take effect. An obvious solution would be to artificially amplify the initial magnetic field strength, increasing the ratio $t_{\\rm dyn}/t_{\\rm A}$ to a computationally manageable size (while still keeping it much less than unity), and then scale the results for smaller ratios. However, if the energy in the seed magnetic field is an appreciable fraction of the gravitational potential energy, care must be taken to incorporate the magnetic field stresses when constructing the initial hydrostatic equilibrium configuration. Another approach might be to use implicit differencing in the calculation to avoid the Courant stability constraint on the evolution timestep. In addition, one might be able treat {\\it part} of the evolution in the quasistatic approximation, as in a typical stellar evolution code, up to the moment that stable equilibrium can no longer be sustained.  Specifically,  one might evolve the magnetic field and the angular velocity on the Alfv\\'{e}n timescale via the MHD equations, but assume hydrostatic equilibrium is maintained on the dynamical timescale in order to determine the density and pressure profiles at each instant. Our idealized incompressible cylindrical star model, which remains radially static, trivially conforms to this behavior and thereby avoids the mismatch between the dynamical and Alfv\\'{e}n timescales. (An additional motivation for treating this idealized problem was to highlight this feature of the analysis.)  We hope to tackle the more general problem in future computational investigations."
    },
    "0010/astro-ph0010346_arXiv.txt": {
        "abstract": "We calculate accurate solar models and report the detailed time dependences of important solar quantities.  We use helioseismology to constrain the luminosity evolution of the sun and report the discovery of semi-convection in evolved solar models that include diffusion.  In addition, we compare the computed sound speeds with the results of $p$-mode observations by BiSON, GOLF, GONG, LOWL, and MDI instruments. We contrast the neutrino predictions from a set of eight standard-like solar models and four deviant (or deficient) solar models with the results of solar neutrino experiments.  For solar neutrino and for helioseismological applications, we present present-epoch numerical tabulations of characteristics of the standard solar model as a function of solar radius, including the principal physical and composition variables, sound speeds,  neutrino fluxes, and functions needed for calculating solar neutrino oscillations. ",
        "introduction": "\\label{sec:introduction} Why are new calculations of standard solar models of interest? After all, solar models have been used to calculate neutrino fluxes since 1962 (Bahcall, Fowler, Iben, \\& Sears 1963)\\nocite{bahcall63} and solar atmospheres have been used to calculate p-mode oscillation frequencies since 1970 (Ulrich 1970;\\nocite{ulrich70} Leibacher \\& Stein 1971).\\nocite{leibacher71} Over the past four decades, the accuracy with which solar models are calculated has been steadily refined as the result of increased observational and experimental information about the input parameters (such as nuclear reaction rates and the surface of abundances of different elements), more accurate calculations of constituent quantities (such as radiative opacity and equation of state), the inclusion of new physical effects (such as element diffusion), and the development of faster computers and more precise stellar evolution codes.  Solar models nevertheless remain at the frontiers of two different scientific disciplines, solar neutrino studies and helioseismology. In an era in which many major laboratory studies are underway to study neutrino oscillations with the aid of very long baselines, $\\sim 10^3$ km, between accelerator and detector, solar neutrinos have a natural advantage, with a baseline of $10^8$ km (Pontecorvo 1968).\\nocite{pontecorvo68} In addition, solar neutrinos provide unique opportunities for studying the effects of matter upon neutrino propagation, the so-called MSW effect (Wolfenstein 1978; Mikheyev \\& Smirnov 1985),\\nocite{wolf,MiSm} since on their way to terrestrial detectors they pass through large amounts of matter in the sun and, at night, also in the earth .  The connection with ongoing solar neutrino research imposes special requirements on authors carrying out the most detailed solar modeling. Precision comparisons between neutrino measurements and solar predictions are used by many physicists to refine the determination of neutrino parameters and to test different models of neutrino propagation. Since the neutrino experiments and the associated analysis of solar neutrino data are refined at frequent intervals, it is appropriate to reevaluate and refine the solar model predictions as improvements are made in the model input parameters, calculational techniques, and descriptions of the microscopic and macroscopic physics.  In this paper, we provide new information about the total solar neutrino fluxes and the predicted neutrino event rates for a set of standard and non-standard solar models. Using the best-available standard solar model, we also present the calculated radial dependence of the production rate for each of the important solar neutrino fluxes.  We publish for the first time the results of a precision calculation with the standard solar model of the electron density throughout the sun, from the innermost regions of the solar core to the solar atmosphere.  We also present for the first time a detailed calculation of the radial profile of the number density of scatterers of sterile neutrinos. These quantities are important for precision studies of neutrino oscillations using solar neutrinos.  We also provide detailed predictions for the time evolution of some of the important solar characteristics such as the depth and mass of the solar convective zone, the radius and the luminosity of the sun, the central temperature, density, pressure, and hydrogen mass fraction, as well as the temperature, density,  pressure, and radiative opacity at the base of the convective zone.  As far as we know, these are the first detailed results submitted for publication on the time evolution of many of these quantities. Some of the calculated time dependences  may be subject to observational tests.  At the present writing, the sun remains the only main-sequence star for which p-mode oscillations have been robustly detected. Thus only for the sun can one measure precisely tens of thousands of the eigenfrequencies for stellar pressure oscillations. The comparison between the sound speeds and pressures derived from the observed p-mode frequencies and those calculated with standard solar models has provided a host of accurate measurements of the interior of the nearest star.  The solar quantities determined by helioseismology include the sound velocity and density as a function of solar radius, the depth of the convective zone, the interior rotation rate, and the surface helium abundance. The excellent agreement between the helioseismological observations and the solar model calculations has shown that the large discrepancies between solar neutrino measurements and solar model calculations cannot be due to errors in the solar models (cf. Figure~\\ref{fig:diffbp00best}).  In this paper, we present a refined comparison between our best standard solar model and measurements of the solar sound speeds obtained using oscillation data from a number of different sources. We use a combination of data from the LOWL instrument and the BiSON network, two sets of data from the GOLF instrument, as well as data from the GONG network and the MDI instrument.  We describe in the following paragraph the organization of this paper. But, since this paper contains a lot of information on disparate topics, we recommend that the reader first turn to the final section, \\S~\\ref{sec:discussion}, and peruse our summary and discussion of the main new results. Section~\\ref{sec:discussion} may help the reader decide which sections of the paper the reader wants to read (or to skip).  The different sections are written so that they can be read more or less independently.   Section~\\ref{sec:standard} defines the slightly-improved Standard solar model and describes the numerical tables that present details of the contemporary characteristics of the Standard model. Section~\\ref{sec:timedependences} describes the main sequence time dependences of some of the most important characteristics of the Standard model. We present in this section the time evolution of the solar radius and luminosity, the properties of the convective zone, and the physical characteristics of the center of the sun. We discuss solar properties from the initial age main sequence to an age of 8 billion years. Section~\\ref{sec:variantdeviant} defines and compares the physical characteristics of seven variant standard models and four deviant (deficient) solar models, which together with the Standard model make up a set of twelve models whose neutrino fluxes we evaluate in this paper.  We previously used a subset of nine of these models to test the robustness of helioseismological inversions (Basu, Pinsonneault, \\& Bahcall 2000, hereafter BPB2000).\\nocite{basu2000} We discuss in this same section two new standard-like solar models with heavy element to hydrogen ratios that differ slightly from our previously adopted value of $Z/X$. Section~\\ref{sec:neutrinos} discusses solar neutrino physics.  We present the predicted present-day neutrino fluxes for the Standard model and for all of the variant and deviant solar models, as well as the electron number density versus solar radius.  We contrast the predicted neutrino event rates with the results of the chlorine, Kamiokande, GALLEX, SAGE, Super-Kamiokande, and GNO solar neutrino experiments. We also give in this section the calculated time evolution of the most important solar neutrino fluxes.  Section~\\ref {sec:soundspeeds} compares calculated and observed sound speeds. We present the results both on a panoramic scale suitable for discussing the implications for solar neutrino physics and on a zoomed-in scale appropriate for detailed investigations of solar physics.  In \\S~\\ref{sec:hepluscz}, we compare, for all 12 of the solar models discussed in this paper, the calculated values of the surface helium abundance and the depth of the convective zone with the measured values for these quantities.  We summarize and discuss our main results in \\S~\\ref{sec:discussion}.   The interested reader may wish to consult the following works that summarize the solar neutrino aspects of solar models (Bahcall 1989;\\nocite{bahcall89} Bahcall \\& Pinsonneault 1992, 1995;\\nocite{bp92,bp95} Berezinsky, Fiorentini, \\& Lissia 1996;\\nocite{berezin96} Castellani et al. 1997;\\nocite{cast97} Richards et al. 1996;\\nocite{ri96} Turck-Chi\\`eze et al. 1993;\\nocite{turck93} Bahcall, Basu, \\& Pinsonneault 1998\\nocite{ba98}) and the helioseismologic aspects of solar models (Bahcall \\& Ulrich 1988;\\nocite{bu88} Bahcall \\& Pinsonneault 1995;\\nocite{bp95} Christensen-Dalsgaard et al. 1996;\\nocite{jcd96}; Guenther \\& Demarque 1997;\\nocite{guen97} Guzik 1998;\\nocite{guzik98} Turck-Chi\\`eze et al. 1998;\\nocite{turck98} Brun, Turck-Chi\\`eze, \\& Zahn 1999;\\nocite{brun99} Fiorentini \\& Ricci 2000).\\nocite{fior00} ",
        "conclusions": "\\label{sec:discussion}  This paper provides new information about four topics: 1) the characteristics of the Standard solar model at the current epoch (see \\S~\\ref{subsec:discusscurrent} below); 2) time dependences of important characteristics of the Standard solar model (\\S~\\ref{subsec:discusstime}); 3) neutrino fluxes and related quantities for standard and variant solar models (\\S~\\ref{subsec:neutrinocharacteristics}); and 4) measured versus calculated solar sound speeds (\\S~\\ref{subsec:discusssoundspeeds}). Extensive numerical data that are useful for applications are available at http://www.sns.ias.edu/$\\sim$jnb .  Just for fun, we provide our favorite list, our ``top three,'' among the many disparate results presented in this paper. Our top three results are listed in \\S~\\ref{subsec:fun}.  \\subsection{Standard solar model: current epoch} \\label{subsec:discusscurrent} We present detailed numerical tabulations of the computed characteristics of our Standard solar model, which is defined and discussed in \\S~\\ref{sec:standard}.  These tables include, as a function of the solar radius, the enclosed mass fraction, the temperature, mass density, electron number density, the pressure, and the luminosity fraction created in a given spherical shell, as well as the mass fractions of $^1$H, $^3$He, $^4$He, $^7$Be, $^{12}$C, $^{14}$N, and $^{16}$O. Over the years, previous numerical versions of our Standard model have been used for a variety of purposes that range from comparisons with other stellar evolution codes, estimating the importance in the sun of newly considered physical effects, searching for possible instabilities in the sun, comparison with helioseismological measurements, and the calculation of processes (especially the MSW effect) that influence the propagation of solar neutrinos.  In the past, we have published in hard copy form increasingly more detailed and precise numerical tables of the characteristics of the solar interior. The capabilities of current calculations and the requirements of some of the most interesting applications have made complete hard copy publication no longer appropriate. We have therefore limited ourselves in \\S~\\ref{sec:standard} to describing briefly the ingredients we use in calculating the current Standard model. We present the numerical results in exportable data files that are available at http://www.sns.ias.edu/$\\sim$jnb .   \\subsection{Standard solar model: time dependences} \\label{subsec:discusstime}  For the first time in this series of papers, we have focused, especially in \\S~\\ref{sec:timedependences}, on details of the time dependence of important characteristics of the Standard solar model.   The total luminosity in the Standard model increases by $48$\\% from the zero-age main sequence stage to the present epoch. Over the same period, the effective temperature varies by only $\\pm 1.3$\\%.  These predictions constitute constraints on models for the evolution of the earth.  The predicted time evolution of the solar luminosity is robust. Figure~\\ref{fig:lumnormalized} shows that all solar models, even those models with deficient physics that are strongly disfavored by helioseismological measurements, predict essentially the same luminosity evolution. The average rms deviation of the deviant models, the Mixed, No diffusion, and $S_{34} = 0$ models, from the standard solar model luminosity is only $1$\\% over the history of the sun from $1$ Gyr to the current epoch (see \\S~\\ref{subsec:radiusetal} for more details).  Table~\\ref{tab:separations} presents the calculated large and small separations of the $p$-mode frequencies as a function of age for the standard solar model.  We have presented in \\S~\\ref{sec:timedependences} the time evolution of some of the principal physical quantities characterizing the solar core (the central temperature, density, pressure, and hydrogen mass fraction, as well as the fractions of the solar luminosity generated by different nuclear reactions). We also present the evolution of important quantities at the base of the convective zone (radiative opacity, temperature, density, and pressure).  We hope that these data and the scaling relations we have inferred will be sufficient to permit a future physical understanding of the time dependences using analytic and semi-analytic arguments.  We find some simple scaling relations. For example, the solar luminosity, $L_\\odot(t)$, is approximately related to the solar radius, $R_\\odot(t)$, as $L_\\odot(t) \\propto R_\\odot(t)^{2.5}$. The depth of the convective zone, $R({\\rm CZ},t)$, scales as $R({\\rm CZ},t) \\propto R_\\odot(t)$, and the central temperature, $T_c(t)$, shows a similar behavior, $T_c(t) \\propto R_\\odot(t)$.  Moreover, we find that the mass of the convective zone, $M({\\rm CZ},t)$ satisfies $M({\\rm CZ},t) \\propto R_\\odot(t)^{-2}$.  The effective temperature is approximately constant, varying by only $\\pm 0.7\\%$ from a solar age of 2 billion years to 8 billion years.     These results make predictions that are potentially testable.  In principle, the measurement of the luminosity (by astrometry) of a star with the same mass and chemical composition ($Z/X$) as the sun would allow the prediction of the star's effective temperature,   the depth and mass of the convective zone, and the large and small separations of the $p$-mode frequencies (see for example Monteiro, Christensen-Dalsgaard,  \\& Thompson 2000\\nocite{mont00} and Christensen-Dalsgaard 1997\\nocite{jcd97}). In practice, it is difficult to make measurements sufficiently accurately to make possible precise tests of stellar evolution theory.  For an appraisal of both the potential and the difficulty of making such measurements, the reader is referred to the recent papers by Guenther and Demarque (2000)\\nocite{guen00} and Morel, Provost, Lebreton, Th\\'evenin, \\& Berthomieu (2000)\\nocite{morel2000} on the binary pair of approximately solar mass stars, $\\alpha$ Centauri AB ($1.1 M_\\odot$ and $0.9 M_\\odot$).    \\subsection{Neutrino fluxes and related quantities} \\label{subsec:neutrinocharacteristics}   Figure~\\ref{fig:ne} and Table~\\ref{tab:ne} give the electron number density as a function of position in the sun for the Standard solar model.  The distribution of the electron density is required to compute the probability for matter-induced oscillations between active neutrinos.  Similarly, Figure~\\ref{fig:nsterile} and Table~\\ref{tab:nsterile} give the radial distribution of the number density of scatterers of sterile neutrinos, $n_{\\rm sterile}$, in the Standard solar model.  We have not previously published precise values for the electron number density, or of $n_{\\rm sterile}$, in the outer regions of the sun. The outer regions are relevant for large mixing angle neutrino oscillations with relatively low neutrino mass differences ($\\Delta m^2 < 10^{-8} {\\rm eV^2}$).   Table~\\ref{tab:bestestimate} presents the neutrino fluxes and the event rates in the chlorine, gallium, lithium, and electron-scattering neutrino experiments that are predicted by the Standard solar model. These predictions assume that nothing happens to solar neutrinos after they are produced.  The table also gives estimates of the uncertainties in the fluxes and the event rates; \\S~\\ref{subsec:fluxestoday} contains a discussion of the physical origin of the uncertainties, as well as the software used to calculate the asymmetric error estimates.  How do the predictions of solar neutrino event rates compare with experiment? Table~\\ref{tab:numeasurements} compares the predictions of BP2000 with the results of the chlorine, GALLEX + GNO, SAGE, Kamiokande, and Super-Kamiokande solar neutrino experiments. This table assumes nothing happens to the neutrinos after they are created in the sun. The standard predictions differ from the observed rates by many standard deviations. Because of an accidental cancellation, the predicted solar neutrino event rates for BP2000 and BP98 are almost identical (see \\S~\\ref{subsubsec:standardneutrino}).   Table~\\ref{tab:fluxesnine} compares the neutrino fluxes and the experimental event rates for all nine of the solar models whose helioseismological properties were investigated in BPB2000, plus two additional standard-like models considered here which have somewhat different heavy-element to hydrogen ratios.  The seven standard-like models (the first seven models in Table~\\ref{tab:fluxesnine}) all produce essentially the same neutrino predictions; the spreads in the predicted $pp$, $^7$Be, and $^8$B fluxes are $\\pm 0.7$\\%. $\\pm 3$\\%. and $\\pm 6.5$\\% , respectively.  The calculated rates for the seven standard-like solar models have a range of $\\pm 0.45$ SNU for the chlorine experiment and $\\pm 2$ SNU for the gallium experiments.  The estimated total errors from external sources (see Table~\\ref{tab:bestestimate}), such as nuclear cross section measurements and heavy element abundances, are about a factor of three larger than the uncertainties resulting from the solar model calculations.  We have investigated one possible source of systematic errors, the relative weights assigned to different determinations of nuclear fusion cross sections. We calculated the neutrino fluxes and predicted event rates using the NACRE (Angulo et al. 1999)\\nocite{angulo99} fusion cross sections rather than the Adelberger et al. (1998)\\nocite{adel98} cross sections. The NACRE parameters lead to slightly higher predicted event rates in solar neutrino experiments. However, all changes in the neutrino fluxes and event rates between the NACRE-based predictions and the Standard predictions (based upon Adelberger et al. nuclear parameters) are much less than the $1\\sigma$ uncertainties quoted for the Standard model (cf. Table~\\ref{tab:bestestimate} and Table~\\ref{tab:nacre}).  The neutrino event rates predicted by all seven of the standard-like solar models considered here are inconsistent at the $5 \\sigma$ level (combined theoretical and experimental errors) with the results of the two gallium experiments, GALLEX and SAGE, assuming no new physics is occurring. The inconsistency with the chlorine experiment is similar but more complex to specify, since the largest part of the theoretical uncertainty in the calculated standard capture rate is due to the electron-type neutrinos from $^8$B beta-decay.  The fractional uncertainty in the $^8$B flux depends upon the magnitude of the flux created in the solar interior. Moreover, the amount by which this $^8$B flux is reduced depends upon the adopted particle physics scenario.  A similar level of inconsistency persists even for the ad hoc deficient models, such as the $S_{34} = 0$ and Mixed models, that were specially concocted to minimize the discrepancy with the neutrino measurements.  For example, the calculated rates for the Mixed model are $6.15^{+1.0}_{-0.85}$ SNU for the chlorine experiment and $115^{+6.8}_{-5.1}$ SNU for gallium experiments. The Mixed model is $4.2\\sigma$ below the measured chlorine rate and $6.3\\sigma$ below the measured gallium rate. In addition, the deficient models are strongly disfavored by the helioseismology measurements.  Figure~\\ref{fig:fluxesvsage} shows the calculated time dependence of the $pp$, $^7$Be, $^8$B, and $^{13}$N solar neutrino fluxes. At the current epoch, the $pp$ flux is increasing at a rate of $4$\\% per billion years, and the $^7$Be, $^8$B, and $^{13}$N fluxes at the rates of $45$\\%, $90$\\%, and $99$\\% per billion years, respectively. Since the age of the sun is estimated to be uncertain by only $5\\times10^{-3}$ billion years (see the Appendix by Wasserburg in Bahcall \\& Pinsonneault 1995),\\nocite{bp95} the age of the sun does not represent a significant uncertainty for solar neutrino predictions.  \\subsection{Sound speeds} \\label{subsec:discusssoundspeeds}  Figure~\\ref{fig:diffbp00best} shows the excellent agreement between the helioseismologically determined sound speeds and the speeds that are calculated for the Standard solar model. The scale on this figure was chosen so as to highlight the contrast between the excellent agreement found with the Standard model and the two orders of magnitude larger rms difference for a solar model that could reduce significantly the solar neutrino problems. One would expect characteristically a $9$\\% rms difference between the observations and the predictions of solar models that significantly reduce the conflicts between solar model measurements and solar model predictions. Averaged over the entire sun, the rms fractional difference is only $0.10$\\% between the Standard solar model sound speeds and the helioseismologically-determined  sound speeds.  The agreement is even better, $0.06\\%$, in the interior region in which the luminosity and the neutrinos are produced. Table~\\ref{tab:soundspeeds} presents numerical values for the sound speeds predicted by the Standard solar model at representative radial positions in the sun.  All eight of the standard-like solar models considered in this paper give acceptable agreement with the measured depth of the convective zone and the surface helium abundance (see Table~\\ref{tab:yandcz}).  Of the twelve Standard, standard-like,  variant, and deviant solar models considered here, only the No Diffusion and Mixed models disagree strongly with the measured convective zone depth.  There are small but robust discrepancies between the measured and the calculated solar sound speeds. Figure~\\ref{fig:differences6} shows the fractional differences between the Standard model sound speeds and the speeds measured in each of six different determinations (using $p$-mode data from LOWL1, BiSON, GONG, MDI, GOLF, and GOLF2). The vertical scale for Figure~\\ref{fig:differences6} is expanded $21$ times compared to the vertical scale of  Figure~\\ref{fig:diffbp00best}.   As can be seen from Figure~\\ref{fig:differences6}, the differences between observed and measured sound speeds are comparable over much of the sun to the differences between different measurements of the sound speed, but there is a clear discrepancy near the base of the convective zone that is independent of which observational data set is used.  There may also be a less prominent discrepancy near $0.2 R_\\odot$.  The agreement between the BP2000 sound speeds and the measured values is improved over what was found earlier with the BP98 model. This improvement may be seen by comparing Figure~\\ref{fig:differences6} and Figure~\\ref{fig:differences5}. The improvement is striking in the region between $0.3R_\\odot$ and $0.7R_\\odot$ and is due to the correction of an error in the interpolation algorithm for the radiative opacity (cf. discussion in \\S~\\ref{subsec:definitionstandard}).  The Rotation model includes a prescription for element mixing that was designed to explain the depletion of lithium. The calculated differences between the Rotation and the Standard models represent a reasonable upper limit to the effects that rotation, sufficient to explain lithium depletion, might produce. The overall agreement between the sound speeds of the Rotation model and the helioseismologically determined sound speeds is slightly worse, $0.12$\\% rather than $0.10$\\%, than the agreement obtained with the Standard solar model. The differences between the solar neutrino fluxes predicted between the Rotation model and the Standard model is typically about $0.3\\sigma_\\nu$, where $\\sigma_\\nu$ represents the uncertainties given in Table~\\ref{tab:bestestimate} for the Standard model neutrino fluxes.  We conclude that further improvements of the theoretical calculations motivated by refinements of $p-$ mode oscillation measurements are unlikely to significantly affect the calculated solar neutrino fluxes.   \\subsection{Top three results} \\label{subsec:fun}  Here are our favorite three results in this paper.  $\\bullet$ The robust luminosity evolution of the sun (see Figure~\\ref{fig:lumnormalized}).  $\\bullet$ The excellent agreement of the Standard model sound speeds with the measured sound speeds on the scale relevant for solar neutrino discussions (see Figure~\\ref{fig:diffbp00best}).  $\\bullet$ The simple relations as a function of time between the solar radius, the solar luminosity, the depth of the convective core, and the mass of the convective core (see Eq.~\\ref{eqn:lrproportional}--Eq.~\\ref{eqn:tcoverr})."
    },
    "0010/astro-ph0010036_arXiv.txt": {
        "abstract": "This paper reviews the theory of protostellar debris disks.  After a brief introduction to accretion disk physics, I describe coagulation models of planet formation in the outer regions of planetesimal disks. Coagulation models for the Kuiper Belt produce Pluto-sized objects on timescales of 10--40 Myr.  These models yield size distributions which agree with observations of Kuiper Belt objects with red magnitudes, R $\\approx$ 20--27.  Velocity stirring models for other debris disk systems demonstrate that 500 km or larger objects can stir the velocities of small objects up to the shattering limit needed to begin a collisional cascade. ",
        "introduction": "In the past two decades, several remarkable discoveries have radically altered our understanding of planet formation.  In 1983, data from the {\\it Infrared Astronomical Satellite} revealed substantial mid-infrared emission from three A-type main-sequence stars, $\\alpha$ Lyr, $\\alpha$ PsA, and $\\beta$ Pic (Aumann \\etal 1983).  In addition to the flux expected from an A-type stellar photosphere, the observations implied a cold, extended source of radiation.  These data led to a beautiful ground-based image of an edge-on dusty disk surrounding $\\beta$ Pic (Smith \\& Terrile 1984). This image provided dramatic evidence that the building blocks for solar systems exist in disk-like structures around other stars.  More recent images with the {\\it Hubble Space Telescope} and large ground-based telescopes have revealed disks with similar sizes, 100--1000 AU, around many pre-main sequence and main sequence stars (e.g., Jayawardhana \\etal 1998; Koerner \\etal 1998; Greaves \\etal 1998; Augereau \\etal 1999; Schneider \\etal 1999; see also the reviews by Beckwith and Koerner in this volume). As in $\\alpha$ Lyr and $\\beta$ Pic, all of these systems have excess near-infrared or mid-infrared emission compared to a normal stellar photosphere.  More than 50\\% of young stars ($\\sim$ 1 Myr) in nearby star-forming regions also have excess infrared emission, suggesting that most stars are born with disks with sizes and masses at least as large as our solar system (e.g., Lada 1999; Jayawardhana \\etal 1999).  Ground-based radial velocity surveys have recently detected $\\sim$ 50 Jupiter-mass and Saturn-mass planets in close orbits around $\\sim$ 1\\% of nearby solar-type stars (e.g., Latham \\etal 1989; Marcy \\& Butler 1996, 1999; Cochran \\etal 1997; Noyes \\etal 1997; DelFosse \\etal 1998).  Several systems have multiple planets. Although some of these planet candidates may have masses close to the hydrogen-burning limit for main sequence stars,  these discoveries demonstrate that gas giant planets are relatively common in the solar neighborhood. However, no planet has yet been unambiguously detected in a debris disk system.  Rings, warps, and other spatially extended structures in some images suggest the tidal influence of planets embedded in several debris disks (e.g., Kalas \\& Jewitt 1995), but direct detection of planetary motion has proved difficult.  The final interesting discovery links extra-solar disks and planets with the formation and evolution of our solar system.  Recent surveys have detected a substantial population of Kuiper Belt objects (KBOs), small icy bodies outside the orbit of Neptune (Luu \\& Jewitt 1988, 1998; Luu \\etal 1997; Jewitt \\& Luu 1993). For an adopted albedo of 4\\%, known KBOs have radii of 50--500 km. The measured surface density of KBOs on the sky suggests a total mass of $\\sim$ 0.1--0.3 \\mearth (Luu \\& Jewitt 1998; Jewitt \\etal 1998; Trujillo \\etal 2000; Kenyon \\& Windhorst 2000).  With semimajor axes of 40--50 AU and orbital inclinations of 0\\degree--30\\degree, the apparent geometry and mass of the KBO population is similar to results derived for the dusty disks observed in $\\alpha$ Lyr and $\\beta$ Pic, suggesting that phenomena currently observed in extra-solar dusty disks probably occurred early in the formation of our own solar system.  These discoveries have fueled interest in theoretical studies of planet formation in a variety of astrophysical contexts.  This paper reviews some of these studies, with an emphasis on the dynamical evolution of planetary debris disks (for other reviews, see chapters in Mannings, Boss, \\& Russell 2000).  After starting with a short summary of current models for planet formation, I consider in detail several recent numerical calculations of planet formation in debris disks and describe observational tests of these models.  I conclude with a discussion of future prospects for the calculations and with suggestions for observations to test different models of planet formation. ",
        "conclusions": "Understanding whether or not coagulation or dynamical instability produce solar systems similar to those we observe involves a comparison of detailed numerical calculations with high quality observations.  The advent of fast, parallel computers now allows fairly realistic calculations of the complete evolution of part of a solar system; calculations of an entire solar system will be possible in 10 years or less.  Observations with ground-based and space-based telescopes now provide data with sufficient sensitivity and spatial resolution to make direct tests of some predictions of the detailed numerical calculations.  These tests yield a better understanding of the physical processes involved in forming terrestrial planets like the Earth and gas giant planets like Jupiter.  As both the data and models improve, we will have a better grasp of the origin of planetary systems similar to our own.  In addition to the tests described in \\S4, ground-based and space-based observations will yield new tests of planet formation theories. Data from {\\it SIRTF}, {\\it SOFIA}, and other NASA-sponsored missions will result in many new debris disk systems and better estimates for the distribution of lifetimes for these systems.  Comparisons with the results of the coagulation calculations will test theories of the collisional cascade.  Direct detection of more large Kuiper Belt objects will test predictions for the merger population; direct detection of the optical and far-infrared background from KBOs will test predictions for the size distribution of the debris population and a measurement of the albedo of KBOs (Kenyon \\& Windhorst 2000). These comparisons -- combined with new calculations for the Kuiper Belt as in Kenyon \\& Luu (1998, 1999) -- will give us a better understanding of planet formation (for Earth-sized planets) in the outer part of a solar system.  Along with simple observational comparisons of the disk lifetime, inclusion of good models for gas drag, Poynting-Robertson drag, and radiation pressure on grains will allow predictions for the surface brightness profile of solar system disks, derived from the model results in Figure 5.  The {\\it HST} and several ground-based telescopes have derived observed surface brightness profiles; the {\\it NGST} mission will provide many hundreds. Comparisons of observations with coagulation model predictions will allow better constraints on the models and on the physical situation in extra-solar debris disks.  The identification of features in the observed profiles may yield constraints on planetary masses in the disks.  Observations may also help guide theorists in producing better models for gas giant planets.  Current theory has trouble understanding the formation of gas giants at distance of $\\simgreat$ 10 AU (coagulation models) and $\\simgreat$ 20 AU (dynamical instability models) from the central star. Identification of gas giant planets in other solar systems at large distances from their parent star would provide a stern test of either model.  Discovery of gas giants in the weak-emission T Tauri stars (wTTs) might be the sternest test of all. Many wTTs have little or no evidence for emission from disk material at $A \\simless$ 1--10 AU, suggesting that planet formation may be complete in these systems. A gas giant planet discovered beyond 20 AU from the central star in a wTTS would probably be luminous due to its young age and might yield important insights into the formation of gas giant planets.  \\vskip 6ex  I would like to thank Jane Luu.  Without Jane's foresight and persistence, our joint projects on planet formation would never have been possible. I also thank B. Bromley, A. Cameron, F. Franklin, M. Geller, M. Holman, and J. Wood for advice and comments. I also acknowledge generous allotments of computer time on the JPL Cray T3D `Cosmos', the HP Exemplar `Neptune' and the Silicon Graphics Origin-2000 `Alhena' through funding from the NASA Offices of Mission to Planet Earth, Aeronautics, and Space Science."
    },
    "0010/astro-ph0010200_arXiv.txt": {
        "abstract": "The LMC is ideal for studying the co-evolution of planetary nebulae (PNe) and their central stars, in that the debilitating uncertainties of the Galactic PN distance scale and selection biases from attenuation by interstellar dust do not apply.  We present images and analyze slit-less spectra which were obtained in a survey of Large Magellanic Cloud PNe.  These data on 29 targets were obtained with {\\it HST} using the Space Telescope Imaging Spectrograph.  The data permit us to determine the nebular dimensions and morphology in the monochromatic light of several emission lines, including those that have traditionally been used for morphological studies in the Galaxy: H$\\alpha$, [\\ion{N}{2}] 6583~\\AA\\ and [\\ion{O}{3}] 5007~\\AA, plus others of varying ionization, including [\\ion{O}{1}], \\ion{He}{1}, and [\\ion{S}{2}]. Together with the 31 resolved LMC PNe for which monochromatic images exist in the {\\it HST} archive, these data show that the incidence of non-symmetric nebulae, including bipolar nebulae (which is an indicator of Population I ancestry in the Galaxy), is significantly higher than that reported for the Galaxy. The onset of asymmetric features appears even in very young nebulae (with dynamical ages of $\\sim1400$ yr), suggesting that at least the gross features of the nebular morphology may be more closely tied to PN formation, and that subsequent shaping of the expanding envelope by the radiation field and wind from the central star may play the lesser role of amplifying these gross features.  There is some evidence of evolution between two morphological types, in the sense that bipolar core (BC) nebulae may evolve to pure bipolars late in the PN lifetime. ",
        "introduction": " ",
        "conclusions": ""
    },
    "0010/nucl-th0010052_arXiv.txt": {
        "abstract": "We show that the synthesis of carbon and oxygen through the triple-alpha process in red giant stars is extremely sensitive to the fine details of the nucleon-nucleon (N-N) interaction. A $\\pm0.5\\%$ change in the strength of the N-N force would reduce either the carbon or oxygen abundance by as much as a factor of 30--1000. This result may be used to constrain some fundamental parameters of the Standard Model. ",
        "introduction": "Almost all the carbon in the Universe is produced through the triple-alpha process \\cite{Rolfs}. Although the $^8$Be nucleus, formed in the collision of two alpha particles ($^4$He), is unbound, it lives long enough to allow the possibility of capturing a third alpha particle to form stable $^{12}$C. However, in order to produce enough carbon, this second reaction must be resonant \\cite{Hoyle}. The $0^+_2$ state of $^{12}$C, lying at 380 keV, relative to the $3\\alpha$ threshold, therefore plays a key role in the synthesis of carbon. As both steps of the triple-alpha process are governed by narrow resonances, the reaction rate is given as \\cite{Rolfs} \\begin{equation} \\label{alphaa} r_{3\\alpha} \\approx 3^{\\frac{3}{2}} N_{\\alpha}^3 \\left(\\frac{2 \\pi \\hbar^2}{M_{\\alpha} k_{\\rm B} T}\\right)^3 \\frac{\\Gamma_{\\gamma}}{\\hbar} \\exp \\left(- \\frac{E_{\\rm res}}{k_{\\rm B} T} \\right), \\end{equation} where $M_{\\alpha}$ and $N_{\\alpha}$ are the mass and the number density of the alpha particles, respectively, $E_{\\rm res}$ and $\\Gamma_\\gamma$ are the resonance energy and radiative width of the $0^+_2$ state of $^{12}$C, respectively, and $T$ is the stellar plasma temperature. ",
        "conclusions": "One can see in Fig.\\ 2 that a 0.5\\% change in the N-N interaction strength would lead to a Universe which does not contain an appreciable amount of carbon or oxygen. This would make the existence of carbon-based life highly unlikely. This very strong fine-tuning effect gives us a possibility to try to constrain the possible values of some fundamental constants in the Standard Model. For example, if one assumes a pion-exchange model of the strong force, then the a smaller (larger) pion mass means a stronger (weaker) force. More detailed analyses show that our result on the fine-tuning of carbon and oxygen production can be used to give constraints on the possible values of the sum of the light quark masses and the vacuum expectation value of the Higgs field at the 1\\% level \\cite{quark}."
    },
    "0010/astro-ph0010566_arXiv.txt": {
        "abstract": "In order to derive statistical properties of a complete X-ray selected sample of AGN we used the classification spectra of the RASS Selected Areas-North survey to study the luminosities, redshifts , X-ray/visual flux ratios, line widths, and various correlations between observed parameters of all Seyfert galaxies and QSOs in this sample. On the basis of these data we discuss implications for the current AGN models. ",
        "introduction": "In two earlier papers (Zickgraf et al. 1997a, Appenzeller et al. 1998, hereafter referred to as ``Paper I'' and ``the Catalog'', respectively) we described and presented the optical identification of a complete sample of X-ray sources discovered in the ROSAT All-sky Survey (RASS, Voges et al. 1999). A statistical analysis of these identifications (which in the literature are also referred to as the ``RASS Selected Area-North survey'') has been presented by Krautter et al. (1999). Subsamples of the Catalog have been discussed by by Zickgraf et al. (1998b), Mujica et al. (1999), and Appenzeller et al. (2000). The present paper presents a discussion of the Seyfert galaxies and QSOs listed in the Catalog as main optical counterparts of RASS X-ray sources with an Identification Quality Index $Q = 1$ or $Q = 2$ (i.e. reliable identification on the basis of our spectra and/or on the basis of reliable literature data). Since the Catalog lists no Seyfert galaxy and only one QSO with $Q = 3$ (``uncertain identification'') the present paper covers practically all objects of these types in our survey.  Not included in the following discussion are the BL Lac objects (which are discussed elsewhere), the two LINERs, and the 7 AGN for which due the weakness of the lines or inadequate $S/N$ no subclass could be assigned in the Catalog. Also omitted are 3 Seyfert galaxies and one QSO which - although inside the error circle of an X-ray position - were not regarded to be the main source of the observed X-ray radiation since another (in our opinion dominant or more likely optical counterpart was (also) present at the position in question.  As described in Paper I our identifications and classifications were based on low-resolution spectra obtained  with a multi-object spectrometer at a 2.1-m telescope. These spectra were exposed to reach a $S/N$ suitable for a reliable classification and redshift derivation. Originally there were no plans for a quantitative evaluation of these spectra. However, after photometric calibration the great majority of the spectra turned out to be of sufficient quality to measure the basic properties of the AGN spectra, such as line strengths and line widths. Therefore, we made use of the opportunity provided by these data to derive statistical information on a well defined complete sample of X-ray selected AGN.  Throughout this paper Seyfert 1 galaxies and QSOs will be discussed jointly since - according to our classification criteria - these two classes are distinguished only by being below or above a certain (arbitrarily chosen) luminosity level. Included in this group are all AGN with permitted emission lines (or components of the permitted emission lines) measurably broader than the forbidden lines and/or showing \\FeII\\ emission. Hence Seyfert 1.5-1.9 and NLS1 galaxies (as defined by V\\'eron-Cetty and V\\'eron, 2000, following Osterbrock and Pogge, 1985) are part of our Seyfert 1 and QSO sample. The (relatively few) Seyfert 2s in the Catalog are discussed separately in Section 3.  In order to keep the classification criteria uniform for the whole sample, the [\\OIII ]5007\\AA /H$\\beta $ flux ratio (which is measurable in only part of our spectra) was not used as a classification criterion to discriminate between Seyfert 1s and 2s (Osterbrock and Pogge, 1985, Laor, 2000). However, in all spectra containing the [\\OIII ]/H$\\beta $ region, the corresponding flux ratio was determined and compared to the critical value 3. While all galaxies classified as Seyfert 1s from their line profiles showed (as expected) [\\OIII ]/H$\\beta <3$, we found (as described in Section 3) 6 objects classified as Seyfert 2s with [\\OIII ]/H$\\beta <3$. Because of this small number the classification of these objects has no influence on the statistical conclusions derived for the Seyfert 1s and QSOs. However, as discussed in Section 3,  the much smaller Seyfert 2 sample is affected if the [\\OIII ]/H$\\beta $ flux ratio is used as classification criterion.  Not discussed in the present paper are AGN number counts, $\\log N - \\log S$ distributions, and the implications of our data for the unresolved X-ray background, since the corresponding results have already been presented by Krautter et al. (1999) and Miyaji et al. (2000). Instead, the present paper will concentrate on conclusions concerning the AGN physics. ",
        "conclusions": ""
    },
    "0010/astro-ph0010085_arXiv.txt": {
        "abstract": "Motivated by observations of supernova SN\\,1987A, various authors have simulated Rayleigh--Taylor (RT) instabilities in the envelopes of core collapse supernovae (for a review, see \\cite{Mueller98}). The non-radial motion found in these simulations qualitatively agreed with observations in SN\\,1987A, but failed to explain the extent of mixing of newly synthesized $^{56}$Ni quantitatively.  Here we present results of a 2D hydrodynamic simulation which re--addresses this failure and covers the entire evolution of the first 5 hours after core bounce. ",
        "introduction": "Our simulation is split into two stages.  The early evolution ($t < 1$\\,s) is computed with the {\\sc HERAKLES} code (Plewa \\& M\\\"uller, in preparation) which solves the multidimensional hydrodynamic equations using the direct Eulerian version of the Piecewise Parabolic Method \\cite{CW84} and which incorporates the physics (equation of state, neutrino source terms and light bulb) described in the simulations of neutrino driven supernovae by \\cite{JM96} with the following modifications. General relativistic corrections are added to the gravitational potential.  A 14-isotope network is implemented to compute the explosive nucleosynthesis including the 13 $\\alpha$-nuclei from $\\He$ to $\\Ni$ and a tracer nucleus which is used to keep track of the neutronization of the material.  The 2D spherical grid consists of 400 radial zones ($3.17 \\times 10^{6}\\,{\\rm cm} \\leq r \\leq 1.7\\times10^9$\\,{\\rm cm}) and 192 angular zones ($0 \\leq \\theta \\leq \\pi$).  The initial data is a 15\\,$\\Msol$ progenitor model \\cite{WPE88} which has been collapsed \\cite{Bruenn93}.  The following set of neutrino parameters is adopted (cf. \\cite{JM96}): $L_{\\nu_e}^0 = 3.094 \\times 10^{52}\\,{\\rm erg/s} , L_{\\nu_x}^0 = 2.613 \\times 10^{52}\\,{\\rm erg/s}, \\Delta Y_l = 0.0963, \\Delta \\varepsilon = 0.0688$. The neutrino spectra and the functional form of the luminosity decay are the same as in \\cite{JM96}.  A random initial seed perturbation is added to the velocity field with a modulus of $10^{-3}$ of the (radial) velocity of the post--collapse model.  The computation begins 20\\,ms after core bounce and is carried up to 800\\,ms when the explosion energy has saturated at $E_{\\rm expl} = 1.77\\times10^{51}$\\,erg.  The subsequent propagation of the shock through the stellar envelope and the growth of RT instabilities is simulated with the adaptive mesh refinement code {\\sc AMRA} (Plewa \\& M\\\"uller, in preparation). Neutrino physics is not included in AMRA because it does influence the explosion dynamics only within the first few seconds.  Newtonian self-gravity is taken into account by solving Poisson's equation in one spatial dimension with an angular average of the density, which is an adequate approximation once the shock has left the iron core.  The equation of state includes contributions from photons, non-degenerate electrons, $\\rm e^+e^-$-pairs, $\\rm ^1H$, and the nuclei included in the reaction network.  The AMR calculations are started with the inner and outer boundaries located at $r_{\\rm in}=10^8$\\,cm and $r_{\\rm out}=4.8 \\times 10^{10}$\\,cm.  No further seed perturbations are added.  The maximum resolution is equivalent to that of a uniform grid of $3072 \\times 768$ zones.  The radial extent of the base grid is extented whenever the supernova shock approaches the outer grid boundary, which attains a maximum value of $r_{\\rm out} = 3.9 \\times 10^{12}$\\,cm at $t= 2095\\,$s.  Reflecting (outflow) boundary conditions are imposed in angular (radial) direction.  The initial data for the AMR simulation consists of three different parts. Interior to 17\\,000 km ($1.63\\,\\Msol$) data from the simulations of the first stage of the explosion, and exterior to that radius data from Bruenn's 1D post-collapse model are used.  Matched to Bruenn's model (covering only parts of the He core) is a new progenitor model of Woosley (private communication; data for the 1988 model are no longer available). ",
        "conclusions": "The simulation of the early evolution shows, in accordance with \\cite{JM96}, that neutrino driven convection leads to the formation of a roughly spherical but very inhomogeneous post--shock shell containing dense $\\Ni$--rich regions and low--density, deleptonized bubbles. These inhomogeneities provide the seed for RT mixing in the stellar envelope at the Si/O and (C+O)/He interfaces of the SN\\,1987A progenitor model within only about a minute after core bounce \\cite{Kif00}.  The stellar metal core is completely shredded only 5\\,minutes after bounce and high velocity clumps of newly synthesized elements are observed to be ejected up to the outer edge of the He--core. While the instability is turning the inner core of the star inside out, a dense shell forms at the He/H interface as a result of the deceleration of the main shock in the H--envelope.  The subsequent interaction of the metal--enriched clumps with this dense He--shell leads to their strong deceleration. After entering the shell the clumps reach transonic speeds and dissipate a large fraction of their kinetic energy. As a result the entire dense shell and the H--envelope are pervaded by bow-shocks and strong acoustic waves from 3000 to 10\\,000\\,s after core bounce.  During this interaction, which has not been reported in any previous calculation of RT instabilities in SNe\\,II progenitors, the composition within the clumps themselves is almost entirely homogenized. Furthermore, acting like a wall the shell shields the H--envelope of the star from becoming enriched with freshly synthesized elements.  According to a linear stability analysis the He/H interface should be RT unstable. However, we do not find a strong growth of the instability at this interface.  Since the supernova shock is almost perfectly spherically symmetric when it emerges from the He--core the evolution in these layers proceeds basically one-dimensional.  Only when the metal--enriched clumps reach the inner boundary of the dense shell behind the He/H interface and start to dissipate their energy about 3000\\,s after bounce, are perturbations from spherical symmetry induced by the waves which are thereby excited.  However, as the interface is only moderately unstable at this time, only small scale variations are observed. If this result also holds for other progenitor models, it indicates that neutrino driven convection alone is not able to provide the perturbations which are needed to induce strong mixing of the helium core and the H--envelope as observed in SN\\,1987\\,A.  Already within the first 300\\,s of the explosion elements like $\\Ox$ and $\\Si$ that made up the original metal core as well as the newly synthesized $\\Ni$ have been mixed almost homogeneously throughout the inner $2.0 \\Msol$, \\ie throughout about the inner half of the He--core.  The extent of mixing increases up to $3.2 \\Msol$ at 10\\,000\\,s.  Species which are not mixed this far out in mass are $\\Ti$ and our neutronization tracer. These nuclei were synthesized in the innermost layers of the ejecta which were located very close to the collapsed core.  The dynamics of the explosion is reflected in the distribution of the mass of $\\Ni$ in velocity space (Fig.~\\ref{fig:nivel}). The most conspicuous feature which can be seen in this plot is the bulk deceleration of the material from velocities as large as $\\sim 5000\\,$km/s at a time of 50\\,s after bounce to less than 1500\\,km/s after 10\\,000\\,s. At 50\\,s the average $\\Ni$ velocity is $\\sim 3500$\\,km/s with a spread of about $\\pm 1400\\,$km/s. At 10\\,000\\,s the corresponding values are $\\sim 900\\,$km/s and $\\pm 500\\,$km/s.  The maximum velocities are significantly smaller than the ones which have been observed in SN\\,1987A.  During no phase of the evolution we do see an acceleration of material from the former metal core of the star. This is in contrast to the results of Herant \\& Benz (1992) who report to have obtained nickel velocities comparable to those observed in SN\\,1987A provided that they premixed the $\\Ni$ in their SPH calculations throughout 75\\% (in mass) of the metal core ($\\sim 2 \\Msol$) of their $20\\,\\Msol$ progenitor.  \\begin{figure}[t] \\begin{center} \\epsfig{file=nivel.ps, width=0.75\\linewidth} \\end{center} \\vspace{-1.5cm} \\caption[]{Mass of $\\Ni$ which is contained within the velocity interval $[v_r,v_r+{\\rm d}v_r]$ as a function of the radial velocity $v_r$ at various epochs. The resolution is ${\\rm d}v_r \\approx 130\\,{\\rm km\\,s^{-1}}$.} \\label{fig:nivel} \\end{figure}  The occurrence of conditions which could give rise to the development of an instability at the He/H interface is harmful for the propagation of the clumps. Even if very strong perturbations are imposed upon the dense unstable shell which forms at this interface the instability is growing too slowly in order to shred this ``wall'' before the clumps will reach it.  On the other hand, with a smoother density profile between the He--core and the H--envelope, the dense shell might either not form at all or become at least less pronounced which might help the clumps to preserve most of their energy.  Such a smoother density profile could be envisaged when the progenitor star of SN\\,1987A was not the result of the evolution of a single star but a merger of two smaller stars \\cite{Pod92}.  The dimensionality of our simulations may also be a possible cause of the problem.  The fingers found in 2D calculations are in fact axially symmetric tori which also experience a larger drag when propagating outwards than genuinely 3D mushroom structures.  Finally, ``missing physics'' in the explosion models itself might be responsible for the small maximum nickel velocities that we obtain. Effects like rotation or anisotropic neutrino emission might have to be considered for the explosion mechanism. These may result in an additional large scale asphericity of the shock or even in jet-like outflows of the ejecta.  In that case it has been claimed \\cite{NSS98} that it is possible to reproduce nickel velocities in excess of 3000\\,km/s. However, these simulations suffer from coarse resolution and from rather unrealistic parameterized initial conditions."
    },
    "0010/astro-ph0010420_arXiv.txt": {
        "abstract": "We report on a comprehensive search for Wolf-Rayet (W-R) stars in the SMC using interference-filter imaging.  Photometry of over 1.6 million stellar images on multiple, overlapping fields covering 9.6 square-degrees found the previously known W-Rs at very high significance levels, two known Of-type stars, plus additional candidates which we examined with slit spectroscopy.  We discovered two new Wolf-Rayet stars, both of type ``WN3+abs\", bringing the total number in the SMC to 11. We discuss their spectra, as well as reclassifying the previously known ones with our new data.  Our survey also revealed 4 newly found Of-type stars, including one of O5f?p type, which is one of the earliest-type stars known in the SMC.  Another newly identified Of star is AV~398 (O8.5~If), a star often used in extinction studies under the assumption that it is of early B type.  We recover S18 (AV154), a B[e] star whose spectrum currently lacks He~II $\\lambda 4686$ emission, but which must have had strong emission a year earlier; we compare this star to S~Dor, suggesting it is indeed a luminous blue variable (LBV).  We also find a previously unknown symbiotic star.  whose spectrum is nearly identical to the Galactic symbiotic AG~Dra.  More important, perhaps, than any of these discoveries is the demonstration that there is not a significant number of W-Rs waiting to be discovered in the SMC. The number of W-Rs is a factor of 3 times lower in the SMC (per unit luminosity) than in the LMC.  This strongly suggests that at the low metallicity which characterizes the SMC only the most massive stars can evolve to W-R type. ",
        "introduction": "For many years, our knowledge of the Wolf-Rayet (W-R) content of the SMC was thought to be complete, with a total of eight discovered. Of these, four had been found by general studies of the SMC's blue stellar population (see Breysacher \\& Westerlund 1978), with an additional four found by an objective prism survey targeted specifically at detecting W-Rs by using an interference filter to reduce sky background and crowding (Azzopardi \\& Breysacher 1979). There matters stood until the discovery of Morgan, Vassiliadis, \\& Dopita (1991) of a ninth W-R. This star was well within the boundary surveyed by Azzopardi \\& Breysacher (1979), but a few tenths of a magnitude fainter than any of the others.  Wolf-Rayet stars are the evolved, He-burning descendents of the most massive stars, and their strong emission lines allow them to serve as important tracers of the young stellar population in other galaxies (Massey 1985). The number of W-R stars in the SMC is of interest, as it appears to be disproportionately low compared to that in the LMC, where roughly 130 are known (Breysacher, Azzopardi, \\& Testor 1999). The (visual) luminosity of the SMC is only about a factor of 3.6 less than that of the LMC ($M_V=-17.1$ and $M_V=-18.5$ respectively, van den Bergh 2000a), and, all other things being equal, we would thus expect $\\sim$35 W-Rs in the SMC. Thus the presence of only 9 W-Rs suggests an under-abundance by a factor of $\\sim 3.5$.  This scarcity, if real, is telling us something interesting either about the progenitor massive star population in the SMC, or about how its massive stars evolve in the SMC.  Let us examine these two possibilities in turn.  Since the visual luminosity is dominated by an older population,  it could simply be that the massive stars from which W-R evolve are scarcer in the SMC due to a steeper initial mass function (IMF) and/or a lower current star formation rate (SFR).  What is the available evidence? Massey (1998) finds that the IMF for massive stars appears to be indistinguishable between the SMC and the LMC (and with the Milky Way for that matter).  In that case, a lack of massive progenitors would require a lower SFR in the SMC. Could the current SFR be 4 times lower in the SMC than in the LMC?  Let us compare the number of OB associations in the two Clouds.  Hodge (1985) catalogs 70 regions in the SMC, compared with the 120 found by Lucke \\& Hodge (1970) in the LMC. (Care was taken to use the same selection criteria in the two galaxies, and indeed the size distributions are indistinguishable.) This would suggest that the SMC has a SFR a factor of 2 {\\it higher} per unit luminosity than does the LMC. A better measure of the SFR woud be the integrated H$\\alpha$ luminosity.  However, what is available is not quite this, but rather the total of all the H$\\alpha$ emission from the discrete H~II regions.  Kennicutt (1991) finds that these require an ionizing flux of $5 \\times 10^{51}$~photons~s$^{-1}$ for the SMC, and $3\\times 10^{52}$~photons~s$^{-1}$ for the LMC. The factor of 6 difference would suggest that the SFR is 1.6 times lower per unit luminosity in the SMC. These totals do not include losses from density-bounded H~II regions, which should result in diffuse emission; such a correction may or may not be the same for the two Clouds. In any event, these data suggest that a lack of massive stars is not the primary cause for the factor of 3--4 under-abundance of W-Rs in the SMC compared to the LMC.  On the other hand, we have good reason to suspect that the low metallicity of the SMC will have a significant effect on the evolution of massive stars, as there are smaller mass-loss rates at a given luminosity. In the ``Conti scenario\" (Conti 1976; Maeder \\& Conti 1994) Wolf-Rayet stars are produced as a massive star loses mass via stellar winds; during the He-burning stage the outer layers have been sufficiently stripped to reveal the H-burning products at the surface and the star is spectroscopically identified as a WN-type Wolf-Rayet.  Further mass-loss leads to a WC-type, with the products of He-burning (C, O) revealed in the spectrum. The strong, broad emission lines are the result of a highly extended and rapidly expanding stellar atmosphere; e.g., optical depth unity occurs at a point in the stellar wind. We would thus expect that stars that evolve to  W-Rs in the SMC would have to be of higher luminosity (and mass) than those in galaxies with higher metallicity. (The oxygen abundance is a factor of $\\sim$2--3$\\times$ lower in the SMC than in the LMC (Lequeux et al.\\ 1979; Russell \\& Dopita 1992). Indeed, the high proportion of WN-type Wolf-Rayets to those of WC-type in the SMC (8:1) is  consistent with the hypothesis that WCs are harder to produce at low metallicities due to smaller mass-loss rates, and is in accord with the SMC's metallicity compared to other Local Group galaxies (Massey \\& Johnson 1998).  Recent studies of coeval clusters and associations suggest that ``cut-off\" mass for becoming a W-R star is about 70$\\cal M_\\odot$ in the SMC, while stars with initial masses greater than 30 $\\cal M_\\odot$ become W-Rs in the LMC (Massey, Waterhouse, \\& DeGioia-Eastwood 2000); in the Milky Way the limit may be as low as 18$\\cal M_\\odot$ (Massey, DeGioia-Eastwood, \\& Waterhouse 2001).  However, the discovery of the ninth SMC W-R by Morgan et al.\\ (1991) raises the question of the completness of previous surveys.  We were naturally curious if a significant additional population of W-Rs remained to be found in the SMC.  We have had good success in searching for W-Rs in nearby galaxies using a set of interference filters optimized to detect W-Rs (e.g., Armandroff \\& Massey 1985, Massey, Armandroff, \\& Conti 1986; Massey \\& Johnson 1998).   Accordingly we decided to survey the SMC using the CTIO/Michigan Schmidt with these filters during a gap in our main observing program last year.  Follow-up spectroscopy this year identified 8 hitherto unknown interesting objects in the SMC, including 4 ``Of\" stars, a luminous blue variable (LBV) candidate, a symbiotic star, and two newly found Wolf-Rayet stars.  We describe our survey,  these newly found objects, and the implications here. ",
        "conclusions": "Our survey discovered two previously unknown Wolf-Rayet stars, bringing the total number in the SMC to 11.  Although we cannot preclude a Wolf-Rayet star or two as having been overlooked in our survey, particularly in crowded regions, we believe this study underscores that there is {\\it not} a significant number of W-Rs waiting to be discovered in the SMC. Thus W-R stars are under-abundant in the SMC by a factor of 3 (per unit luminosity) compared with the LMC. As we argue in Section 1, we believe that this supports the view that it requires a higher mass to become a W-R in the SMC than it does in the LMC, due to the effect of lower metallicity on stellar winds.  It is iteresting that both of the newly-found W-Rs show the presence of absorption lines. When Breysacher \\& Azzopardi (1979) first announced the discovery of four additional W-R stars in the SMC, and noted that all appeared to show some sign of OB companions, the inital reaction was that probably Roche-lobe enhanced mass-loss was crucial to a massive star becoming a W-R at low metallicity---that presumably stellar winds by themselves were not sufficient.  Subsequently, Conti et al.\\ (1989) re-examined this point, noting that the case for {\\it bona-fide} binarity was poorly established for most of the SMC W-Rs.  An alterantive explanation for the prevelance of absorption lines in the spectra of the SMC W-Rs could simply be that the stellar winds are weaker, allowing normal photospheric lines to be seen. Over a decade later, orbit solutions still exist for only the same three SMC W-Rs, with the situation for a fourth (HD~5980) made even more complex by its LBV outburst. A radial velocity study is currently underway for all of the SMC W-Rs, which we expect to be complete after the 2001 Magellanic Cloud observing season.   We do note that the statistics for the SMC are very different than for the LMC: for the latter, an absorption-line spectrum is seen only for the brighter systems, consistent with the fact that an OB companion will make the system brighter than on average (Vanbeveren \\& Conti 1980).  The presence of absorption lines in even the faintest of the SMC W-Rs {\\it could} suggest that they are in fact all binaries---or equally well that the absorption lines are seen thanks to a weak stellar wind.  We note that the SMC W-Rs are, on average, more luminous than their counterparts in the Milky Way or LMC.  Ignoring the ``+abs\" systems, Vacca \\& Torres-Dogden (1990) find and average $M_v=-3.7$ for the early-type WNs; the average for the same type stars in the SMC is $-5.3$, where we of course have to include ``+abs\" systems as most of the stars show absorption lines.  Such an average is considerably more luminous even than the $-4.6$ adopted by Conti \\& Vacca (1990) for the ``WNE+abs\" systems.  This cannot be a selection effect, as we argued earlier, as our survey went penty deep enough to find both weaker-lined, fainter stars.  This {\\it could} mean that in general the W-Rs in the SMC do have binary companions which dominate the continuum. Or, it could simply be a reflection of the higher luminosity requirement needed for stellar winds to lead to the W-R stage at lower metallicities.  Our discovery of an O5f?p star, as well as three other hitherto unknown Of stars, emphasizes that much spectroscopic work remains to be done in the Magellanic Clouds."
    },
    "0010/astro-ph0010602_arXiv.txt": {
        "abstract": "We investigate the development of bipolar outflows during the early post-AGB evolution. A sample of ten OH/IR stars with irregular OH spectra and unusually large expansion velocities is observed at high angular resolution.  The sample includes bipolar nebulae (e.g. OH231.8+4.2), bright post-AGB stars (HD 101584) and reflection nebulae (e.g. Roberts 22). The IRAS colour--colour diagram separates the sample into different types of objects. One group may contain the immediate progenitors to the (few) extreme bipolar planetary nebulae. Two objects show colours and chemistry very similar to the planetary nebulae with late IR-[WC] stars. One object is a confirmed close binary.  A model is presented consisting of an outer AGB wind which is swept up by a faster post-AGB wind, with either the AGB or Post-AGB wind being non-spherically symetric.  The interface of the two winds is shown to exhibit a linear relation between velocity and distance from the star, giving the impression of an accelerating outflow. The OH data confirms the predicted linear velocity gradients, and also reveals torus-like, uniformly expanding components.  All sources are discussed in detail using optical/HST images where available.  ISO data for Roberts 22 reveal a chemical dichotomy, with both crystalline silicates and PAHs features being present.  IRAS 16342$-$3814 shows a dense torus with mass $0.1\\,\\rm M_\\odot$ and density of $10^8\\,\\rm cm^{-3}$; HST data shows four point-like sources located symmetrically around the nebula,  near the outer edge of the dense torus.   Lifetimes for the bipolar OH/IR stars are shown to be in excess of $10^4$ yr, longer than normal post-AGB timescales.  This suggests that the toruses or disks are near-stationary. We suggest that accretion from such a disk slows down the post-AGB evolution. Such a process could explain the link between the long-lived bipolar nebular geometry and the retarded star. ",
        "introduction": "Stars at the end of the Asymptotic Giant Branch (AGB) experience a phase of high mass loss during which they expel most of the material from their outer atmospheres and form optically-thick circumstellar envelopes (CSEs) (e.g.\\ Sch\\\"onberner 1989; for a review see Habing 1996).  After the cessation of the mass loss, the star heats up rapidly and eventually begins to ionize the expanding envelope. During the post-AGB evolution, a fast wind from the hot star sweeps up the earlier slow AGB wind to form a planetary nebula (PN) (Kwok, Purton \\& Fitzgerald 1978).  Approximately half of all PNe show bipolar structures.  A bipolar structure may form if the slow AGB wind possesses an asymmetry, which will be amplified by the post-AGB wind--wind interaction (Icke 1988).  Hydro-dynamical models predict the eventual formation of a dense, slowly expanding equatorial torus, with a faster outflow weakly collimated within two polar cones (Kahn \\& West 1985; Balick, Preston \\& Icke 1987; Icke 1988; Frank \\& Mellema 1994). The predicted structures compare well with the observed range of morphologies of PNe (e.g. G\\'orny et al. 1999).  This interacting-wind model predicts that the AGB wind will show deviations from spherical symmetry.  Present evidence shows that the outer winds of AGB stars have a high degree of symmetry.  In the OH 1612 MHz survey of te Lintel Hekkert et al.\\ (1991), only 16 out of 852 OH/IR stars showed irregular line profiles. Sevenster et al.\\ (1997a,b) also found almost exclusively regular profiles. The inner wind may be less symmetric, however.  Optical interferometry (e.g. Tuthill et al.\\ 1994; Bedding et al.\\ 1997) has shown elongation of the stellar photospheres, and OH main-line masers of Mira variables have shown axi-symmetric structures (Bowers, Johnston \\& de Vegt 1989; Chapman \\& Cohen 1985; Chapman, Cohen \\& Saikia 1991; Chapman et al.\\ 1994). Johnson \\& Jones (1991) found significant optical and infrared polarization in AGB and post-AGB stars.  The second prediction of the interacting-wind model is that strongly bipolar morphologies will only develop once the star is hot enough to exhibit a fast wind. Asymmetries during the early post-AGB evolution should not yet be amplified, since these stars are too cool (stellar temperature between 5000 and 15000 K) to have a hot wind.  However, images of early post-AGB objects show a large proportion of highly bipolar morphologies, in sharp contrast with the model prediction. In combination with the insufficient evidence for significant asymmetries in the AGB winds, this indicates that the standard model is incomplete.  Sahai \\& Trauger (1998) therefore propose that early in the post-AGB evolution, a fast jet-like wind operates. An alternative possiblity is that at the very end of the AGB, the mass loss suddenly becomes strongly equatorially enhanced, caused by e.g. the spiral-in of a Jupiter-like companion.  In this study we present aperture-synthesis observations of the OH maser emission from one or more of the ground-state transitions at 1612, 1665 and 1667 MHz, for 10 likely post-AGB stars with unusually broad and irregular spectral profiles. The principal aim of this study is to investigate the envelope geometries of post-AGB stars and to determine whether these can be explained using a wind-wind interaction model.  Section 2 below describes the OH properties of post-AGB stars and the selection of targets. Section 3 describes the observations. Section 4 discusses momentum-driven wind--wind interactions and gives a geometric model which is used to interpret the results. Section 5 describes the results and their interpretation for individual sources. In Section 6 we discuss the nebular and stellar parameters, lifetimes, two proposed evolutionary sequences and the possible effects of Post-AGB accretion on the stellar evolution. Conclusions are summarized in Section 7. ",
        "conclusions": "We have presented an investigation into the onset of bipolar structure formation for OH/IR stars. The selected stars show irregular OH spectra with high expansion velocities, consistent with a possible classification as post-AGB star. We postulate an interaction between an outer slow wind and an inner, somewhat faster wind, with at least one component showing a dependency of mass-loss rate or expansion velocity with polar angle.  The main conclusions are:  1. In the presence of a post-AGB--AGB wind--wind interaction, the velocity of the swept-up shell is predicted to increase linearly with distance from the star. The OH data confirms the presence of such a linear outflow component.  A torus-like component is also seen with expansion velocity of typically 25 km $s^{-1}$.  2. All objects with a well-defined bipolar morphology and high-velocity outflows have IRAS colours closely similar to those of young outflow sources. This indicates that similar morphologies may be present in both evolutionary phases.  3. Two of the selected stars show evidence for a chemical dichotomy, with both crystalline silicates and PAHs or other carbon molecules being present. The silicates are probably located in the torus (as shown by the OH data for Roberts 22).  4. IRAS 16342$-$3814 shows four point-like sources located symmetrically around the central star, at a distance consistent with the outer edge of the circumstellar torus. The nature of these (scattering?) sources is not clear.  5. For five objects where data for the central stars are available, the observed spectral types are M9, F5, A2, B9 and Be. However, the F star has been suggested to have an earlier underlying type (late or early B), affected by an extended atmosphere. The A star (Roberts 22) shows indications for ionization also suggesting an earlier type. For IRAS 16342$-$3814, modeling implies a central star with a temperature around 15 kK, or late B.  The present wind velocities of up to a few 100 km s$^{-1}$ appear consistent with spectral type around late B.  6. Two evolutionary sequences are indicated. The strongly bipolar objects appear related to the (few) planetary nebulae with similar morphologies and may constitute their progenitors. Two objects show colours and chemistry similar to the infrared-bright, late IR-[WC] stars and may be evolutionary related. One bipolar OH/IR star is known on the AGB (OH 19.2$-$1.0), which is a possible progenitor of the bipolar post-AGB stars.  7. OH231.8+4.2 has a Mira central star but also a fast wind. Its wind velocity also indicates a B-star. The central star is therefore likely a binary.  The presence of a Mira precludes a B-type main sequence star, and therefore the companion appears to be a post-AGB star. Following Cohen et al. 1985, we suggest OH231.8+4.2 is a symbiotic OH/IR star.  8. The lifetime of this phase of bipolar post-AGB OH/IR stars is found to be of the order of $10^4$ yr. This is much longer than expected from normal post-AGB evolution. Accretion from a circumstellar torus on a post-AGB star could stall its evolution near the observed spectral types. Such a process could explain the apparent link between retained disks and retarded stellar evolution."
    },
    "0010/astro-ph0010328_arXiv.txt": {
        "abstract": "We report on an \\xmm\\ observation of the most distant known quasar, \\sdssfull, at $z=5.80$. We have detected this quasar with high significance in the rest-frame 3.4--13.6~keV band, making it the most distant cosmic object detected in X-rays; $32\\pm 9$ counts were collected. \\sdssfull\\ is notably X-ray weak relative to other luminous, optically selected quasars, with $\\alpha_{\\rm ox}=-1.91\\pm 0.05$ and a 3.4--13.6~keV luminosity of $\\approx 1.8\\times 10^{44}$~erg~s$^{-1}$. The most likely reason for its X-ray weakness is heavy absorption with $N_{\\rm H}\\simgt 10^{24}$~cm$^{-2}$, as is seen in some Broad Absorption Line quasars and related objects; we discuss this and other possibilities. High-quality spectroscopy from 0.95--1.10~$\\mu$m to search for blueshifted C~{\\sc iv} absorption may elucidate the origin of the X-ray weakness. ",
        "introduction": "\\label{intro}  X-ray observations of quasars at the highest redshifts should reveal the physical conditions in the immediate vicinities of their central black holes. Measurement of the intrinsic X-ray continuum's shape, normalization relative to the rest of the spectral energy distribution, and variability can provide information on the inner accretion disk and its corona, and thus ultimately about how the black hole is being fed. There are already some reports that the X-ray continuum shapes of quasars may evolve with redshift at $z\\simlt 2.5$ (e.g., Blair et~al. 2000; Vignali et~al. 1999); these intriguing results require further study and extension to the highest redshifts. In addition, X-ray absorption measurements can probe material in the environments of high redshift quasars. For example, the fraction of radio-loud quasars (RLQs) with heavy intrinsic and/or associated X-ray absorption (i.e., $N_{\\rm H}\\simgt 5\\times 10^{22}$~cm$^{-2}$) appears to rise with redshift (e.g., Fiore et~al. 1998; Elvis et~al. 1998; Reeves \\& Turner 2000). The absorbing gas in RLQs may be circumnuclear, located in the host galaxy, or entrained by the radio jets.  At $z>4$ our knowledge about the X-ray properties of quasars is unfortunately quite limited (e.g., Kaspi, Brandt \\& Schneider 2000 and references therein). To date it has only been possible to obtain X-ray spectra for four $z>4$ quasars. All four are radio-loud blazars in which the X-ray emission is probably dominated by jets; three of these appear to have X-ray absorption, as would be expected from the trend mentioned above (Moran \\& Helfand 1997; Boller et~al. 2000; Fabian et~al. 2000; Yuan et~al. 2000). The more representative radio-quiet quasars (RQQs) have been too X-ray faint to allow spectroscopy to date; indeed, only 7 optically selected quasars have even been detected in X-rays at $z>4$. At present, measurements of the X-ray spectral shapes of RQQs are scarce and of limited statistical quality above $z\\approx 2.5$ (e.g., Bechtold et~al. 1994; Vignali et~al. 1999).  We have started a project to determine the X-ray properties of the highest redshift quasars using the new generation of X-ray observatories. Our main focus is on quasars with $z>4.8$ discovered by the Sloan Digital Sky Survey (SDSS; see York et~al. 2000). The SDSS multicolor selection method provides a uniform sample of quasars that have been consistently selected in a well-defined manner, and more than 100 SDSS quasars at $z>5$ are expected over the next five years (Fan 1999; Schneider 1999). After the recent discovery of \\sdssfull\\ (hereafter \\sdss; Fan et~al. 2000), an optically bright RQQ (AB$_{1280}=19.28$; $R=f_{\\rm 6~cm}/f_{4400~\\mbox{\\footnotesize \\AA}}<11$) at $z=5.80$, we proposed an \\xmm\\ Target of Opportunity observation of this object. From the Eddington limit, \\sdss\\ appears to contain a $\\simgt 1.6\\times 10^9$~$M_\\odot$ black hole formed within a billion years of the Big Bang (we adopt $H_0=70$~km~s$^{-1}$~Mpc$^{-1}$ and $q_0=0.5$ throughout; note this cosmology differs from those of Fan et~al. 2000). The quasar's luminosity density at 2500~\\AA\\ is $l_{2500~\\mbox{\\footnotesize \\AA}}=2.4\\times 10^{31}$~erg~s$^{-1}$~Hz$^{-1}$, about twice that of 3C~273 (we adopt an optical continuum slope of $\\alpha_{\\rm o}=-0.79$ for \\sdss\\ throughout; see Schneider, Schmidt \\& Gunn 1991 and Fan et~al. 2001). Typical X-ray to optical flux ratios for optically selected $z>4$ quasars of similar luminosity (Kaspi et~al. 2000) suggested that \\xmm\\ might be able to obtain a spectrum of this object in a reasonable integration time. This paper describes the analysis and interpretation of the \\xmm\\ data. ",
        "conclusions": "\\subsection{The X-ray weakness of \\sdss}  With this \\xmm\\ observation, \\sdss\\ surpasses GB~1428+4217 ($z=4.72$; Fabian et~al. 1997) as the highest redshift cosmic object detected in X-rays. The most notable X-ray characteristic of \\sdss\\ is its faintness in this band (see Figure~2); we had expected to obtain an X-ray spectrum for \\sdss\\ but were foiled by the surprisingly low X-ray flux. X-ray weak quasars are known to exist at low redshift where they comprise $\\approx 10$\\% of the optically selected quasar population (e.g., Brandt, Laor \\& Wills 2000, hereafter BLW). There is increasing evidence that the majority of such objects are X-ray weak because they suffer from intrinsic absorption: $\\simgt 80$\\% of optically selected, X-ray weak quasars show strong (4--80~\\AA\\ equivalent width), blueshifted C~{\\sc iv} absorption in the ultraviolet (BLW), and direct X-ray spectral fitting has revealed absorbed (but otherwise normal) X-ray continua in several X-ray weak quasars (e.g., Gallagher et~al. 2000 and references therein). Broad Absorption Line quasars (hereafter BAL~QSOs), for example, are notoriously X-ray weak (e.g., Green \\& Mathur 1996; Gallagher et~al. 1999). The absorbing column densities associated with X-ray weak quasars are often $N_{\\rm H}\\simgt 10^{23}$~cm$^{-2}$ and in some cases appear to be `Compton thick' (i.e., $N_{\\rm H}\\geq 1.5\\times 10^{24}$~cm$^{-2}$).  \\subsubsection{Comparison with the Bright Quasar Survey}  Sharper insight into the X-ray weakness of \\sdss\\ can be gained by comparing it with the $z<0.5$ Bright Quasar Survey (BQS; Schmidt \\& Green 1983). This sample is fairly large and has been extensively studied at many wavelengths; the high-quality constraints on X-ray and ultraviolet absorption are particularly relevant here.  If \\sdss\\ is not absorbed in the X-ray band, its X-ray weakness is extreme compared to other unabsorbed RQQs. In Figure~3, we show a histogram of \\aox\\ for the 53 luminous ($M_{\\rm V}<-22.27$; we use the absolute magnitudes of Boroson \\& Green 1992 converted to our cosmology) BQS RQQs. The \\aox\\ values used are from BLW (calculated following Footnote~1 of that paper). We have shaded the histogram for 7 of these RQQs, since there is strong evidence that their \\aox\\ values have been depressed by intrinsic X-ray absorption. The shaded RQQs are the BAL~QSOs 0043+039, 1001+054, 1700+518 and 2112+059 and the absorbed RQQs 1126--041, 1351+640 and 1411+442 (see Brinkmann et~al. 1999; Gallagher et~al. 1999; Wang et~al. 1999; BLW; Gallagher et~al. 2000). The mean \\aox\\ and \\aox\\ standard deviation for the unshaded BQS RQQs in Figure~3 are $-1.56$ and 0.14, respectively. \\sdss, with $\\alpha_{\\rm ox}=-1.91\\pm 0.05$, has a larger negative value of \\aox\\ than {\\it all but one\\/} of the unshaded BQS RQQs in Figure~3.  The three unshaded BQS RQQs in Figure~3 with the lowest \\aox\\ values are 1259+593, 1552+085 and 2214+139. As we shall now describe, there is some evidence that at least two of these have had their \\aox\\ values depressed by absorption (we have not shaded them in Figure~3 since the data are limited, and we wanted to be appropriately conservative). 2214+139 ($\\alpha_{\\rm ox}=-2.06$) has been observed by \\rosat, but unfortunately the exposure time was fairly short. While detailed spectral fitting is not possible due to the small number of counts collected, its flat X-ray continuum shape ($\\Gamma=-0.07\\pm 0.94$ when only Galactic absorption is assumed) strongly suggests intrinsic absorption (Rachen, Mannheim \\& Biermann 1996). 1552+085 ($\\alpha_{\\rm ox}=-1.80$) is one of the most highly polarized BQS quasars in the optical (1.9\\%; Berriman et~al. 1990); a significant fractional contribution from scattered, polarized light most readily occurs when the direct continuum is absorbed, and most optically selected RQQs with polarization percentages $>1.5$\\% show X-ray absorption (e.g., BLW and references therein). Furthermore, based on International Ultraviolet Explorer (\\iue) data, Turnshek et~al. (1997) argue that 1552+085 is a BAL~QSO, and the notably weak [O~{\\sc iii}] emission of 1552+085 is also often found in absorbed RQQs (e.g., Turnshek et~al. 1997; BLW). 1259+593 ($\\alpha_{\\rm ox}<-1.79$) has the poorest X-ray coverage of any of the $z<0.5$ BQS RQQs; the only limit available is a fairly loose one from the \\rosat\\ All-Sky Survey. We are not aware of any multiwavelength observations which either strongly suggest or argue against intrinsic absorption in 1259+593.  \\subsubsection{Comparison with Other Quasar Samples}  The X-ray weakness of \\sdss\\ is also apparent when it is compared with other quasar samples having higher median luminosities than the BQS. Such comparisons are germane since \\aox\\ appears to generally decrease (i.e., move towards larger negative values) with increasing luminosity (e.g., Avni, Worrall \\& Morgan 1995 and references therein); for our cosmology, \\sdss\\ is $\\approx 40$ times more luminous than the typical BQS RQQ shown in Figure~3. Unfortunately, these higher luminosity samples do not have as complete multiwavelength coverage as the BQS, so it is more difficult to account for the significant effects of absorption that plague many \\aox\\ studies (see BLW and \\S3.1.1 for discussion). In addition, precision comparisons with a sample matched in luminosity are hindered because, for a reasonable range of $q_0=$~0.1--0.5, the luminosity of \\sdss\\ at $z=5.80$ varies by a factor of $\\approx 3.5$; all well-studied comparison samples have substantially lower median redshifts, and thus their luminosities depend much less on $q_0$.  In Figure~4, we compare the \\aox\\ value of \\sdss\\ with \\aox\\ values derived for other RQQs found in optically selected surveys. In particular, the Large Bright Quasar Survey (LBQS; Hewett, Foltz \\& Chaffee 1995) RQQs provide a well-defined, optically selected comparison sample with a significantly higher median luminosity than the BQS; the X-ray properties of these RQQs were studied by Green et~al. (1995) using an X-ray counts stacking technique. Comparison of \\sdss\\ with the highest luminosity LBQS data point in Figure~4 affirms its X-ray weakness. Furthermore, it is important to note that the LBQS data points include some absorbed RQQs (e.g., BAL~QSOs; see \\S6 of Green et~al. 1995). Removal of the absorbed RQQs would shift the LBQS data points towards less negative values of \\aox, increasing their separation from \\sdss. The magnitude of this shift is expected to be $<0.1$ (P.J. Green 2000, private communication), largely due to the limited sensitivity \\rosat\\ All-Sky Survey observations used by Green et~al. (1995).  We have used the weighted orthogonal regression for the RQQs in Table~2 of Green et~al. (1995) to predict the expected \\aox\\ value for \\sdss\\ given its luminosity density at 2500~\\AA\\ (see \\S1).\\footnote{Revised values for the slope and intercept in Table~2 of Green et~al. (1995) are $0.157$ and $-3.267$, respectively (P.J. Green 2000, private communication).} We predict $\\alpha_{\\rm ox}=-1.71$, in good agreement with expectations from Figure~4.  Figure~4 also includes the optically selected $z>4$ RQQs studied by Kaspi et~al. (2000). We have re-calculated the 2500~\\AA\\ luminosity densities and \\aox\\ values from this paper using the optical continuum slope measurements from Schneider et~al. (1991) or the mean value of $\\alpha_{\\rm o}=-0.79$ from Fan et~al. (2001). The \\aox\\ value for \\sdss\\ is substantially below the range found for these quasars. Finally, we note that \\sdss\\ has a larger negative \\aox\\ value than all but one of the RQQs studied by Bechtold et~al. (1994) even though these RQQs are significantly more luminous in the rest-frame optical band than \\sdss.  In conclusion, if \\sdss\\ does not suffer from intrinsic and/or associated X-ray absorption, then its anomalously weak X-ray continuum marks it as an unusual quasar.  \\subsection{Intrinsic and/or associated X-ray absorption?}  The results of the above discussion suggest that heavy absorption may be affecting the X-ray spectrum of \\sdss, and heavy absorption (specifically, a low-energy cutoff) is indeed commonly found in the X-ray spectra of $z\\simgt 2.5$ RLQs (see the references in \\S1). Is it possible that X-ray absorption generally increases with redshift in RQQs at $z\\simgt 2.5$? In this case, the large negative \\aox\\ value for \\sdss\\ might be the result of this trend in extremis, and a few RQQs at $z\\approx 2.5$ do appear to show significant X-ray absorption (see Figure~9 of Reeves \\& Turner 2000). While we consider this to be an enticing possibility requiring further investigation, we note that the \\aox\\ values for the $z>4$ RQQs in Kaspi et~al. (2000) do not appear to support it. Thus, the most likely interpretation at present appears to be that \\sdss\\ is just one of the minority of optically selected RQQs that have intrinsic and/or associated absorption depressing their X-ray continua (rather than a representative example of a widespread, general RQQ trend). For example, \\sdss\\ may be a BAL~QSO or mini-BAL~QSO. Finally, we note that because the high redshift of \\sdss\\ has provided us with access to penetrating, high-energy X-rays, any absorber must be very thick with $N_{\\rm H}\\simgt 10^{24}$~cm$^{-2}$ (perhaps with partial covering of the X-ray continuum source to explain the photons detected at $\\simlt 8$~keV).  The integrated column density of the intergalactic medium out to $z=5.80$ is almost certainly too small to produce the required X-ray absorption, especially given the intergalactic medium's low metallicity (e.g., Miralda-Escud\\'e 2000).   \\subsection{An intrinsically weak X-ray continuum?}  We must also admit the possibility that \\sdss\\ has an intrinsically weak X-ray continuum; such objects are rare at low redshift, but some may exist (e.g., BLW; Gallagher et~al. 2000). Speculating for a moment, \\sdss\\ could be a `precursor quasar' where the relativistically deep potential well needed for efficient X-ray production is still in the process of formation (see \\S3.3 of Haehnelt \\& Rees 1993); at ultraviolet and longer wavelengths, such contracting precursors may be able to sustain quasar-level luminosities for $\\sim 10^7$ years. Alternatively, the black hole in \\sdss\\ could be accreting at a highly super-Eddington rate where `trapping radius' effects cause the X-rays created in the inner region of the accretion flow to be dragged back into the hole (e.g., Begelman 1978; Rees 1978); such an accretion flow has a luminosity of about the Eddington limit even though the mass accretion rate is super-Eddington. To first order, radiation trapping would truncate the spectrum above a critical frequency depending on the trapping radius; it could therefore lead to weak X-ray emission. Super-Eddington accretion would help to explain the formation of such a massive black hole in less than a billion years (see \\S1).  Even if \\sdss\\ is intrinsically X-ray weak, this should not dramatically change the bolometric luminosity estimated in \\S4.2 of Fan et~al. (2000). In luminous RQQs, X-rays typically contribute $\\simlt 10$\\% of the bolometric luminosity (e.g., \\S6.1 of Elvis et~al. 1994).  \\subsection{Future studies}  Searches for intrinsic and/or associated absorption lines in the rest-frame ultraviolet spectrum of \\sdss\\ are now of key importance for further understanding the origin of its X-ray weakness. Specifically, most of the X-ray weak quasars at low redshift that suffer from X-ray absorption also show significant blueshifted ultraviolet absorption; BQS RQQs with comparable \\aox\\ values to \\sdss\\ typically have C~{\\sc iv} absorption equivalent widths of 4--80~\\AA\\ (BLW). Due to the high redshift of \\sdss, it does not yet have spectral coverage of C~{\\sc iv} (this line has an observed-frame wavelength of 1.05~$\\mu$m). Absorption lines from N~{\\sc v} and Si~{\\sc iv} are frequently seen in objects with strong C~{\\sc iv} absorption, and these lines are covered by the spectrum shown in Figure~2 of Fan et~al. (2000). One can make a case for a moderate strength Si~{\\sc iv} absorption line blueshifted by $\\approx 6000$~km~s$^{-1}$ from the emission line (with a maximum rest-frame equivalent width of $\\approx 6$~\\AA), but this could simply be a dip in the continuum. Any N~{\\sc v} absorption at this velocity would be difficult to disentangle from the Ly$\\alpha$ forest; there is none apparent at lower velocities. High-quality spectroscopy of the C~{\\sc iv} line is required to settle this issue. Rest-frame ultraviolet absorption studies may well be the only way to better understand the X-ray weakness of \\sdss\\ in the short term. Since the low X-ray flux of \\sdss\\ puts it below the practical spectroscopy limit for even \\conx\\ (see Figure~2), directly addressing the origin of its X-ray weakness with X-ray spectroscopy will almost certainly be impossible until \\xeus\\ or a comparable mission begins operation.  More generally, further X-ray observations of the highest redshift quasars are now needed to determine the distribution of \\aox\\ for these objects. \\chandra\\ and \\xmm\\ studies of additional quasars found by the SDSS will soon address this issue."
    },
    "0010/astro-ph0010434_arXiv.txt": {
        "abstract": "{ The UV resonance transitions of neutral argon Ar\\,{\\sc i}\\,$\\lambda1066$ \\AA\\, and of singly ionized phosphorus P\\,{\\sc ii}\\,$\\lambda963$ \\AA\\, originated  in the damped Ly$\\alpha$ system  (DLA) at \\zabs = 3.3901 towards  QSO 0000--2620 have been detected by means of the UVES spectrograph at the 8.2m ESO KUEYEN telescope.   So far, this is the first measurement of Ar\\,{\\sc i}\\, and the second of P\\,{\\sc ii}\\, ever performed in  damped galaxies  and in  high redshift objects. This DLA is well known for having one of the lowest metal abundances and dust content, and the lowest fractional abundance of molecular hydrogen H$_2$. The measured Ar abundance is    [Ar/H] = $- 1.91 \\pm 0.09$ which is equal to   the abundances of the other $\\alpha$-chain elements  (O, S and Si). The similarity of the Ar abundance with the other $\\alpha$-chain elements implies the absence of significant photoionization by either UV background  or stellar sources along the sightline throughout the damped Ly$\\alpha$ system.  Both  $\\log$(Ar/O) and $\\log$(Ar/S) ratios are found  close to those  measured in the extragalactic H\\,{\\sc ii} regions and in blue compact galaxies where O is more abundant by at least one order of magnitude . This strengthens the universality of the Ar/O and Ar/S ratios and lends support to the existence of a universal IMF.  The abundance of the non-refractory element phosphorus [P/H] = $- 2.31 \\pm 0.10$ confirms the low amount of chemical evolution  in the DLA.   This is the  measurement of P in the most metal-poor material and shows a  subsolar   [P/Fe] = --0.27  value. The measured ratios [P/Si] = $- 0.40 \\pm 0.13$ and [P/S] = $- 0.33 \\pm 0.13$ provide evidence for a mild odd-even effect.   Finally, a stringent upper limit to the population of the $^3$P$_1$ level in the ground state of O\\,{\\sc i} is derived, which provides a lower limit to the physical dimensions of the  \\zabs = 3.3901 system of $L > 7$~pc. } ",
        "introduction": "Damped Ly$\\alpha$ (DLA) systems are absorption systems observed in quasar spectra characterized by large column densities ($N$(H\\,{\\sc i})$\\ge$ 2$\\times 10^{20}$ cm$^{-2}$). They span the whole epoch of galaxy formation and are generally considered the progenitors of the present day galaxies. DLA systems are of particular interest because it is possible to measure  a variety of chemical elements in their interstellar gas-phase with unprecedented precision (Lu et al. 1996; Prochaska \\& Wolfe 1999; Pettini et al. 1997, 1999 and 2000). However, the observed abundance patterns  in the DLAs proved difficult  to account for,  due to the possibility that a significant fraction of some elements contribute  to dust grain formation  thus depleting   the  gas phase abundances (Kulkarni et al. 1997; Vladilo 1998). At face value the  observed Si, S, Cr, Al, Ni abundances relative to Fe are similar to those observed in the old stellar population of the Galaxy, with the notable exception of  Zn which is overabundant compared  to iron. However, if the [Zn/Fe] {\\it anomaly} is interpreted as a Fe depletion  in the interstellar medium,  the observed abundance pattern would be similar to   those observed in the warm phase of the interstellar medium with solar chemical composition. Zinc has not been considered a totally reliable element by Prochaska et al. (2000), because of its uncertain nucleosynthesis. A  sample of chemical elements larger than those presently available, and in particular of elements with little affinity with dust grains, is  required  to better address the kind of  chemical pattern present in  DLAs.  In this paper we report on the  detection of UV lines of the non-depleted species -- neutral argon Ar\\,{\\sc i} and singly-ionized phosphorus P\\,{\\sc ii} -- in a damped Ly$\\alpha$ absorption (DLA) system at \\zabs = 3.3901 towards the quasar Q~0000--2620. So far,  phosphorus has been measured by Outram et al. (1999) in the DLA at \\zabs=2.625 towards GB 1759+7539 and an upper limit on  phosphorus abundance has been set for the DLA at \\zabs = 2.309 towards PHL~957 (Molaro et al. 1998). A possible identification of the Ar\\,{\\sc i} lines in the \\zabs = 2.8112 damped Ly$\\alpha$ system towards PKS~0528--250 was reported by Srianand \\& Petitjean (1998), however, in their spectrum the Ar\\,{\\sc i} lines were considerably contaminated by hydrogen absorption which yields a rather uncertain Ar\\,{\\sc i} column density.  The Ar we derive here is the first one ever obtained in any DLA and one of the few values measured by means of unsaturated lines unlike with the interstellar studies in the Galaxy (e.g. Sofia \\& Jenkins 1998). Being  a noble gas with very low condensation temperature, $T_{\\rm c} = 30$~K (Spitzer \\& Jenkins 1975), neutral argon atoms do not form chemical bonds with solid compounds, and the depletion of phosphorus is found to be very small in the ISM where variations in refractory elements do occur (e.g. Dufton et al. 1986). The dust-to-gas ratio in the DLA at z$_{abs}$=3.3901 towards QSO 0000-2620 was found to be less than 0.2\\% than  that in the Milky Way (Levshakov et al. 2000b), and thus the measured [Ar/H] and [P/H] ratios\\footnote{[X/H] $\\,\\equiv\\, \\log\\, [N({\\rm X})/N({\\rm H})] - \\log\\,[N({\\rm X})/N({\\rm H})]_\\odot$} at \\zabs = 3.3901 should not be significantly affected by  dust depletion, thus  providing a unique probe of pure nucleosynthetic yields at high redshift.  The study of phosphorus may eventually allow to clarify the dust formation processes in DLA galaxies. Since P and Fe have nearly the same condensation temperature $T_{\\rm c} \\simeq 1300$~K, these species should deplete together in equal proportions if dust is produced in dense and hot shells of supernova remnants where thermodynamic calculations of condensation are appropriate. A possible disparity between   P and Fe gas-phase abundances would favour a different  model where the dust grains grow in the interstellar medium (see Jura \\& York 1978, and references therein).  We discuss here the implications of the Ar and P abundances in connection with the chemical evolution of such damped galaxies. We underline that for research  on the element production history, Ar is a quite unique element due to its ability of  tracing how strongly the apparent abundances of other species from H\\,{\\sc i} regions could be influenced by ionization.  The paper is organized as follows: in \\S~2 we briefly describe observations and analyze  Ar\\,{\\sc i} and P\\,{\\sc ii} profiles. The results on Ar and P abundances are discussed in \\S~3 (sections 3.1 and 3.2, respectively). In Section~3.3, the estimations of the mean gas density and the linear size of the \\zabs = 3.3901 system are also given. Conclusions are   summarized in \\S~4. ",
        "conclusions": "UVES observations of the quasar Q~0000--2620 gave as a result the first detection of the Ar\\,{\\sc i}\\,$\\lambda1066.660$ \\AA\\, and P\\,{\\sc ii}\\,$\\lambda963.801$ \\AA\\, transitions originated in the damped Ly$\\alpha$ system at \\zabs = 3.3901. The  derived abundances are [Ar/H] = $-1.91 \\pm 0.09$ and   [P/H] = $-2.31 \\pm 0.10$ with most  errors originated in the neutral hydrogen column density.  The phosphorus abundance provided here is one of the few available for this element in any astronomical site and unique in  material poor of metals. The low abundance of the non-refractory element P confirms the low  metallicity level of the  galaxy which is therefore in the early stages of its chemical evolution. The P abundance is slightly below the iron-peak elements and when compared to its close nuclei S and Si, P shows mild evidence of an odd-even effect, but less extreme than what predicted by theoretical models by Woosley \\& Weaver (1995) or Limongi et al. (2000).  Ar abundance is remarkably similar to that of oxygen, sulphur and silicon, [O/H] = $- 1.86 \\pm 0.15$, [S/H] = $- 1.98 \\pm 0.09$, and [Si/H] = $- 1.91 \\pm 0.08$. Ar has a large photoionization cross-section and the similarity of its abundance with the other $\\alpha$-chain elements implies the absence of significant photoionization  either by external or internal sources.  Log~(Ar/O) and $\\log$(Ar/S) ratios are almost the same as  those found in the blue compact galaxies and in present day galaxies where  O  is more abundant by one or two orders of magnitude. This strengthens the constancy of  Ar/O and Ar/S ratios and lends support to the existence of a universal IMF.  In the DLA under study the relative abundances of $\\alpha$-chain and iron-peak elements, as measured by an unprecedented  number of indicators, show only a mild overabundance at variance with the pattern of the old stellar population of the Galaxy. Since this DLA is a dust-free system, its chemical composition and the relative abundances of the observed species may be used to constrain model parameters in the theoretical calculations of stellar nucleosynthesis in the metal-poor gas.  Measurements of the volatile and nonvolatile element abundances have shown that none of the known processes of dust formation or destruction is effective  in the \\zabs = 3.3901 galaxy.  Finally, the upper limit to the population of the first fine-structure level in the ground state of O\\,{\\sc i} is used to infer a lower limit to the physical dimension of the DLA system, which is placed at $L > 7$~pc."
    },
    "0010/astro-ph0010091_arXiv.txt": {
        "abstract": "We present the analysis of new X-ray observations of two Galactic radio supernova remnants (SNRs), G337.2$-$0.7 and G309.2$-$0.6. Both remnants exhibit line-rich thermal spectra that require highly non-solar elemental abundance ratios. In each case, Si and S are unequivocally overabundant compared to solar based purely on the measured equivalent width of the K$\\alpha$ line complex.  Detailed nonequilibrium ionization spectral analysis of these remnants, using a single temperature, single ionization age model, confirm the overabundances, yielding Si, S and Ar abundances many times their solar values: 3 to 5 times solar for G337.2$-$0.7 and greater than 3 to 10 times solar for G309.2$-$0.6. We also find that for both remnants, Ne and Mg are underabundant relative to Si, and for G309.2$-$0.6, Ca and Fe as well are relatively underabundant.  This is the first detection of metal-rich, non-solar abundance material in these remnants, identifying them as young ejecta-dominated SNRs.  Further support for their youth comes from the distances that we estimate from the fitted column densities (less than 15~kpc for G337.2$-$0.7, and $4 \\pm 2$~kpc for G309.2$-$0.6). From the implied sizes of the remnants at these distances we infer relatively young ages (2000 to 4500 years for G337.2$-$0.7, and 700 to 4000 years for G309.2$-$0.6). Finally we note that no compact object has yet been found within G337.2$-$0.7, but an X-ray point source is evident within the radio shell of G309.2$-$0.6, although its nature remains unknown. ",
        "introduction": "Supernovae (SNe) are the birthplace of almost all metals that exist in the universe. Models of the explosive nucleosynthesis that occurs in SNe are used to study the evolution of the interstellar medium, external galaxies, and even clusters of galaxies. Young supernova remnants (SNRs), whose emission is still dominated by the ejecta, are the best test for our models of nucleosynthesis.  However, these models have primarily been tested in the ensemble, against the abundances of large regions, such as the solar neighborhood, where those abundances are assumed to be the product of many representative supernovae.  Direct measurements of the nucleosynthetic yields of individual supernova, using the abundances found in their ejecta, has been much more limited.  Cas A, which has been extensively studied across the wavebands (e.g., Chevalier \\& Kirshner~1979; Fesen, Becker, \\& Blair~1987; Douvion, Lagage, \\& Cesarsky~2000; Hughes et~al.~2000) is more the exception than the rule.   Considering that the Galactic SNe rate is widely believed to be a few per century, there should have been on the order of 5--10 Galactic SNe since the last historical one observed by Kepler in 1604. Additionally there must be a sizeable population of SNRs less than $\\sim$2000 years old or so. However, prior to {\\it ASCA} there were only a handful of young remnants known and they fell into two classes: the remnants of young historical SNe (e.g. Tycho, Kepler) and oxygen-rich SNRs (e.g. Cas A, G292.0$+$1.8) classified based on their optical characteristics. Studies with the {\\it Advanced Satellite for Cosmology and Astrophysics} ({\\it ASCA}) have added a third class of presumably young remnants: those for which optical evidence of ejecta is absent, but whose X-ray spectra show enhanced, non-solar abundances, and hence are likely to be ejecta-dominated (e.g. W49B in the Galaxy, Fujimoto et~al. 1995; or N103B in the LMC, Hughes et~al.~1995).  Clearly, identifying and studying new X-ray ejecta-dominated SNRs will be vital for understanding the SNR population and the process of nucleosynthesis.  G337.2$-$0.7 and G309.2$-$0.6 are part of our ongoing project surveying small Galactic supernova remnants in an attempt to identify more ejecta-dominated SNRs through their X-ray emission.  G337.2$-$0.7 and G309.2$-$0.6 were first identified as SNR candidates based on their possible non-thermal radio emission (Green 1974). These identifications were confirmed in the radio in the MOST supernova remnant catalogue (Whiteoak \\& Green 1996). G309.2$-$0.6 was the subject of a detailed radio study by Gaensler, Green, \\& Manchester (1998) (hereafter GGM) using the Australia Telescope Compact Array (ATCA).  The radio images of G337.2$-$0.7 and G309.2$-$0.6 are given in Figures~1a and~2a.  G337.2$-$0.7 shows a barely resolved, possibly shell-like morphology, with a diameter of $\\sim$6~arcminutes and an overall radio surface brightness at 843 MHz of $11\\times 10^{-21}$~W~m$^{-2}$~Hz$^{-1}$~sr$^{-1}$ (Whiteoak \\& Green 1996). G309.2$-$0.6 exhibits a more complex morphology in the radio with evidence for 2 shells, diameters 12 and 15~arcminutes, and a possible jet-like feature to the northeast (GGM) which appears as a faint wisp in our reproduction. The SNR has a surface brightness at 843 MHz of $5.4\\times 10^{-21}$~W~m$^{-2}$~Hz$^{-1}$~sr$^{-1}$ (Whiteoak \\& Green 1996). GGM find a distance for G309.2$-$0.6 of between 5.4$\\pm$1.6 and 14.1$\\pm$0.7 kpc using H{\\small I} absorption measurements.  GGM's lower limit was found based on absorption towards the SNR, and the upper limit was constrained by the lack of a positive velocity absorption feature which was present in other nearby sources, indicating that the remnant must be closer than those sources.   These SNRs are positionally coincident with bright sources from the {\\it ROSAT} All-Sky Survey (RASS) (Voges et~al.~1999).  G337.2$-$0.7 corresponds to source 1RXS~J163931.4$-$475019, with a count rate of $0.05 \\pm 0.02$ counts s$^{-1}$, and G309.2$-$0.6 is close to source 1RXS~J134651.1$-$624843 with a count rate of $0.06 \\pm 0.02$ counts s$^{-1}$.  These measurements indicated the presence of soft X-ray emission from these remnants; our {\\it ASCA} observations were required to determine the nature of this emission.   The {\\it ASCA} X-ray observations of these remnants provide new insight into their physical properties. The {\\it ASCA} band (0.7$-$10 keV) spans emission lines of Ne, Mg, Si, S, Ar, Ca, and Fe, allowing for examination of the abundances, temperature, density and ionization state in the X-ray emitting region, as well as the line-of-sight absorbing column density, although individual emission lines are, in general, not resolved by {\\it ASCA}. This first examination of the X-ray spectra from these remnants reveals line-rich thermal emission with strong indications of highly non-solar elemental abundance ratios.  In section 2 we outline the data reduction, background subtraction and general image analysis. In section 3 we present the spectroscopy and nonequilibrium ionization (NEI) analysis, along with the resulting abundances. Section 4 discusses the distances, ages, and nucleosynthesis models. Finally, we summarize our results and highlight areas for further investigation in section 5. ",
        "conclusions": "The enhanced abundances derived from the strong X-ray emission lines of highly ionized Si, S and Ar reveal the ejecta-dominated nature of both G337.2$-$0.7 and G309.2$-$0.6. The EWs of the Si and S lines for both remnants are greater than the maximum possible EWs for solar abundance plasmas with temperatures between 0.4 and 5.0 keV.  In addition the prominence of these lines compared to the Ne and Mg lines, requires non-solar abundance ratios.  Using nonequilibrium ionization thermal plasma spectral models we derive abundances for these remnants that are highly non-solar. Over the entire range in temperatures and ionization timescales, the abundances of Si, S, Ar and Ca are many times the solar values for G337.2$-$0.7 and for G309.2$-$0.6 a pure metal plasma of only Ne, Mg, Si, S and Ar (i.e., no H or He) is allowed. In addition, the elemental abundances are not well described by a single multiplicative factor times their solar abundances. For G337.2$-$0.7, Ne, Mg, and possibly Fe are significantly underabundant relative to the Si-group elements.  For G309.2$-$0.6 Ne, Mg, Ar, Ca and Fe are all underabundant compared to Si and S.  We thus conclude that G337.2$-$0.7 and G309.2$-$0.6 are new examples of ejecta-dominated remnants.  Furthermore our spectral fits provide us with a measure of the column densities to each remnant, which when coupled to estimates of the absorption as a function of distance along a given line of sight, provides distance estimates for both remnants. For G309.2$-$0.6 we find a distance of $4 \\pm 2$~kpc consistent with both GGM's estimate from \\ion{H}{1} absorption lines, and the position of the Scutum-Crux spiral arm. For G337.2$-$0.7 we find an upper limit on the distance of 15~kpc, placing it no further than far side of the Norma spiral arm. Using these distances, we can place lower limits on the ages of the remnants by assuming free-expansion. G309.2$-$0.6 is at least 700 to 2000 years old, while G337.2$-$0.7 is greater than 1700 to 2600 years old. A nominal Sedov solution, $E_{0} = 10^{51}$~ergs, $n_{0} = 0.2$~cm$^{-3}$ yields upper limits on the ages of 4500 years for G337.2$-$0.7, and 4000 years for G309.2$-$0.6.  These age estimates corroborate our finding that both remnants are young and ejecta-dominated.  We also draw attention to the X-ray point source 1WGA~J1346.5$-$6255, apparent within the radio shell of G309.2$-$0.6. We found no pulsations for this source in the 0.002~Hz to 32~Hz frequency range, but this only limits the pulsed fraction to be less than $\\sim$85\\%. Although we were able to estimate its distance using the absorbing column density, we can neither exclude the possibility that it is associated with the SNR nor with the foreground open cluster NGC~5281. Higher spatial and temporal resolution observations as are possible with the {\\it Chandra X-ray Observatory} are vital to resolve the nature of this source as well as to unambiguously distinguish between the emission from this point source and the SNR itself.  Continued investigation of ejecta-dominated SNRs such as G337.2$-$0.7 and G309.2$-$0.6 is crucial.  Higher spectral resolution observations as will be possible with {\\it XMM-Newton} and {\\it Chandra} will allow further constraints on both the abundances and the thermodynamic state of the shocked plasma in these remnants. Further, the breadth of possible investigations that take advantage of the 0.5$^{\\prime \\prime}$ spatial resolution of {\\it Chandra} is staggering. Now it is possible to study the abundances of individual knots of ejecta and compare them with models of explosive nucleosynthesis from different layers in the exploding SN and thereby trace the dynamics of the explosion, as we have already begun for Cassiopeia A (Hughes et~al.~2000). We can also isolate the spectrum of small-scale features in a remnant (e.g., the outermost blast-wave) and thereby study how SNe shocks impart energy to the ISM, heat electrons and ions, generate cosmic rays, and so on (see Hughes, Rakowski, \\& Decourchelle 2000). Thus studies of young ejecta-dominated remnants, like those presented here, with the new generation of X-ray observatories, have the potential to make a significant impact on a number of important astrophysical questions."
    },
    "0010/astro-ph0010572_arXiv.txt": {
        "abstract": "%  We explore the thermal evolution of a neutron star undergoing episodes of intense accretion, separated by long periods of quiescence. By using an exact cooling code we follow in detail the flow of heat in the star due to the time-dependent accretion-induced heating from pycno-nuclear reactions in the stellar crust, to the surface  photon emission, and the neutrino cooling. These models allow us to study the neutron stars of the Soft X-Ray Transients.  In agreement with Brown, Bildsten and Rutledge (1998) we conclude that the soft component of the quiescent luminosity of Aql X-1, 4U 1608-522, and of the recently discovered SAX J1808.4, can be understood as thermal emission from a cooling neutron star with negligible neutrino emission. However, we show that in the case of Cen X-4, despite its long recurrence time, strong neutrino emission from the neutron star inner core is necessary to understand the observed low ratio of quiescent to outburst luminosity. This result implies that the neutron star in Cen X-4 is heavier than the one in the other systems and the pairing critical temperature $\\Tc$ in its center must be low enough (well below $10^9$ K) to avoid a strong suppression of the neutrino emission. ",
        "introduction": "%   Neutron stars in Soft X-Ray Transients (SXRTs) undergo recurrent surges of activity separated by long phases of relative \"quiescence\". The detection of type I bursts during the  X-ray brightening observed in outburst ($\\sim 10^{37-38}\\ergs$) unambiguously indicate that episodes of intense accretion occur onto the stellar surface. The origin of the faint X-ray emission, observed in quiescence, at the level of $\\sim 10^{32-33}\\ergs,$ is instead uncertain and may result  either from surface thermal emission (Brown et al. 1998\\markcite{BBR98}) and/or from shock emission due to an hidden fastly spinning pulsar (Stella et al. 1994\\markcite{Setal94}; Campana et al. 1998a,b\\markcite{Cetal98a}\\markcite{Cetal98b}).  Brown, Bildsten and Rutledge (1998\\markcite{BBR98}, BBR98) showed, using simple but concrete arguments, that the \"rock-bottom\" emission, in quiescence, could result from the cooling of the neutron star made hot during the events of intense accretion: the inner crust compressed by the loaded material becomes the site of pycno-nuclear reactions that may deposit enough heat into the core to establish a thermal luminosity \\begin{equation} \\Lq\\sim {Q_{\\rm {nuc}}\\over m_u} \\mdotav \\sim 6\\,\\times 10^{32} {\\mdotav\\over 10^{-11} \\Msun \\rm {yr}^{-1}} \\,\\,\\,\\ergs \\end{equation} detectable when accretion halts ($\\mdotav$ is the time averaged accretion rate, $Q_{\\rm {nuc}}\\sim 1.5$ MeV, the nuclear energy deposited per baryon, and $m_u$ the atomic mass unit). Following this consideration, Rutledge et al. (1999\\markcite{Retal 99}, 2000\\markcite{Retal00}) showed that spectral fits with accurate hydrogen atmosphere models lead to emitting areas consistent with a neutron star surface, pointing in favor of a thermal origin of the quiescent emission (or of part of it).  The interpretation of the quiescent emission seen in SXRTs as due to ``cooling'' gives the opportunity of probing the physics of the neutron star interior in an unprecedented way. We study in detail the thermal evolution of transiently accreting sources with a full cooling code, computing the ``exact'' quiescent luminosity and relating it to the accretion history. We then explore the consequences of the cooling hypothesis to infer properties of the underlying neutron star. ",
        "conclusions": ""
    },
    "0010/astro-ph0010058_arXiv.txt": {
        "abstract": "We have obtained follow-up observations of the quasar pair LBQS 0107$-$025A,B and new observations of the nearby quasar LBQS 0107$-$0232 with the Hubble Space Telescope Faint Object Spectrograph. Extended wavelength coverage of LBQS 0107$-$025A and B using the G270H grating was also obtained. This triple system is unique in providing sensitivity to coherent Ly$\\alpha$ absorption on transverse scales of approximately 1 Mpc at $z <$ 1. Monte Carlo simulations were used to establish the confidence level for matches between absorption features in different lines of sight as a function of velocity separation. Pairwise, there are 8, 9, and 12 lines that match between spectra. Three instances of matches between all three lines of sight were found with velocity separations of less than 550 $\\rm{km\\ s^{-1}}$. Two of the pairings have coincident lines within $\\rm{|\\Delta v| < 200 kms^{-1}}$ that would occur with less than 10\\%\\ probability by chance. Taking into account the equivalent widths of the lines, one of these triple coincidences is significant at the 99.99\\% confidence level, based on Monte Carlo simulations with random line placements. Matches with strong lines preferentially have small velocity separations. These same simulations are used to demonstrate that the distribution of matches for a population of absorbers randomly distributed in velocity space is peaked toward $|\\Delta$v$|$ = 0, which has implications for the statistical significance of matches. One of the triple coincidences appears to be a strong absorber with a sheet-like, but inhomogeneous, geometry and a coherence length approaching or exceeding 1 Mpc. ",
        "introduction": "Quasar absorption lines offer a remarkable opportunity to study the large scale structure of the universe. Given a sufficiently bright background illuminating source and a combination of observations from the ground and space, their properties can be studied over 90-95\\% of the Hubble time. Much recent progress has been made using the unique spectroscopic capabilities of the Hubble and Keck telescopes. Interest in low column density hydrogen absorbers -- the Lyman $\\alpha$ forest -- is also motivated by theoretical insights. The absorbers are valuable cosmological tools because (a) they are as common as luminous galaxies, (b) they contain most of the baryons in the universe, (c) they accurately trace the underlying dark matter potential, and (d) at low column density they may be primitive enough to retain a memory of initial conditions, unlike highly non-linear objects such as galaxies \\citep{cen94, zha95, her96, mir96, dav99}. The fact that supercomputer simulation can reproduce the basic features of the Lyman $\\alpha$ forest has become an in important argument in support of the standard picture of hierarchical structure formation.  The statistical properties of quasar absorbers are usually deduced from the combination of observations along single, widely separated sight-lines. At high redshift, the HIRES spectrograph of the Keck telescope has provided the sensitivity and resolution needed to resolve the Lyman $\\alpha$ forest \\citep{hue95, kim97, kir97}. Below a redshift of $z = 1.6$, we must rely on the modest aperture and moderate resolution of the spectrographs aboard the Hubble Space Telescope. The demographics of the absorbers have been defined by the Quasar Absorption Line Key Project \\citep{bah96, jan98, wey98}. Quasar pairs or groups that are closely aligned in the sky can be used to measure the transverse size or the coherence or clustering scale of the absorbers. The experiment is conceptually simple: if Lyman $\\alpha$ absorption is seen at a similar velocity in two lines of sight and the chance probability of such a line match is small, then the transverse separation is a firm lower limit on the size of the cloud. Guided by the insights of supercomputer simulations, it is clear that the idea of the absorber as ``clouds'' with any simple geometry is naive. Even the notion of a characteristic size or radius is imprecise -- the absorbers trace an extensive and complex topology.  We use the term 'size' throughout this study but we realize that it could refer to the physical size of a single absorbing entity or the coherence length of an ensemble of smaller absorbing clouds.  Until a few years ago there were few size constraints on the hydrogen clouds. Common Lyman $\\alpha$ absorption in quasars that were gravitationally lensed had been used to derive lower limits to the sizes of a few tens of kpc \\citep{fol84, sme92}. The first evidence of large absorber sizes (hundreds of kpc) came from studies of the quasar pair LBQS 1343$+$264A,B \\citep{din94, bec94}. The characteristic size deduced from these studies, $\\sim$ 300 $h_{70}^{-1}$ kpc at $z \\sim 2$ ($q_0 = 0.5$), rules out simple absorber models such as cold dark matter mini-halos \\citep{ree86} and pressure-confined clouds \\citep{sar80}. Since then, a number of pairs have been observed to extend these experiments to larger transverse separations and lower redshifts \\citep{din95, elo95, fan96, cro98, din98, pet98}. These experiments are still limited by the modest number of lines available for any analysis of coincidence. it is still difficult to distinguish between models where the absorbers are (a) non-spherical, or (b) clustered, or (c) spherical but drawn from a non-uniform size distribution. In all these analyses the addition of a third sight-line adds important information about the two-dimensional structure of absorbers in the plane of the sky.  The next obvious step is to relate observations of the Lyman-$\\alpha$ absorbers to the results from the supercomputer simulations. This is non-trivial --- there are major differences in methodology between the ways observers and modelers measure absorber properties \\citep{mir96, cha97, dav99}. Observers have one dimensional path-lengths but cannot sample one contiguous volume even with multiple sight-lines. Supercomputer simulations can be used to measure large scale structures on all scales up to the maximum size of the experiment, about 15-40 $h_{70}^{-1}$ Mpc. Spectral extractions from the simulations have high spectral resolution, infinite signal-to-noise and a known continuum level, while observers must deal with photon statistics and the vagaries of continuum-fitting and line identification. Modelers know the physical state of the gas; observers must make operational definitions for the detection, profile-fitting, and de-blending of spectral features. Our long term goal is to make direct comparisons with the results of N-body simulations that incorporate gas dynamics \\citep{kat96}.  Although we defer a detailed interpretation of the absorber coincidences until a future paper, we make several improvements to the methodology of defining coincident lines herein. First, we impose no velocity window on the definition of a line match in an adjacent sight-line, in contrast to virtually every published paper in this subject. The rationale is that coincident line pairs should be selected without any {\\it a priori} kinematic assumption about the absorber. We also include cosmic variance in line counting in the determination of the errors associated with line matching, which has also been rarely done in the past. Finally, we employ a full Monte Carlo simulation of a random line matching experiment as a way of deriving reliable significance levels to the results presented here.  This paper presents new observations of the quasar pair LBQS 0107$-$025A,B, along with observations of a third quasar, LBQS 0107$-$0232, which creates baselines with up to 2.5 times the transverse separation of the original pair. The next section discusses the HST/FOS observations and data reduction procedure. We then proceed to describe the line matching procedure and the observed matches along with their significance as determined by comparison with Monte Carlo simulations of random populations of absorbers. The last section of the paper summarizes the implications of these observations for the geometry of the absorbers. The details of the Monte Carlo procedure are presented in Appendix A. ",
        "conclusions": "We have presented new Hubble Space Telescope observations of the quasar pair LBQS 0107$-$025A,B, along with observations of a third quasar, LBQS 0107$-$0232. \\citet{din97} found 5 coincident Lyman $\\alpha$ lines in the quasar pair. The new data increase that yield to 12 coincident lines, albeit with a more generous velocity window for declaring a match. The comparisons of the LBQS 0107$-$025A,B pair to the third quasar yield 8 and 9 matches, respectively. There are 3 triple matches within a velocity window of 550 km s$^{-1}$.  The significance of the coincident absorbers is established with a Monte Carlo simulation that lays down lines randomly in velocity space, with numbers that account for the Poisson variance in the number of absorbers seen per sight-line and the varying signal to noise between spectra. Simulations of random lines give a distribution of coincident pairs that peaks slightly towards zero velocity difference, with important implications for determining the significance of matches in this (and other) experiments.  The main results are, (1) coincident lines within $|\\Delta$v$|$$\\ <$ 200 $\\rm{km\\ s^{-1}}$ that are significant at the 90\\%\\ level for two quasar pairings, (2) coincident pairs of stronger lines preferentially have small velocity separations, and (3) a strong absorber that represents a highly significant triple coincidence.  The simplest interpretation of the double and triple coincidences is in terms of absorbers with coherence lengths spanning 0.5 to 1 Mpc at $0.4 < z < 0.9$ and extending in two dimensions on the plane of the sky. The correlation between line strengths for the coincident pairs is weak, so the absorbers must be extended but inhomogeneous. It is difficult to distinguish between single, coherent absorbers and clustered but distinct absorbers with a pair experiment. However, the detection of a highly significant match in all three quasars in the asterism indicates that at least one strong absorber has a sheet-like geometry. As more data on multiple quasar sight-lines accumulates, the line matching statistics can be used for an important new measure of the dimensionality of large scale structure at modest redshifts."
    },
    "0010/astro-ph0010214_arXiv.txt": {
        "abstract": "We collected optical and near IR linear polarization data obtained over 20--30 years for a sample of 51 blazars.  For each object, we calculated the probability that the distribution of position angles was isotropic.  The distribution of these probabilities was sharply peaked, with 27 blazars showing a probability $<$ 15\\% of an isotropic distribution of position angles. For these 27 objects we defined a preferred position angle. For those 17 out of 27 blazars showing a well-defined radio structure angle (jet position angle) on VLBI scales (1--3mas), we looked at the distribution of angle differences -- the optical polarization relative to the radio position angles. This distribution is sharply peaked, especially for the BL Lac objects, with alignment better than 15\\deg\\ for half the sample. Those blazars with preferred optical position angles were much less likely to have bent jets on 1--20mas scales. These results support a shock-in-jet hypothesis for the jet optical emission regions. ",
        "introduction": "Polarization observations have long been a very important probe of the internal structure of blazar jets.  Bright spatially resolved knots often show radio polarization ({\\bf E} vector) aligned with the projected jet direction, indicating a perpendicular magnetic field.  This suggests that shocks are responsible for compressing the jet magnetic field and accelerating the synchrotron-emitting electrons (e.g. Aller, Aller \\& Hughes 1985).  The relation between optical polarization and VLBI structure provides a unique tool for investigating the regions of jet formation on $\\la$ parsec scales. While previous statistical investigations have shown a tendency for optical polarization to be aligned with the jet (Impey et al. 1991, Rusk \\& Seaquist 1985), the interpretation of optical polarization is less clear because the emitting regions in blazars are not resolved, blazars often show violent short-term optical polarization variability, and the old radio observations did not have sufficient angular resolution to probe the region near the optically emitting core. A few quasi-simultaneous optical-VLBI observations indicate that the optical polarization is aligned with the direction of newly-ejected blobs at the highest VLBI resolutions (Gabuzda \\& Sitko 1996, Lister \\& Smith 2000). The optical polarization may originate in shocks at the base of the jet.  We re-address the question of optical alignment, taking advantage of a more extensive optical polarization database, and more and improved VLBI maps. ",
        "conclusions": "A natural explanation for the result that optical polarization tends to align with the jet, is that the optical synchrotron emission arises from a shock front in which the jet magnetic field has been compressed, on average, perpendicular to the jet. The large scatter in the optical polarization angles for a given blazar suggests that, in the inner jet region, the compressed magnetic field changes direction with time. Possible explanations are that the inner jet is internally unstable, or shocks may form via interaction with gas surrounding the central engine.  The jets of QSOs may be affected by gas in the NLR and BLR, present in QSOs but absent in BLLs. This may explain why QSOs show preferred optical polarization angles less frequently. The variations may be enhanced by the effects of foreshortening and relativistic beaming.   The tendency that objects with  well-defined preferred optical polarization directions also have very small VLBI scale jet bending indicates that a well-behaved straight jet on parsec to Kpc scales corresponds to a well-behaved jet on sub-parsec (optical) scales. Large curvature is likely to be the effect of projection of small jet curvature at very small viewing angles (Gower et al. 1982). Possible causes of jet curvature are [1] an interaction with the environment, or [2] an apparent curvature. In the first case, how does the base of the jet know about the environment on much larger scales? The angular resolution of optical observations is at best a factor of 100 worse than VLBI, often $> 100$mas. So we do not have direct evidence to test our assumption that the more energetic optical photons arise near the base of the jet. The optical emission could arise in the same shocks giving rise to cm-wavelength emission. The observation of rapid polarization variation at cm wavelengths, outside the core, gives credence to this idea (Gabuzda et al. 2000). In the second case, the direction of particle ejection may vary with time, for example, via a precession jet (e.g. Hummel et al. 1997). Present data are inadequate to address changes in optical polarization position angles on precession time scales."
    },
    "0010/astro-ph0010508_arXiv.txt": {
        "abstract": "A general method of deprojecting two-dimensional images to reconstruct the three dimensional structure of the projected object -- specifically X-ray, Sunyaev-Zel'dovich (SZ) and gravitational lensing maps of rich clusters of galaxies -- assuming axial symmetry (\\cite{zshs98}), is considered. Here we test the applicability of the method for realistic, numerically simulated galaxy clusters, viewed from three orthogonal projections at four redshift outputs. We demonstrate that the assumption of axial symmetry is a good approximation for the 3D structure in this ensemble of galaxy clusters. Applying the method, we demonstrate that a unique determination of the cluster inclination angle is possible from comparison between the SZ and X-ray images and, independently, between SZ and surface density maps. Moreover, the results from these comparisons are found to be consistent with each other and with the full 3D structure inclination angle determination. The radial dark matter and gas density profiles as calculated from the actual and reconstructed 3D distributions show a very good agreement. The method is also shown to provide a direct determination of the baryon fraction in clusters, independent of the cluster inclination angle. ",
        "introduction": "Recent years have witnessed significant improvements in the quality and quantity of data from clusters of galaxies in all wavebands, from the X-ray, through optical, to radio wavelengths. Together with advances in the theoretical modeling of the physical processes in clusters, these improvements call for new and more accurate methods of data analysis and comparison with various theoretical predictions.  In particular, uncovering the 3D structure of clusters is of great interest as it plays an important role in the determination of the Hubble constant from X-ray and SZ measurements (see Appendix), cluster mass and baryon fraction determination, and the underlying galaxy orbit structure. In order to reconstruct the 3D geometry of clusters researchers normally apply the spherical symmetry assumption, either within a parametric (\\eg\\ Cavaliere \\& Fusco-Femiano 1976) or non-parametric (Fabian \\etal 1981; and more recently Yushikawa \\& Suto 1999) framework. Although this simplifying assumption is very useful for many problems, it can lead to significant biases in others, \\eg\\ the Hubble constant determination (\\eg\\ Fabricant et al 1984).  Recently, a few studies have developed methods to take advantage of the full two dimensional information retained in the cluster images and extend the modeling of the underlying three dimensional structure to have a more general, aspherical distribution (\\eg\\ Zaroubi \\etal 1998; Reblinsky \\& Bartelmann 1999). In particular, Zaroubi \\etal (1998; hereafter referred to as paper I) presented a model-independent method of image deprojection for probing the 3D structure in clusters of galaxies, assuming only that the underlying 3D structure has axial symmetry.  The original testing of this method demonstrated that given images of the cluster in the X-ray, radio and a map of the projected total mass distribution, one could uniquely determine the inclination angle of the system symmetry axis, and determine the underlying 3D structure in the gas and dark matter (see also Grego \\etal\\ 2000). They studied the stability of the inversion method in the context of a specific analytical gas density profile: an axially symmetric, elliptical isothermal model; which is an extension of the widely used isothermal sphere model.  A few questions arise from the study in paper I: 1) Is the fundamental assumption of axial symmetry valid for real galaxy clusters?  2) Does substructure and dynamical evolution in galaxy clusters (as predicted to be abundant by hierarchical structure formation models) permit a robust, unique and stable numerical application of the method?  3) Can the cluster symmetry axis inclination angle be uniquely determined for more complex cluster gas/mass density distributions? 4) With realistic expected resolution and signal-to-noise from current observations, can the method be applied?  To probe these questions, we have performed a study of the inversion algorithm on a set of numerically simulated galaxy clusters. The goal is to test the axisymmetric deprojection algorithm, using as input a representative realization of the cluster population. We would like to emphasize at this stage that testing the applicability of the technique to current data sets is beyond the scope of the current paper. The method should be applied to each data set while taking into account its unique specifications, \\eg\\ resolution, scanning strategy, data acquisition domain (UV plane {\\it vs.} real space), etc. We intend indeed to tackle these issues for specific data sets in a forthcoming paper.   The outline of this study is as follows.  In \\S2 the deprojection method is briefly reviewed, and we develop a new feature of the method that allows an inclination angle-independent determination of the baryon fraction in clusters.  In \\S3, we briefly describe the input cluster simulations.  In \\S4, we show in detail the results for a single, prototype simulated cluster. In \\S5, we apply the method to the full cluster simulation output. The paper is concluded with a discussion (\\S6). ",
        "conclusions": "We have tested a method for deprojecting realistic galaxy cluster X-ray, SZ decrement and mass surface density maps, assuming only that the underlying gas/mass distributions are axially symmetric.  Conceptually, the deprojection procedure is straightforward: we first determine the image axis of symmetry, employing three different estimators accounting for the Poisson nature of the noise, and a technique giving increased weight to counts at large radii from the cluster center. Having a solution for the projected symmetry axis, we deproject the SZ, X-ray and surface mass distributions using equation (\\ref{eqn:maineqn}), allowing for the possibility that the cluster is either oblate or prolate.  As the inclination angle is the primary unknown, we deproject each image over a wide range of assumed inclination angles, searching for the angle that consistently yields the best reconstructed ratios between the Sunyaev-Zel'dovich, X-ray and the total mass deprojections.  At each inclination angle, we fill the information lost in the COI by either an elliptical isothermal model fitted to the data outside the COI, or a linear extrapolation into the COI with the amplitude fixed by the value at the cone boundary.  The results are fairly insensitive to the precise algorithm applied to fill the COI, primarily because the information lost inside the COI is restricted to moderately high spatial frequencies.  For this study, we demonstrated this procedure in depth for a single, prototype cluster, and then applied the method to the entire sample.  We demonstrated the goodness and robustness of the inclination angle determination in several ways: we compared the 3D gas density distribution as inferred from the SZ and X-ray deprojection, using the appropriate best fit angle for each case. The agreement between the two, in terms of both shape and amplitude, is very good (with a $5-10\\%$ difference at the center).   We also determined the inclination angle by calculating the cumulated, normalized absolute difference between the inferred and true gas densities as a function of inclination angle (see Figure~\\ref{fig:chi2vstheta}). The results agreed with our previous determinations to an accuracy of $\\lsim \\pm 5\\deg$, demonstrating the reliability of the inclination angle determination even in the case of a very wide COI.  With the prototype cluster, we mimic observations of a rich galaxy cluster by CHANDRA (X-ray) and BIMA or OVRO (SZ) by degrading the resolution of the simulated data and adding noise. We again compared the deprojections of the two maps to determine the inclination angle and found $26^\\circ$ (assuming a prolate cluster) which is within $6^\\circ$ from the angle obtained from noiseless images; this is an excellent agreement that shows the potential of this method when applied to real data.  We note here, that the purpose of this exercise is to show that the method has the potential of reproducing the 3D structure from the current data sets, this however does not mean that we have tested the application of the method to those data sets. The application of the method to each of those data sets and its unique specifications should be tested much more extensively, something we intend to carry out in a forthcoming study.  We also note, that since the current method is based on Fourier slice theorem it is well suited to analyze interferometric data, like the data obtained by the BIMA and OVRO experiments, where the raw data is given in the Fourier space U-V plane. For this kind of data one can for example, manipulate the X-ray or lensing data and transform them to the Fourier U-V plane and make the comparison with the SZ raw data at the wavenumbers covered in this plane which has the advantage of avoiding the U-V plane extrapolation needed in order to obtain a real space SZ image.   Finally, with the prototype cluster we tested the accuracy of the deprojection by comparing the true and inferred 3D radial profiles of the gas and total mass distributions (Figure~\\ref{fig:spherical}). We found that the reconstruction underestimated the density at the innermost center of the cluster but it yields a very accurate profile at radii $\\gtrsim 0.3 \\; \\Mpc$ for the gas density profile and $\\gtrsim 0.5 \\, \\Mpc$ for the dark matter density profile. The discrepancy between the real and reconstruction radial profiles in the innermost region, especially in the dark matter radial profile, is primarily due to the loss of the high frequency information in the COI. However, the {\\em total}, estimated over the whole the simulation volume, gas and mass differences are only of the order of few percent.  We also developed a feature of the method that provides a direct determination of the baryon fraction in clusters, independent of the cluster inclination angle. The only uncertainty comes from whether the underlying cluster gas density distribution is assumed to be prolate or oblate. For all of the clusters studied here, we note that the relative variations in the radial dependence for the baryon fraction for the prolate case are much smaller than if the cluster is oblate (see Figures \\ref{fig:z00_densityratio}, \\ref{fig:z03_densityratio}, \\ref{fig:z06_densityratio}, and \\ref{fig:z09_densityratio}).  The shape of the baryon fraction radial profile provides an interesting possibility for discriminating between the prolate/oblate hypotheses: Due to the expected strong correlation between the gas and dark matter distributions, one expects their density and mass ratios to show relatively small variability as a function of radius from the cluster center. Indeed, for low- to intermediate redshift clusters $(z \\lesssim 0.5)$, the baryon fraction has a relatively slight radial dependence $f_b(r) \\propto r^{0.2}$ (\\cite{ettori99}).  In conclusion, we have demonstrated that: 1) The fundamental assumption of axial-symmetry is a reasonable assumption when applied to a realistic subset of the general cluster population. 2) Under the assumption of axial symmetry, the method works very well for the purpose of recovering the 3D structure. 3) For a reasonable COI extrapolation scheme, the deprojection is robust, stable and unique even when most of k-space information has been lost in the projection. 4) The method is applicable for deprojecting cluster maps throughout the evolutionary history."
    },
    "0010/astro-ph0010022_arXiv.txt": {
        "abstract": "Details of a theory of self-broadening of hydrogen lines are presented.  The main features of the new theory are that the dispersive-inductive components of the interaction (van der Waals forces) have been included, and the resonance components have been computed by perturbation theory without the use of the multipole expansion.  The theory is applied to lower Balmer lines and the theoretical and observational impact of the new broadening theory is examined.  It is shown that this theory leads to considerable differences in the predicted line profiles in cool stars when compared with previous theories which include only resonance interactions. In particular, the effect is found to be very important in metal poor stars. The theory provides a natural explanation for the behaviour of effective temperatures derived from Balmer lines by others using a theory which includes only resonance broadening. When applied to Balmer lines in the solar spectrum the theory predicts an improved agreement between observed and computed profiles for models which also match limb darkening curves and rules out a model which does not. However significant discrepancies still remain which could be due to inadequacies in our theory or the atmospheric model or both. ",
        "introduction": "Hydrogen line wings are one of the strongest tests of model stellar atmosphere structure. In the majority of stars hydrogen is a dominant source of continuous opacity and thus for strong hydrogen lines the abundance parameter is excluded.  In hot stars, where the broadening of hydrogen lines is dominated by protons and electrons produced by the ionisation of hydrogen, the hydrogen line profiles are dependent only on atmospheric structure and properties of the hydrogen atom.  In cool stars the metallicity is important as ionisation of metals is the principal source of ions and electrons which contribute to the Stark broadening of the lines while hydrogen atoms in their ground state produce self-broadening of the lines.  Further, hydrogen lines can be observed in all stars, unlike for example limb darkening curves which are one of the strongest tests for solar photosphere models.  The large range of opacities within a single line means that many different depths are probed.  If the behaviour of the hydrogen atom in the conditions of stellar atmospheres is understood, the hydrogen absorption lines can be a powerful diagnostic.  The transition probabilities are known with extremely high accuracy.  Furthermore, the line profile shape is particularly sensitive to atmospheric structure, due to the unique situation of the broadening which derives from the ``accidental degeneracy'' of states in the hydrogen atom.  However, this degeneracy means that the broadening is extremely complex by comparison with metallic lines.   Recently there have been a number of applications of hydrogen lines, particularly lower Balmer lines, to the analysis of stellar photospheric models, and in particular models of stellar convection (Fuhrmann~et~al.~\\cite{fuhrmann1,fuhrmann2}, Van't Veer-Menneret \\& M\\'egessier~\\cite{vanV}, Castelli~et~al.~\\cite{castelli}, Gardiner~et~al.~\\cite{gardiner}).  Of course, such analyses will be dependent on the accuracy of the theories describing the hydrogen atom properties.  Of particular importance for analyses of photospheres is the broadening of the wings in stellar photospheric conditions.  This is especially true when testing convection theories, as convection affects the atmosphere in the deeper layers which do not contribute to the core of the line.  It was pointed out by Lortet \\& Roueff~(\\cite{lortet_roueff}) that the effect of neglecting dispersive-inductive forces should be significant, however, this seems to have gone largely unnoticed. They showed that relative to resonance broadening, dispersive interactions make a significant contribution to the self-broadening of Balmer lines.  For Paschen lines they demonstrated that dispersive interactions should dominate the self-broadening.  However, in their analysis they used an inadequate theory of the dispersive-inductive interaction which is known to underestimate this type of broadening by typically a factor of two. In this paper a theory of self-broadening of hydrogen lines which includes a better treatment of both resonance and dispersive-inductive interactions, which was announced in an earlier letter (Barklem~et~al.~\\cite{bpo:let}, hereafter Paper I), is presented. It is shown that the inclusion of these interactions has a significant effect on the predicted profiles of Balmer lines in cool stars. ",
        "conclusions": "We have presented a theory of self-broadening of hydrogen lines, which includes both resonance and dispersive--inductive interactions. The theory was used to synthesise Balmer lines in cool stars and shown to make a considerable difference compared to the commonly used Ali \\& Griem~(\\cite{ali_griem:errata}) theory which does not include the dispersive-inductive interactions.  The new theory perhaps explains behaviour observed by Gardiner~et~al.~(\\cite{gardiner}) and Castelli~et~al.~(\\cite{castelli}) when using Balmer lines and Ali \\& Griem~(\\cite{ali_griem:errata}) theory to obtain effective temperatures for stars including the sun. It is superior to Ali \\& Griem~(\\cite{ali_griem:errata}) theory when applied to Balmer lines in the solar spectrum as it reduces the discrepancy between observed and computed profiles when the HOLMUL model is used and leads to the KNOVER model being classed as unacceptable both for its failure to adequately model the observed limb darkening and Balmer line profiles. However significant discrepancies between theory and observation for the Balmer lines still exist for the KOVER and HOLMUL models which could be due to our theory or the photospheric models.  Work is in progress on the determination of the effective temperatures of a sample of dwarf stars using our theory and observed Balmer lines for the stars in the sample. Preliminary results indicate a stronger correlation between the effective temperatures obtained from the Balmer lines and the effective temperature obtained by other methods such as the infrared flux method when our theory is used compared with that found when Ali \\& Griem~(\\cite{ali_griem:errata}) theory is used.   We plan to extend the theory to Paschen lines. We expect that in these lines dispersive-inductive interactions will dominate resonance interactions but not Stark broadening. Many of the approximations made in developing the theory are only valid in cool stars with weak quasistatic ion fields. For stars of earlier spectral type it may be necessary to develop a Unified Theory to correctly include self-broadening. The true limit of validity of the impact approximation in H$\\alpha$ and H$\\beta$ needs to be established."
    },
    "0010/astro-ph0010352_arXiv.txt": {
        "abstract": "We develop a method for estimating the shear power spectra from weak lensing observations and test it on simulated data.  Our method describes the shear field in terms of angular power spectra and cross correlation of the two shear modes which differ under parity transformations.  Two of the three power spectra can be used to monitor unknown sources of noise in the data.  The power spectra are decomposed in a model independent manner in terms of ``band-powers'' which are then extracted from the data using a quadratic estimator to find the maximum  of the likelihood and its local curvature (for error estimates).  We test the method against simulated data from Gaussian realizations and cosmological $N$-body simulations. In the Gaussian case, the mean bandpowers and their covariance are well recovered even for irregular (or sparsely) sampled fields.  The mild non-Gaussianity of the $N$-body realizations causes a slight underestimation of the errors that becomes negligible for scales much larger than several arcminutes and does not bias the recovered band powers. ",
        "introduction": "Weak lensing of background galaxies by foreground large-scale structure has now been convincingly detected (\\cite{BacRefEll00} 2000; \\cite{KaiWilLup00} 2000; \\cite{vanetal00} 2000; \\cite{Witetal00} 2000) and is rapidly becoming a valuable tool for studying the distribution and clustering evolution of dark matter in the universe. Lensing produces a correlated distortion of the ellipticities of background galaxies, at the percent level, which can be used to measure a two-dimensional projection of the mass distribution.  While there are numerous observables that can be defined from the shear maps produced by such surveys, one of the most important is the angular power spectrum.  The angular power spectrum contains valuable information on cosmological parameters that complements other astrophysical measurements (\\cite{JaiSel97} 1997; \\cite{Kai98} 1998; \\cite{HuTeg99} 1999). In this respect weak lensing is very similar to anisotropies in the CMB and in fact much of the theory and data analysis is very analogous as well.  In this paper, we suggest a technique for the presentation and interpretation of weak lensing data which is commonly employed in CMB analysis: the use of ``band-powers'' extracted from the data by an iterated quadratic estimator of the maximum likelihood solution. This technique has the advantage of automatically taking into account irregular survey geometries and varying sampling densities.  It provides an optimal estimate of the power spectrum in the Gaussian regime and makes efficient use of all of the data on the relevant angular scales. Error estimates that include the sampling and noise variance of the survey are also automatically produced by this method. Similar claims cannot be made for methods based on correlation functions or simple Fourier Transforms of the data.  Throughout our focus will be on weak lensing by large-scale structure, specifically on angular scales larger than several arcminutes (or wavenumbers $l \\la 10^3$). These are the scales which will be probed by ongoing wide field surveys, e.g.~the Deep Lens Survey\\footnote{http://dls.bell-labs.com}, surveys with the VLT and the Hawaii/IfA lens survey. On these angular scales the signal reduces simply to a projection of the density contrast along the line of sight (\\cite{BlaSauBraVil91} 1991; \\cite{Mir91} 1991; \\cite{Kai92} 1992), which is well approximated by a Gaussian probability distribution.  In this limit the angular power spectrum encodes all of the relevant information about the field, and in particular can be predicted from the 3D power spectrum of the density field using Limber's equation. On subarcminute scales, the field becomes substantially non-Gaussian, lens-lens coupling and perturbations to the photon trajectory become increasingly important and the weak lensing approximations break down. On these angular scales the signal is dominated by individual objects along the line of sight (e.g.~clusters of galaxies) and a correlation function or power spectrum analysis becomes less useful.  The outline of this paper is as follows.  In \\S \\ref{sec:theory} we describe the shear signal covariance matrix and its relationship to the three fundamental shear power spectra. We then develop  (\\S\\ref{sec:likelihood}) and test (\\S\\ref{sec:tests}) the iterated likelihood method for band power extraction using simulated data.  We conclude in \\S\\ref{sec:conclusions}.  \\begin{figure*}[tbh] \\centerline{\\epsfxsize=6truein\\epsffile{shear.eps}} \\caption{\\footnotesize The fundamental shear modes and their cross correlation. (a) A pure $\\epsilon$-field obtained from a convergence map from \\protect{\\cite{WhiHu00}} (2000).  (b) A pure $\\beta$-field obtained by a rotation of the shears by $\\pi/4$.  (c) A correlated mixture of $\\epsilon$ and $\\beta$ with $C_l^{\\epsilon\\epsilon}=C_l^{\\beta\\beta}=C_l^{\\epsilon\\beta}$ obtained by a rotation of the shears by $\\pi/8$.} \\label{fig:shear} \\end{figure*} ",
        "conclusions": "\\label{sec:conclusions}  On large angular scales the shear field induced by weak gravitational lensing of background galaxies by large-scale structure is close to Gaussian.  In this regime the relevant information is encoded in the angular power spectrum, $l(l+1)C_l/(2\\pi)$. We have suggested a technique, commonly used in CMB analysis, for determining the angular power spectrum in this regime: the use of ``band-powers'' extracted from the data by an iterated quadratic estimator (\\cite{BonJafKno98} 1998) of the maximum likelihood solution. This technique has the advantage of automatically taking into account irregular survey geometries and varying sampling densities.  It provides an optimal estimate of the power spectrum which makes efficient use of all of the data on the relevant angular scales.  We have tested the technique against simulated Gaussian and realistically non-Gaussian data, regular and irregularly sampled data, and with known and unknown amplitudes of shot noise from the intrinsic ellipticities of galaxies. In all cases, the mean band powers are recovered correctly.  The technique introduced here is a result of one of many possible cross-fertilizations of CMB and weak lensing research.  Indeed the experience gained in measuring the shear power spectra from noisy windowed data may feed back into the analogous problem for future CMB polarization studies. Techniques for handling large CMB data sets where the likelihood algorithm used here becomes prohibitively expensive (e.g.~\\cite{WanHivGor} 2000; \\cite{Szapudi} 2000) will also be useful to lensing studies as the lensing fields become ever larger.  \\smallskip {\\it Acknowledgments:} W. Hu and M. White were supported by Sloan Fellowships. M.~White was additionally supported by the US National Science Foundation.  \\smallskip"
    },
    "0010/astro-ph0010487_arXiv.txt": {
        "abstract": "We have carried out an analysis of about 160 eclipse timings spanning over 30 years of the Hyades eclipsing binary V471 Tauri that shows a long-term quasi-sinusoidal modulation of its observed eclipse arrival times. The O--Cs have been analyzed for the ``light-time'' effect that arises from the gravitational influence of a tertiary companion. The presence of a third body causes the relative distance of the eclipsing pair to the Earth to change as it orbits the barycenter of the triple system. The result of the analysis of the eclipse times yields a light-time semi-amplitude of $137.2\\pm12.0$~s, an orbital period of $P_3=30.5\\pm1.6$~yr and an eccentricity of $e_3=0.31\\pm0.04$. The mass of the tertiary component is $M_3\\sin i_3\\simeq0.0393\\pm0.0038$~M$_{\\odot}$ when a total mass of $1.61\\pm0.06$~M$_{\\odot}$ for V471 Tau is adopted. For orbital inclinations $i_3\\ga35$\\arcdeg, the mass of the third body would be below the stable hydrogen burning limit of $M\\approx0.07$ M$_{\\odot}$ and it thus would be a brown dwarf. In the next several years (near maximum elongation), it should be feasible to obtain IR images and spectra of V471~Tau~C that, when combined with the known mass, age, distance, and $[Fe/H]$, will serve as a benchmark for understanding the physical properties and evolution of brown dwarfs. ",
        "introduction": "Over a hundred brown dwarf (BD) candidates have now been discovered primarily from IR surveys such as 2MASS, DENIS, SDSS, and others (see Kirkpatrick et al. 2000). These objects have been classified as BDs from their spectra (e.g., presence of lithium and methane) and IR colors. A spectroscopic classification scheme recently proposed by Kirkpatrick et al. (1999) divides these objects into T-type (cooler BDs, methane BDs) and L-type (hotter BDs). A number of different models have been published to explain the photometric and spectroscopic characteristics of BDs (e.g., Burrows et al. 1997; Chabrier et al. 2000) but a unified scenario is yet to be provided. Unfortunately, nearly all of the BD candidates are free-floating objects in which it is not possible to determine dynamical masses. Therefore, the major problem confronting these models is the paucity of information on the physical properties of the BDs. For example, the ages and luminosities of solitary BDs in nearby clusters (e.g., Pleiades) are independently known, as well as their spectroscopic and photometric properties, but, regrettably, no direct measures of the masses of the candidates (the primary variable that determines the properties and the evolution of any object) are possible.  For a dynamical mass of a BD to be known, it has to be a member of a binary or multiple system. There are only a handful of cases where BDs have been imaged around nearby stars (see Burgasser et al. 2000), the most well-known of these is Gliese 229B. This wide ($\\sim$8\\arcsec) astrometric binary has the first {\\em bona fide} cool (T-type) BD (Nakajima et al. 1995), and has been studied by many observers. However, the long orbital period of Gl 229B would require centuries to determine dynamically (i.e., directly) a definitive mass. Even though the precision of current Doppler techniques in planet search surveys is higher than needed for the discovery of BDs, only a few candidates have been found so far.  This indicates a probable low frequency of short period BDs orbiting main sequence stars (Halbwachs et al. 2000). In any case, these objects fail to meet the requirements for being critical to understanding the physics and structure of BDs because: their observed masses are based on model-dependent calibrations, the orbital inclination is not known, and, also, the glare of the bright component makes direct imaging impossible.  In addition to direct imaging and radial velocity techniques, eclipsing binaries showing a ``light-time'' effect offer an opportunity for detecting low-mass companions. The light-time effect occurs as the relative distance (and light travel time) changes as the eclipsing binary moves around the barycenter of the triple system. The detection of low-mass objects orbiting eclipsing binary stars is of great interest because the mass of the tertiary component can be directly determined (the analysis of the light and radial velocity curves of the eclipsing binary yields fundamental masses for the stars). Also, it is often possible to estimate the age of the system via isochronal fits or using other indicators such as cluster membership. If direct imaging of the tertiary component was possible (long period orbit), the orbital inclination, the mass, the age, the photometry, and the spectroscopy of the third component could all be simultaneously obtained.  The 9th mag Hyades eclipsing binary V471 Tauri (BD+16$^{\\circ}$516; HIP17962; $P=12.5$~hr) has been under intense study at nearly all wavelengths since its discovery in 1969 (Nelson \\& Young 1970). V471 Tau is a detached binary that consists of a hot DA2 white dwarf and a chromospherically active K2~V star (Guinan \\& Sion 1984). This system plays a key role in deciphering binary star evolution (post common envelope and pre-cataclysmic variable phases) as well providing fundamental information on the formation, structure and evolution of white dwarfs. V471 Tau also serves as a laboratory for studying dynamos and magnetic activity of cool stars. For example, during the primary eclipse, the hot white dwarf acts as a beaming probe of the K2 star's atmosphere (see Guinan et al. 1986). In this {\\it Letter}, we present yet another aspect that demonstrates the importance of V471~Tau. From the analysis of over 30 yr of eclipse timings, V471 Tau shows periodic variations of the arrival times of its primary eclipse.  Such variations appear to arise from the light-time effect caused by the gravitational pull of a low mass tertiary companion that could likely be a BD with a mass of $\\sim0.04$~M$_{\\odot}$. ",
        "conclusions": "The results presented in the previous section yield strong evidence for the presence of a tertiary component in a long-period orbit around V471~Tau. Adopting a total mass for the eclipsing binary of $M_1+M_2= 1.61\\pm0.06$~M$_{\\odot}$ ($M_{\\rm K2}=0.79\\pm0.04$~M$_{\\odot}$, $M_{\\rm WD}=0.82\\pm0.04$~M$_{\\odot}$; E. M. Sion, priv. comm.), the mass of the third body is $M_3 \\sin i_3\\simeq0.0393\\pm0.0038$~M$_{\\odot}$ and its orbital semi-major axis is $a_3\\simeq11.2\\pm0.4$~AU. The mass of the third component as a function of the inclination (for several selected values) is provided in Table \\ref{tab3}. As can be seen, as long as the inclination of the orbital plane of the tertiary component is $\\ga35$\\arcdeg, the mass of this object is below the theoretical threshold of $\\sim0.07$~M$_{\\odot}$ for a hydrogen burning star. Even for the minimum mass, however, the third body would be probably too massive to be a planet. The predicted orbit projected on the sky as a function of the orbital inclination has also been calculated. Plots for two different values of $i_3$ are shown in Figure \\ref{fig3}. In these calculations a distance of $46.8\\pm3.5$~pc determined from Hipparcos astrometry of V471 Tau (ESA 1997) has been adopted. The maximum apparent separation of the tertiary component from the eclipsing pair reaches $\\approx 260$~mas near one of the quadratures of its orbit.  Thus, according to our calculations, V471~Tau~C is likely to be a BD for which the mass, age, and chemical composition can be well known. The age and chemical composition of the system are known from its membership in the Hyades cluster (Barstow et al. 1997). Recent determinations of the Hyades age yield a value of $625\\pm50$~Myr (Perryman et al. 1998). From its mass, age, and $[Fe/H]$, most of the observable properties of the object can be estimated from theoretical models. We adopted very recent evolutionary models of Chabrier et al. (2000) that consider non-gray atmospheres with dust formation and opacity. The predictions for several values of $i_3$ (obtained from a double linear interpolation in age and mass) indicate values of $T_{\\rm eff}=1370$~K, $\\log g=5.0$, and $M_{\\rm K}=12.7$~mag for the minimum mass of the third body (see Table \\ref{tab3}). The apparent K-band magnitude of the low-mass object would be $m_{\\rm K} \\la 16.0$ mag. The BD companion is therefore expected from theory to be $\\approx9$~mag fainter than the eclipsing binary in the K-band (V471 Tau: $m_{\\rm V}=9.43\\pm0.05$~mag, $m_{\\rm K}\\simeq7.21\\pm0.10$~mag).  We have proposed HST/{\\em Fine Guidance Sensors} astrometry (to be combined with existing Hipparcos data) to ascertain the presence and determine the properties of the tertiary component. When these astrometric measurements are available (covering half of the orbital period) they will unambiguously yield the orbital inclination and the semi-major axis with an error below 0.5 mas, corresponding to a few percent uncertainty of the tertiary object's mass. This will represent the first direct dynamical mass determination of a BD with known age, chemical composition, and distance. Moreover, in the next several years it should be feasible to image directly V471 Tau C in the IR (using coronographic observations made with adaptive optics or observations made from space) as it moves toward maximum angular elongation from the eclipsing pair. Once we carry out these observations, and if V471 Tau C is indeed confirmed to be a BD, it will make an excellent benchmark for understanding the properties and evolution of BDs."
    },
    "0010/astro-ph0010164_arXiv.txt": {
        "abstract": "\\noindent Observations show significant H$\\alpha$-emissions in the Galactic halo near the edges of cold gas clouds of the Magellanic Stream. The source for the ionization of the cold gas is still a widely open question. In our paper we discuss the critical velocity effect as a possible explanation for the observed H$\\alpha$-emission. The critical velocity effect can yield a fast ionization of cold gas if this neutral gas passes through a magnetized plasma under suitable conditions. We show that for parameters that are typical for the Magellanic Stream the critical velocity effect has to be considered as a possible ionization source of high relevance. ",
        "introduction": "The Magellanic Stream (MS) consists of a narrow band of high-velocity neutral hydrogen gas (HI) which stretches over almost 100$^{o}$ on the sky, trailing the Magellanic Clouds in their orbit around the Milky Way (Mathewson, Cleary \\& Murray 1974; Mirabel, Cohen \\& Davies 1979). The Stream was observed at 21 cm. Thereby a chain of clouds connected by a gas of lower density was identified. The clouds have been labeled MS I through MS VI (Mathewson, Schwarz \\& Murray 1977). They are characterized by a high density concentration with a relatively sharp gradient on the leading edge, which is defined by the direction of the proper motion of the Large Magellanic Cloud (Jones, Klemola \\& Lin 1994). The Stream has no known stellar component (Mathewson et al. 1979) and its distance to the Milky Way is unknown, although its leading end (MS I) connects to the Magellanic Clouds and is presumably located at a distance of 50- 60 kpc. Moore and Davis (1994) estimate the distance of the tip of the Stream (MS VI) to be 20 kpc, whereas Gardiner, Sawa \\& Fujimoto (1994) give a distance value of 60 kpc.  Bregman (1979) considered the MS as the hydrodynamic result of the tidal interaction between the Magellanic Clouds and the gaseous halo of the Milky Way (Murai \\& Fujimoto 1980; Lin \\& Lynden-Bell 1982). This tidal scenario is supported by an extension to the model (Moore \\& Davis 1994) which was able to reproduce the kinematics of the Stream  by including the stripping associated with an extended ionized disk and drag due to a diffuse halo. Nevertheless, the models for the origin of the MS remain controversial (Wakker \\& van Woerden 1997; Murali 2000), so that the inferences about halo gas properties are correspondingly uncertain.  What do we know from observations about the characteristic values for the density and temperature of the MS?   The northern part of the MS (MS III - MS MVI) has been surveyed with high sensitivity by Mirabel, Cohen \\& Davies (1979) in order to study the small scale structure. Five regions were investigated, which are characterized by narrow filaments running parallel to the main direction of the Stream, and small cloud-like concentrations. These clouds have typical masses of 15 $D^2 M_\\odot$ (here D is their distance in kpc). The authors conclude that the clouds appear to be unstable over the lifetime of the Stream as a whole and some external confinement mechanism seems to be necessary to preserve the observed small scale structure. This could be provided by a hot plasma of density $\\sim 4\\cdot 10^{-4}\\, {\\rm cm}^{-3}$ at a temperature of $10^6\\, $K.  Weiner \\& Williams (1996) detected faint H$\\alpha$-emission from several points of the MS, i.e. the leading edges of the HI clouds MS II, MS III and MS IV. The emission is strongest (emission measures EM: 0.5-1 ${\\rm cm}^{-6}$ pc) on the sharp leading-edge density gradient. They suggest that ram pressure from the halo gas plays an important role in stripping the Stream out of the Magellanic Clouds and they suggest the existence of a relatively dense ($n_H\\sim 10^{-4}\\,  {\\rm cm}^{-3}$) and almost fully ionized gas with a temperature of $\\sim 1.7 \\times 10^6\\, $K in the Galactic halo at a radius of 50 kpc.  Weiner and Williams (1996) rule out photoionization as a cause for the H$\\alpha$-emission. They argue that the HI column densities in all observed fields are optically thick to ionizing radiation which means that any uniform ionizing flux should produce ionization rates and H$\\alpha$ emission which are roughly constant from field to field, which is not observed. The low number of required UV-photons ($10^{48}\\, {\\rm s}^{-1}$) could, in principle, be provided by a few OB stars located  $\\sim 100$ pc away from the observed field. However, such stars that would be easily detectable by the associated luminous HII-region are not observed.   Murali (2000) recently discussed some theoretical implications. He reexamined the interaction of the MS with the surrounding medium at large distances in the halo of the Galaxy. He was able to show that heating dominates over drag forces. Due to their high relative velocities ($\\ge  200$ km ${\\rm s}^{-1}$) the HI-clouds should evaporate if the density of the halo was too high. Thus, he concluded a limit for the halo density at 50 kpc to be less than $10^{-5}\\, {\\rm cm}^{-3}$.  The strength and orientation of the magnetic field in the halo of our Galaxy is still a matter of debate (e.g. Vallee 1998). According to Beuermann, Kanbach \\& Berkhuijsen (1985) the scale height of the field is $\\sim 6 $kpc. The distance of the MS is several tens of kpc. Given a disk magnetic field of about $6\\mu$G and taking into account local field compression by the MS clouds the magnetic field in the MS is expected to be of the order of several $\\mu$G. Amplifications of magnetic fields by infalling gas clouds is well-known from high velocity clouds (e.g. Crutcher 1999; Vallee 1998).  The interaction of the high velocity clouds (HVC) of the MS and the medium of the galactic halo is an example for the interaction of a fully ionized plasma and a cold weakly ionized gas. Such a situation has been considered by Alfv\\'en (1954) in connection with his model for the origin of the solar system. He proposed that when  a neutral gas streams through a plasma across magnetic field lines at a velocity that exceeds a certain value $v_c$ a discharge like process can occur in which the neutral gas component is rapidly ionized. The critical velocity is proportional to the ionization potential and inversely proportional to the square root of the mass of the neutral atom. This effect is usually considered in the context of plasmas in which the ion-neutral collision rate is negligible. This is exactly the case in the interaction regions of the MS HI-clouds and the galactic halo. Given the problematic issue of lacking ionization sources for the H$\\alpha$, the critical velocity effect (CVE) presents an interesting fundamental plasma process that is capable of providing the required ionization. The CVE occurs when a neutral gas streams through a plasma across magnetic field lines with a velocity that exceeds a critical value measured by the ionization potential of the neutrals. The kinetic energy of the flow is partly converted to heating the electrons which can efficiently ionize the neutral gas component.  It is the aim of our contribution to discuss the CVE in the astrophysical system MS-Galactic halo as an alternative ionization mechanism far away from standard ionization sources like stars.  In Sect. 2 we review the basic conditions for the CVE to work. We give limits for the magnetic field strength, neutral gas, plasma density, and plasma temperature. Sect. 3 presents the application to the ionization of the MS cloudlets, followed by Sect. 4 which contains the discussion and some conclusions, in particular, on the general applicability of the CVE-process in astrophysical plasma-neutral gas interactions. ",
        "conclusions": "In our paper we discussed the critical velocity effect as a possible explanation for the observed H$\\alpha$ emission at the leading edges of cold cloudlets in the MS. For suitable parameter regimes this effect can yield  a rapid ionization of neutral gas streaming through a magnetized plasma.  Our work has been motivated by intensive studies of the CVE with respect to observed ionization phenomena in laboratory and space plasmas. In laboratory plasmas ionization caused by the CVE has been clearly identified in numerous experiments (e.g. Fahleson 1961; Danielsson \\& Brenning 1975; Piel et al. 1980). The conditions, however, under which the CVE becomes sufficiently efficient in space plasmas are still under debate.  Ionization due to the CVE has been clearly identified in context with a space experiment where an artificial cloud of barium was released in the Earth's magnetosphere (Haerendel 1982). Besides, ion production in the Io plasma torus around Jupiter and ionization in Io's ionosphere are interpreted as the consequences of the CVE (Cloutier et al. 1978; Galeev \\& Chabibrachmanov 1983).  Further ionization phenomena in space which are discussed to be the consequence of the CVE are the space shuttle glow (Papadopoulos 1984) and, especially, ionization in cometary environments due to the interaction between neutral gas sublimented from the cometary nuclei and the magnetized solar wind plasma (Formisano et al. 1982).  All these examples yielding evidence for the importance of the CVE refer to plasma - neutral gas interactions in laboratory and space plasmas which differ from a large scale astrophysical system like the MS in several aspects. However, the presented first quantitative studies of the CVE considering parameter regimes being typical for cloudlets in the MS strongly indicate that the CVE can yield an efficient ionization of the neutral gas at the boundaries of cold cloudlets in the halo plasma.  Thus, our kinematical studies have shown that the CVE has to be considered as a possible explanation for the observed H$\\alpha$ emissions. To get deeper insight into the question of to what extent the CVE can contribute to the ionization of halo clouds detailed dynamical multifluid studies are required.        \\ \\\\ \\ \\\\   {\\sl Acknowledgments.} This work has been supported by the Deutsche Forschungsgemeinschaft through the grant LE 1039/5-1."
    },
    "0010/astro-ph0010047_arXiv.txt": {
        "abstract": "We summarize the current observational status of the interstellar medium in terms of the McKee/Ostriker model. We find both agreement and disagreement---hence the title of the paper. Some of the newest observations differ from older ones,  which underscores the need for both more and better data. ",
        "introduction": "In 1977 Mckee \\& Ostriker (MO) extended the two-phase model of Field, Goldsmith, \\& Habing (1969; FGH) by including supernova and wrote their famous paper entitled ``A theory of the interstellar  medium - Three components regulated by supernova explosions in an inhomogeneous substrate''.   Their ISM is dominated by individual supernova (SN) explosions; today some believe we should replace the individual explosions by correlated ones producing superbubbles.  The Hot Ionized Medium (HIM) fills the interior of SN remnants and powers their blast waves, which sweep up the gas and magnetic field inside the bubble and pile them into the shell.  Soon this shocked gas starts to cool and recombine rapidly, forming the Cold Neutral Medium (CNM) with an amplified field.  Soft X-rays produced by immediately adjacent HIM penetrate the outsides of CNM clouds, heating the gas to form the partially ionized Warm Neutral Medium (WNM), whose more partially ionized portions are the Warm Ionized Medium (WIM). Their model is impressive in its physical development and internal consistency.  It makes many predictions and stands as the accepted paradigm against which most discussions and interpretations are compared.  The FGH and MO models rely on a broad range of physical processes, ranging from atomic interactions with energetic particles to gasdynamics, and the uninitiated may be easily overwhelmed. Fortunately, there is a straightforward path to nirvana, namely excellent, classical reviews that remain useful for experts. Dalgarno \\& McCray (1972) present the physics of heating and cooling for atomic regions, the basis for the FGH model. MO's paper is a bit intimidating  for many of us, but their Figures $1 \\rightarrow 3$ and Table 1 present an excellent three-minute overview of the basic picture. McCray \\& Snow (1979) present a highly readable discussion of the details and basic physics behind all aspects of the MO model, together with some critical comments that remain relevant today.  At the first Tetons meeting, Kulkarni \\& Heiles (1987; KH) interpreted ISM observations of the cool and warm phases in terms of the MO paradigm with great success. Cox (1995) presents a historical overview of our ISM understanding, and lack of it, and emphasizes some specific problems with the MO model.  The MO model doesn't cover everything. The magnetic field and cosmic rays play important roles in the overall dynamics of the ISM. Their pressures are larger than the typically adopted thermal pressures of the gas components.  Even for the large-scale $z$-distribution, where the pressure gradients may be small, their pressures contribute to the hydrostatic balance (Boulares \\& Cox 1990).  Also, the magnetic field links the different gas phases with each other, and also with the cosmic rays, so that components cannot act independently.  Rather, analysis of the dynamics and equilibria must include the whole ensemble, and this leads to new modes of behavior such as magnetically-linked clouds (Elmegreen 1994) and the Parker (1966) instability.  As important as these two nonthermal, non-material components are, we will not emphasize them in this review.  Rather, we will define and review {\\it five} (not MO's three) diffuse gas phases. These are the HIM, the HII-region like Reynolds WIM (RWIM), the MO-like WIM (MOWIM), the WNM, and the CNM. We begin with a definition of and brief introduction to the five phases in \\S2 and then go on to discuss various aspects in more detail. ",
        "conclusions": ""
    },
    "0010/astro-ph0010337_arXiv.txt": {
        "abstract": "s{Summary and discussion of some projects to use SCUBA to target sources selected at other wavebands, as well as to find new sub-mm galaxies in `blank fields'. } ",
        "introduction": "The sub-mm waveband has opened up for cosmology, and through this window we can hope to glimpse some answers to a number of related puzzles: What are the brightest sub-mm galaxies? What sorts of galaxies make up the Far-IR Background (FIB)? When did the Universe form the bulk of its stars? How important is dust obscuration for obtaining a full star-formation census?  SCUBA\\cite{scuba} has been instrumental in establishing this new field of sub-mm high redshift astronomy.  This meeting has been dominated by SCUBA-based surveys, and stands as a testament to the instrument builders who produced the best camera of its kind at just the right time.  \\begin{figure}[ht] \\epsfxsize=30pc \\center\\epsfbox{firb.ps} % \\label{fig:firb} \\caption{Estimates for the cosmic background in the IR and neighbouring wavebands.  This is a blown-up and up-dated plot of Figure~26.2 in {\\it Astrophysical Quantities}. Lower limits generally come from the integrated brightness of resolved sources. Limits which are more model-dependent are indicated by dotted lines.  The FIB has been slowly emerging from careful analyses of (largely) {\\sl COBE\\/} data.  It appears to contain more energy than the optical background (usually assumed to be the sum of the deepest HST counts).  It also appears to be broader than the blackbody CMB peak, possibly indicating that it comes from a range of redshifts. References for the points are given in the text.} \\end{figure}  Many basic questions have already been answered by the SCUBA data which are in hand, and a coherent picture has developed -- typical SCUBA sources are the $z\\,{\\sim}\\,$2--3 counterparts of locally well-studied Ultraluminous Infrared Galaxies, and these are considerably more common at an epoch which was perhaps important for the formation of elliptical galaxies. However, there are still a great many details to understand about the exact composition of the SCUBA-bright sources, including how they overlap with populations selected at other wavelengths, how the ${\\sim}\\,$mJy sources might differ from the brighter ones, what the clustering strength of the SCUBA sources might be, and what fraction could be at still higher redshifts. Here we will avoid reviewing all of the different studies (many of which are covered in other presentations), but will concentrate on projects carried out by our own group.  We have been undertaking a set of related studies, focussing on either blank fields to detect sub-mm sources, or on characterizing the sub-mm properties of objects selected at other wavelengths. ",
        "conclusions": "Further clues to the nature of sub-mm sources and how they overlap with optical-, IR- and radio-selected galaxies will be found through the joint approach of investigating the sub-mm properties of already known objects, and finding identifications for possibly unknown objects which are bright in the sub-mm.  Studies of the clustering of SCUBA sources are just in their infancy, but should provide further cosmological information.  The crucial thing is to identify the SCUBA sources at other wavelengths -- this has proved to be difficult in practice, and the sources for which is {\\it most\\/} difficult are potentially the most interesting!"
    },
    "0010/astro-ph0010101_arXiv.txt": {
        "abstract": "We have used the Submillimetre Common-User Bolometer Array (SCUBA) detector on the  James Clerk Maxwell Telescope (JCMT) to measure bright sub-mm emission associated with a recently discovered extensive (${>}\\,100h^{-1}$kpc) and highly luminous, `blob' of Ly$\\,\\alpha$ emission at $z=3.09$. The blob lies within a known large overdensity of optical sources in the $z=3.07$--3.11 range, and is centered on a locally overdense peak within this region. The best explanation for the copious sub-mm emission is a dust obscured continuum source, which may produce the ionizing flux for the Ly$\\,\\alpha$ cloud. Cooling gas explanations are plausible but excessively complicated, and the 450/850$\\,\\mu$m ratio rules out a significant fraction of the signal arising from the Sunyaev-Zel'dovich increment. At least two additional ${\\simeq}\\,10\\,$mJy sub-mm detections in the SCUBA map, with a surface density significantly higher than in blank field surveys, suggests that they may be associated with the $z\\,{=}\\,3.09$ structure. A SCUBA `photometry' observation of a second nearby Ly$\\,\\alpha$ blob tentatively detects a weaker sub-mm counterpart. % ",
        "introduction": "Deep sub-mm surveys with SCUBA have detected distant luminous galaxies which emit most of their energy at far infrared wavelengths (e.g., Smail et al.~1997; Barger et al.~1998, Hughes et al.~1998, Eales et al.~1998). Although some of these sources have been identified with very faint and sometimes very red optical counterparts (Smail et al.~1999) their poorly constrained redshift distribution (Barger et al.~1999, Smail et al.~2000) has made their interpretation unclear. Bright distant SCUBA sources appear similar to local ultra-luminous infra-red galaxies (e.g.~Sanders 2000), but represent a more significant fraction of the global star formation at high redshift.  How these differ at high $z$, and how they are related to optically selected galaxies remains to be understood.  We might expect such luminous sub-mm sources to be especially prevalent in regions that will one day become rich clusters of galaxies.  For instance, if early type galaxies (which are dominant only in rich cluster environments) form by major merger events, then during the epoch of their formation they might produce a large fraction of the bright sub-mm sources in the Universe.  There might therefore be a strong clustering signal from sub-mm sources at high redshift (e.g.~Ivison et al.~2000). Even if the sub-mm sources are primarily AGN-driven, we might expect them to be strongly clustered, given the bulge to black hole mass correlations (cf., Gebhardt et al.~2000).  Since it is not yet possible to measure the clustering properties of sub-mm sources, the next best approach is to examine regions that are already known to be overdense from other techniques.  The most likely time in a cluster's history for a bright sub-mm phase is when smaller groups are collapsing, prior to virialization of the whole proto-cluster environment, when compact groups of objects with relatively low velocity dispersion present high probabilities for merger events. Therefore, galaxy concentrations within known large-scale overdensities are good places to search for significant sub-mm emission.  Extensive spectroscopic surveys have recently shown strong clustering among the star forming Lyman break galaxy (LBG) population at $z\\,{\\sim}\\,3$ (e.g., Adelberger et al.~1998). In one striking example, a high-contrast redshift overdensity of LBGs at $z\\,{=}\\,3.09$ in the SSA22 survey field was discovered (Steidel et al.~1998), with deep, narrow band \\lya\\ imaging used to identify at least 160 likely members of this structure (Steidel et al.~2000). Most dramatic in this $z\\,{=}\\,3.09$ spike was the discovery of two extensive (${>}\\,100h^{-1}$kpc), highly luminous ($L_{{\\rm Ly}\\alpha}\\,{\\sim}\\,10^{37}$W) `blobs' of Ly$\\,\\alpha$ emission. The `blobs' resemble the giant Ly$\\,\\alpha$ nebulae associated with high redshift radio galaxies (cf.\\ van Ojik et al.~1996), but have ${<}\\,$1\\% of the associated radio continuum flux and no obvious source of UV photons bright enough to excite the nebular emission.  Fabian et al.~(1986) suggested that such extended emission line nebulae might be the signature of `maximal cooling flows' in  proto-cluster  and proto-galaxy environments -- these `blobs' resemble just such objects.  The SSA22 $9\\times9$ arcmin field at $z\\,{=}\\,3.07$--3.11 is a factor of 6 overdense in LBGs relative to the general field, and about twice as overdense again within the local density peak centered on `blob~1'. At present, the SSA22 structure is by far the best studied candidate for a clustered region at $z\\,{>}\\,1$. The space density of the LBG redshift `spikes' is similar to that of rich x-ray clusters at low redshift, supporting the plausible identification of the LBG overdensities as proto-cluster sites.  We have obtained SCUBA 850$\\mu$m and 450$\\mu$m maps of the Ly$\\,\\alpha$ blob lying at an overdensity peak within the already overdense field, revealing an extremely luminous sub-mm counterpart. In this letter, we describe the available multi-wavelength data for the blob regions and explore the relevant excitation mechanisms. ",
        "conclusions": "The central `blob~1' SCUBA source is among the most luminous high-$z$ sub-mm sources known. It is especially interesting that this object was discovered through the presence of tremendous amounts of \\lya\\ emission, usually thought of as easily destroyed by the presence of dust.  This begs the question: would other bright sub-mm sources (if their redshifts were known) exhibit similar \\lya\\ `halos'?  One luminous sub-mm {\\it selected} source, SMMJ02399-0136, is known to be associated with an extended ($>$8\\arcsec) \\lya\\ region (Ivison et al.~1998), although the actual extent has not yet been constrained through a narrow-band image. Both star forming and AGN components have been identified, although the primary power source for the sub-mm emission has not been constrained. A few high-$z$ radio galaxies with giant \\lya\\ halos have also been detected in the sub-mm, including 4C41.17 (Chambers et al.~1990) which has a possible `cluster' of sub-mm sources associated (Ivison et al.~2000). However, in the case of this SSA22 blob, the \\lya\\ halo is hard to explain through excitation of the ambient medium by a radio jet, since there is no known radio counterpart here. Archibald et al.~(2000) have observed a sample of high redshift radio galaxies using SCUBA, detecting several with known \\lya\\ halos. The sub-mm emission from the `blob~1' object is more luminous than any radio galaxy in this survey of 47 sources over $1\\,{<}\\,z\\,{<}\\,4$.  Indeed, the lack of significant radio emission from our blob~1 SCUBA object suggests that the \\lya\\ halos surrounding distant radio galaxies may not be entirely related to the radio sources.  The questions of the ionizing source for the Ly$\\,\\alpha$ blobs and the nature of the sub-mm emission remain. The blobs are probably excited by either photoionization from a central source, or as a result of virialization in a deep potential. The sub-mm emission could be either a UV-luminous star forming galaxy or an obscured AGN. More extended sub-mm emission such as from a cooling flow or even from the Sunyaev-Zel'dovich effect are also possible. The lack of a UV source in the optical images, coupled with the large dust mass implied by the sub-mm detections, make direct photoionization of the \\lya\\ from such a central AGN or starburst difficult to imagine. However, it is conceivable that complicated geometrical and radiative transfer effects could conspire to allow substantial \\lya\\ fluxes at the same time that the continuum sources are heavily obscured by dust along our line of sight (e.g.~Neufeld 1991). % The fact that the second \\lya\\ blob is not a strong sub-mm source suggests that an obscured continuum source need not be the primary driver for the \\lya\\ emission.  For a pure star-burst galaxy, the star formation rate extrapolated from the sub-mm would be in excess of $500\\,h^{-2}{\\rm M}_\\odot{\\rm yr}^{-1}$ for `blob~1'.  By analogy with the giant Ly$\\,\\alpha$ halos around high-$z$ radio galaxies, there is reason to suspect a luminous AGN at the heart of the Ly$\\,\\alpha$ blob.  Although the redshift of the \\lya\\ `blob~1' ($z\\,{=}\\,3.09$) is relatively high for radio detection, the farIR/radio correlation for radio quiet sources (Helou et al.~1995) suggests a 1.4\\,GHz flux density of ${\\sim}\\,170\\,\\mu$Jy if it is a steep spectrum object (Carilli \\& Yun 1999), and possibly a comparable 8.5\\,GHz flux if it had the flatter spectrum characteristic of an AGN. Our Bayesian 95\\% upper limits of 142$\\,\\mu$Jy (1.4\\,GHz) and 95$\\,\\mu$Jy (8.5\\,GHz) lie well within the uncertainties of the Carilli \\& Yun (2000) relation, and deeper observations will be required to further constrain the properties of the system. The ratio $S_{450}/S_{850}$ for the central source is $3.8\\pm1.2$, which is consistent with a starburst or radio quiet AGN lying at $z\\,{\\simeq}\\,3$ (cf.\\ Hughes et al.~1998).  Another possibility is that the 850\\mum\\ flux could be partly Sunyaev-Zel'dovich increment,  although the detection of 450\\mum\\ flux argues against this being a significant component. On theoretical grounds, the surface brightness of the increment for rich local clusters would be expected to be around $1\\,$mJy per SCUBA beam, and all other things being equal, S-Z surface brightness is independent of redshift. The S-Z signal is approximately $\\propto M\\times\\rho$: the background density is proportional to $(1+z)^3$, and so the core of a cluster virializing at high redshift would be much denser. However, the characteristic non-linear mass in an $\\Omega_M=1$ universe goes like $(1+z)^{-6/(3+n)}$, where $n$ is the power-spectrum slope.  For CDM models $n$ is -1 to -2 on the relevant scales. Hence, though the estimate is dependent on cosmology, the simple $M\\times\\rho$ argument suggests that the S-Z contribution is not likely to explain all our signal. The expected {\\it decrement\\/} at $8.4\\,$GHz would be ${\\simeq}\\,100\\,\\mu$Jy, and hence at the limit of being detectable currently.  If the blobs are associated with cooling-flow-like phenomena (e.g.~Hu~1992), then some of the sub-mm emission could come from dust associated with distributed star-formation in the cooling flow. Edge et al.~(1999) and Chapman et al.~(2000c) have searched for such emission in well-known rich clusters with null results.  However, local cooling flow clusters are not necessarily a proper benchmark, given that the associated gas densities could conceivably be much higher at $z\\,{\\sim}\\,3$. Fardal et al.~(2000) and Haiman et al.~(2000) have recently modelled the cooling radiation from forming galaxies and concluded that sub-virial temperatures for the gas could lead to much lower x-ray luminosities than one would expect from extrapolating local cooling flows. While it is not inconceivable that some of the flux has this source, even optimistic estimates for the strength of the signal, as discussed in the above papers, come out at a few mJy.  Future observations will continue to elucidate the nature of the `blobs'. For optically thin sub-mm emission, we can estimate the expected x-ray signal of the `blob~1' sub-mm object using different analogs for the central source at $z\\,{=}\\,3.09$, and representative sub-mm/x-ray indices (cf.\\ Fabian et al.~2000). Even the most heavily obscured (Compton-thick) AGN, analogous to NGC\\,6240, should be detectable with Chandra or Newton with soft and hard x-ray fluxes of order ${\\simeq}10^{-18}{\\rm W}\\,{\\rm m}^{-2}$, while an Arp220 analog would be even fainter. Deeper radio maps or higher frequency interferometer maps would be able to distinguish any SZ component, as well as confirm the other sub-mm sources and their positions.  The extreme sub-mm luminosity from the central \\lya\\ halo in `blob~1' (and weaker emission detected from `blob~2') is an intriguing phenomenon, given the fragility of the \\lya\\ line in the presense of dust.  This combination of deep, large-area \\lya\\ surveys and sub-mm images may be interesting to explore further as a way of unearthing information about clusters in their early stages of formation."
    },
    "0010/astro-ph0010384_arXiv.txt": {
        "abstract": "The emergence of a new population of radio galaxies at mJy and sub-mJy levels is responsible for the change in the slope of the radio source counts.  \\\\ This population seems to include both star forming galaxies and classical (AGN-powered) radio sources, but the relative importance of the two classes is still debated. \\\\ We present results from the ATESP radio survey and its optical follow-up and show that the fraction of starburst galaxies changes from $\\sim$ 15$\\%$ at fluxes $\\ge$ 1 mJy to $\\ge$ 50\\% at lower fluxes. ",
        "introduction": "Recently, deep radio surveys ($S \\la 1$ mJy) have shown that normalized radio counts show a flattening below a few mJy, (see figure 1 for counts at 1.4 GHz). This change of slope is generally interpreted as being due to the presence of a new population of radio sources (the so--called sub--mJy population) which does not show up at higher flux densities (see e.g. Condon 1989), where the counts are dominated by classical radio galaxies and quasars.  The sub-mJy population is essentially composed by low luminosity AGN and starburst galaxies, but their relative comtribution is still not firmely established. Unfortunately, due to the long observing times required to reach faint fluxes, the existing samples in the sub--mJy region are generally small.  The identification work and the subsequent spectroscopy are very demanding in terms of telescope time. Typically, no more than $\\sim 50-60\\%$ of  the radio sources in sub--mJy samples have been identified on optical images, even though in the $\\mu$Jy survey in the Hubble Deep Field  and in SSA13 80$\\%$ of the 111 radio sources have been identified (Richards et al. 1999). On the other hand, the typical fraction of spectra available is only $\\sim 20\\%$. The best studied sample is the Marano Field, where 50$\\%$ of the sources have spectra (Gruppioni et al. 1999a), which permitted the determination of spectral type and redshift for 29 objects. To alleviate the identification work, regions with deep photometry already available (possibly multicolor) provide a significant advantage. The region we have selected fulfills these requirements at least partially.  \\begin{figure} \\epsfxsize=10cm \\epsfysize=10cm \\epsfbox{counts.ps} \\caption{Normalized 1.4 GHz differential source counts for different samples. The ATESP source counts are indicated by filled circles. The fit obtained by Windhorst et al. (1990) is also indicated (solid line). } \\end{figure} ",
        "conclusions": ""
    },
    "0010/astro-ph0010498_arXiv.txt": {
        "abstract": "The 2dF Galaxy Redshift Survey has obtained 135,000 redshifts for galaxies in two broad strips. Here we present the first results of a 3-dimensional search for galaxy clusters based on known 2-dimensional compilations. We derive new redshifts and velocity dispersions for clusters, assess the level of contamination in the sample, analyze the accuracy of photometric redshift estimates and study the space distribution of clusters. ",
        "introduction": "It has long been known that the distribution of galaxies is not homogeneous over scales of at least 200 $h^{-1}$ Mpc; the first redshift surveys revealed an intricate pattern of filaments, voids and walls (e.g., Da Costa 1999) and showed that samples of the order of 10$^5$ galaxies were needed to reach the scale of homogeneity and derive cosmologically useful quantities from analysis of the 3-dimensional distribution of galaxies. The 2dF Galaxy Redshift Survey (e.g., Colless 1999) is the first of the new generation of surveys to be able to obtain such samples with reasonable efficiency. Among papers being published by the 2dFGRS Team it is worth mentioning: the $b$ band type-selected luminosity function (Folkes et al. 1999), the $K$-band luminosity function from 2MASS photometry (Cole et al. 2000), the bivariate brightness distribution (Cross et al.  2000), an accurate estimate of the $\\beta$ parameter (Peacock et al. 2000) and Principal Component Analysis of galaxy populations (Madgwick et al., this volume and 2000).  Clusters of galaxies are the largest bound structures in the observable universe and the only ones that can be observed and identified to cosmologically significant redshifts. The mass distribution of clusters yields limits to the cosmological density parameter; for instance, the standard Cold Dark Matter (CDM) model, normalized to the COBE data, yields cluster densities in excess of observations by an order of magnitude and this provided the first hint of an open universe model.  However, most studies of clusters are still based on photographic catalogs, such as those of Abell and collaborators (1958; 1989), or the APM (Dalton et al. 1992) or Edinburgh-Durham (Collins et al. 1995). Once the 2dF survey is complete, it will be possible to determine a catalog of groups of galaxies on scales extending from compact groups to giant clusters, using 3-dimensional selection algorithms. While this is not easily done at this stage, it is possible to consider clusters selected from 2-dimensional catalogs in 3-dimensional space and assess their reality and level of contamination, derive their space density and study the properties of their members. In turn, this will allow us to 'define' a cluster for later 3-dimensional searches, matching the properties of known template objects. ",
        "conclusions": ""
    },
    "0010/astro-ph0010517_arXiv.txt": {
        "abstract": " ",
        "introduction": "Be stars are non-supergiant early-type stars with Balmer emission lines, whose spectral types range from late O- to early A-type. Extensive studies have revealed that a Be star has a two-component extended atmosphere, a polar region and a cool ($\\sim 10^4\\,{\\rm K}$) equatorial disk. The polar region consists of a low-density, fast outflow emitting UV radiation. The terminal velocity of the polar wind is of the order of $10^3\\,{\\rm km}\\,{\\rm s}^{-1}$. The wind structure is well explained by so-called line-driven wind model, in which the radiative acceleration results from the scattering of the stellar radiation in an ensemble of spectral lines (Castor et al.\\ 1975; Abott 1982).  The mass loss rate of this region inferred from the UV lines is about $10^{-11}-10^{-9}\\,M_\\solar\\,{\\rm yr}^{-1}$ (Snow 1981).  On the other hand, the equatorial disk consists of a high-density plasma from which the optical emission lines and the IR excess arise. The radial velocity of the disk is smaller than a few ${\\rm km\\,s}^{-1}$, at least within $\\sim 10$ stellar radii (Hanuschik 1994, 2000; Waters, Marlborough 1994). The small radial velocity and the small pressure gradient force, as in viscous accretion disks, suggest that the equatorial disk is pressure-supported and geometrically thin in the vertical direction and is rotationally-supported in the radial direction. In fact, the presence of near Keplerian disks around Be stars is supported observationally as well as theoretically. For example, the relation between the size of the H$\\alpha$-emitting region of Be disks resolved with the optical interferometer (Quirrenbach et al.\\ 1997) and the separation of the double peaks of the H$\\alpha$ line is in agreement with that expected for near Keplerian disks. In addition, the long-term line-profile variations are well explained by global disk oscillations with a one-armed density perturbation pattern, which can be present only in near Keplerian disks (Okazaki 1991; Papaloizou et al.\\ 1992; Hanuschik et al.\\ 1995). The mass loss rate through the disk is not known. The density in the disk derived from the IR excess and the observed upper limit of the radial velocity impose the upper limit of the mass loss rate of several $\\times 10^{-9}\\,M_\\solar\\,{\\rm yr}^{-1}$ (Waters, Marlborough 1994).  At present, the viscous decretion disk model proposed by Lee et al.\\ (1991) is the only model that naturally yields near Keplerian disks around Be stars. In this model, matter supplied from the equatorial surface of the star gradually drifts outward by the viscous torque and forms the disk. Assuming Keplerian rotation and neglecting the advective term in the equation of motion, Lee et al.\\ (1991) obtained steady and thermally-stable structure of viscous decretion disks around Be stars.  Later, using a 3D Smoothed Particle Hydrodynamics code, Kroll and Hanuschik (1997) studied the evolution of the gas ejected temporarily from a Be star to model the transient disk formation/decay process around $\\mu$~Cen. They found that the gaseous particles gradually expand and form a near Keplerian disk in the viscous timescale.  These works show that the viscous torque is likely the agent that makes a disk near Keplerian. However, since the specific angular momentum in the near Keplerian disk increases with radius, Be disks are unlikely near Keplerian far from the star. A general belief is that the disk which is Keplerian near the star becomes angular momentum conserving far from the star, although no theoretical or observational study has confirmed this transitional feature of Be disks.  The purpose of the present paper is to investigate whether Be disks have this transitional feature, based on the viscous decretion disk scenario. In section~2, we describe the disk model and formulate the eigenvalue problem for obtaining the radial structure of viscous decretion disks. We present the transonic solutions and discuss the topology of the sonic points in section~3. In section~4, we draw our conclusions and discuss several issues which remain unsolved. ",
        "conclusions": "\\label{sec:summary}  We have examined the characteristics of the outflow in viscous decretion disks around Be stars. For simplicity, we have assumed the disk to be isothermal and adopted the Shakura-Sunyaev $\\alpha$-prescription for the viscous stress. As the radiative force, we have adopted the parametric form used by Chen and Marlborough (1994).  Solving the resulting wind equations for a wide range of parameter values, we have found that a transonic solution exists for any combination of parameters. When the radiative force is much smaller than the gravity, it is basically the pressure force that accelerates the flow up to a supersonic speed. The sonic radius $r_{\\rm s}$ is, then, roughly given by $r_{\\rm s} \\sim {2\\over5}GM/c_{\\rm s}^2$. The sonic radius is smaller for higher disk temperature. A larger radiative force also gives a smaller sonic radius. The outflow is highly subsonic in the inner part of the disk. Approximately, the radial velocity increases as $r$ and the surface density decreases as $r^{-2}$ in the inner subsonic part. For smaller $\\alpha$, slopes of these quantities become steeper in the outer subsonic part. Interestingly, the disk is near Keplerian ($V_\\phi \\sim r^{-1/2}$) in the inner subsonic part, while it is angular momentum conserving ($V_\\phi \\sim r^{-1}$) in the outer subsonic part and in the supersonic part.  We have also found that the topology of the sonic point is nodal for $\\alpha$ greater than a critical value, $\\alpha_*$, whereas it is of saddle type for $\\alpha < \\alpha_*$. In our model, $\\alpha_*$ lies between 0.9 and 0.95 irrespective of values of other parameters. From the similarity between the physics of accretion and decretion and the study of the transonic accretion disks, we expect that the transonic decretion disks with $\\alpha < \\alpha_*$ are stable, whereas those with $\\alpha > \\alpha_*$ are unstable.  We have seen that the viscous transonic decretion disks have many characteristics in agreement or consistent with the observed features. Thus, the current model is satisfactory as a first step to obtain a sophisticated model of Be disks. We have to admit, however, that there are several issues which remain unsolved by the current version of the model. They are the observed high mass-loss rate through the disk or short disk-formation timescale, a variety of the observed density distribution, and the mass supply mechanism from the star. We would like to discuss these issues in the rest of this paper.  First, we consider the mass loss rate $|\\dot{M}|$ through the disk. In principle, the mass loss rate is determined by how much torque is exerted by the star at the inner disk boundary. Unfortunately, however, we have no satisfactory model for the star-disk interaction or observational estimate of $|\\dot{M}|$ in quasi-steady disks ($|\\dot{M}|$ in the disk formation stage will be discussed later). Therefore, in the below we give an indirect estimate of $|\\dot{M}|$ the present model for the steady, viscous transonic solution allows.  In isothermal decretion, as in isothermal accretion, $\\dot{M}$ is not a parameter which characterizes the problem. It does not appear in the wind equations~(\\ref{eqn:windeq1}) and (\\ref{eqn:windeq2}), so the absolute value of the surface density $\\Sigma$ is not determined from the model. Using the observed base density (i.e., the equatorial density at the inner disk radius $r=R$) to normalize $\\Sigma$, however, we can estimate $\\dot{M}$ by \\begin{eqnarray} |\\dot{M}| & \\sim & 2 \\times 10^{-11} \\cdot {{\\rho_{00}(R)} \\over {10^{-11} {\\rm g}\\,{\\rm cm}^{-3}}} \\cdot {{V_r(R)} \\over {10^{-3} c_{\\rm s}}} \\nonumber \\\\ && \\times \\left( {T_{\\rm d} \\over T_{\\rm eff}} \\right)^{1/2} M_\\solar\\,{\\rm yr}^{-1}. \\label{eq:Mdot} \\end{eqnarray} The base density, $\\rho_{00}(R)$, for most stars analyzed from Fe~II lines (Hanuschik 1986) and the IR excess (Waters et al.\\ 1987; Dougherty et al.\\ 1994) ranges from $10^{-11}$ to $10^{-12}$ g\\,cm$^{-3}$. Figure~1 shows that the radial velocity at the inner disk radius of the transonic decretion disks, $V_r(R)$, is roughly of the order of $10^{-3} \\alpha c_{\\rm s}$, as far as the radiative force is much smaller than the gravity. As a result, we obtain $|\\dot{M}|$ of the order of $10^{-11} - 10^{-12} \\alpha M_\\solar\\,{\\rm yr}^{-1}$ for steady, viscous transonic disks.  It should be noted that this value does not increase very much even if we use $V_r(R)$ derived from the observed upper limit of the disk outflow velocity, i.e., a few km\\,s$^{-1}$ for $r \\ltsim 10R$. Since $V_r$ approximately linearly increases with $r$ in viscous decretion disks, this upper limit of the disk outflow, $V_r(10R) <$ a few km\\,s$^{-1}$ $\\sim 0.1 c_{\\rm s}$, suggests that $V_r(R) \\ltsim 10^{-2} c_{\\rm s}$ and $r_{\\rm s} \\gtsim 10^2 R$. Consequently, using the upper limit of the observed radial flow, we have $|\\dot{M}| \\sim 10^{-10} - 10^{-11} M_\\solar\\,{\\rm yr}^{-1}$ from equation~(\\ref{eq:Mdot}). Our results, together with the observed range of the base density of Be disks, suggest that, in steady viscous decretion, the mass loss rate through the disk is at most comparable with that in the polar wind region.  It is interesting to note that another constraint on $\\dot{M}$ is obtained by estimating the spin-down timescale of the Be star. Porter~(1998) derived the constraint on a quantity equivalent to $\\dot{M}$, for which Be stars do not show significant spin-down during their main-sequence lifetimes. Porter adopted that each Be star loses its angular momentum at the rate of $|\\dot{M}| \\ell_{\\rm K}(R)$, where $\\ell_{\\rm K}(R)$ is the Keplerian angular momentum at $r=R$ (here, for the sake of convenience, we have adopted the notation different from Porter's). In viscous disks, however, this rate is largely an underestimate, because it is likely that the Keplerian region extends at least up to $100R$ where the specific angular momentum is one order of magnitude larger than $\\ell_{\\rm K}(R)$. Thus, we have to use $\\sim 10|\\dot{M}| \\ell_{\\rm K}(R)$ instead of $|\\dot{M}| \\ell_{\\rm K}(R)$ to evaluate the angular-momentum loss rate from the star. Then, the constraint for the star not to show significant spin-down during their lifetimes becomes more stringent than that derived by Porter (1998).  Since it is beyond the scope of the paper to present a detailed calculation, we roughly estimate the spin-down timescale below. For this purpose, we adopt the same stellar model as that adopted by Porter, i.e., a rigidly rotating polytrope with the index of 3/2. The stellar angular momentum is, then, given by $\\sim 0.2 f M \\ell_{\\rm K}(R)$, where $f$ is the ratio of the stellar rotation velocity to the break-up velocity, which ranges from 0.4 for early Be stars to 0.8 for late Be stars. Then, the spin-down timescale is given by $\\sim 0.2 f M \\ell_{\\rm K}(R)/10|\\dot{M}| \\ell_{\\rm K}(R)$ $\\sim 0.02 f M/|\\dot{M}|$ $\\sim 10^9 (|\\dot{M}|/10^{-10}M_\\solar\\,{\\rm yr}^{-1})\\,{\\rm yr}$. Since the spin-down timescale has to be much longer than the main-sequence lifetime, which is $\\sim 10^7\\,{\\rm yr}$ for early Be stars and $\\sim 10^8\\,{\\rm yr}$ for late Be stars, we have the constraint on $\\dot{M}$ for Be stars: $|\\dot{M}| \\ll 10^{-8}M_\\solar\\,{\\rm yr}^{-1}$ for early Be stars and $|\\dot{M}| \\ll 10^{-9}M_\\solar\\,{\\rm yr}^{-1}$ for late Be stars. Since the observed disk structure of Be stars is insensitive to the spectral type, the latter constraint ($|\\dot{M}| \\ll 10^{-9}M_\\solar\\,{\\rm yr}^{-1}$) should be taken.  Observationally, the rate of equatorial mass loss has been derived for several stars only in the disk formation stage. Since 1977, $\\mu$~Cen has exhibited only flickering emission caused by episodic mass loss events. Each outburst lasts only for 2--5\\,d and makes a transient, small disk which decays in 20--80\\,d. Hanuschik et al.\\ (1993) estimated the mass loss rate for $\\mu$~Cen during each outburst to be $4 \\times 10^{-9} M_\\solar\\,{\\rm yr}^{-1}$. Recently, Telting et al.\\ (1998) studied the long-term behavior of the Be disk in X Per (4U~0352$+$30), a Be/X-ray binary system, and derived the mass loss rate of $4 \\times 10^{-9} M_\\solar\\,{\\rm yr}^{-1}$ during a disk build-up phase which lasted about 1\\,y in 1994--1995.  These values show that Be stars can eject mass at a rate much higher than the upper limit predicted by the viscous decretion model for steady disks. This suggests that the disk formation is a violent and, consequently, non-steady process to which the the steady disk model cannot be applied. To study such a process, we need numerical simulations such as those by Kroll and Hanuschik (1997).  The short timescale of disk formation around these stars also suggests that the disk-formation process is essentially non-steady. For steady, viscous transonic disks shown in figure~1, the drift time defined by $\\int_R^r V_r^{-1} dr$ is $\\sim 30 V_r(R)/10^{-3}c_{\\rm s}\\,{\\rm yr}$ at $r=10R$ and $\\sim 80 V_r(R)/10^{-3}c_{\\rm s}\\,{\\rm yr}$ at the sonic radius. This means, even if $V_r(R)$ is as fast as $10^{-2}c_{\\rm s}$, the formation of an $r \\sim 10R$ disk takes about 3\\,yr and that of a transonic disk takes about 8\\,yr. This long drift time suggests that only disks which are persistent for more than 10\\,yr will have the structure similar to those studied in this paper.  Next, we discuss the density distribution in Be disks. In isothermal decretion disks, the radial exponent of density, $n$, defined by $n=-d \\ln \\rho_{00}/d \\ln r = -d \\ln (\\Sigma/H)/d \\ln r$, is a little bit bigger than 3.5. More generally, as pointed out by Porter (1999), $2n+3m \\gtsim 7$ must hold in order for the decretion to occur, where $m$ is the radial exponent of the disk temperature. Since Be disks are roughly isothermal except for the very vicinity of the star (Millar, Marlborough 1998, 1999), we would expect that the observed density gradient index $n$ is distributed around $n \\sim 3.5$ with a small dispersion.  On the other hand, the observed values of $n$ are distributed around $n\\sim3-3.5$, which is in agreement with the current model, with a dispersion much wider than that predicted. Adopting a simplified disk model, in which the disk has a constant opening angle and the density depends only on radius, and applying the curve of growth method by Waters (1986) to the observed far-IR excess for 59 Be stars, Waters et al.\\ (1987) found $n = 2-4$. Adopting the same disk model as that of Waters et al.\\ (1987), Dougherty et al.\\ (1994) obtained $n=2-5$ from the near-IR excess colors for 144 Be stars. Moreover, applying the curve of growth method with a more realistic disk model to a Be star, Porter (1999) obtained a value of $n$ quite similar to that obtained by Waters et al.\\ (1987).  Consequently, the viscous decretion disk model is roughly in agreements with the observed density distribution, but it has to explain why the observed density distribution ranges widely. In particular, the model has to explain why there are Be stars with $n \\ltsim 3$, for which the decretion does not seem to occur. Porter (1999) pointed out that the viscosity dependent on the radius and/or the stellar radiation field may help make a decretion disk with a shallow density gradient. More detailed study of these possibilities is desired.  Finally, we come to the most important and long-standing issue to be solved, that is, the mechanism which enables a star to supply mass into the surrounding disk. Be stars rotate significantly slowly compared with the break-up velocities. Therefore, in order to supply mass, stars have to accelerate matter up to break-up velocities. Moreover, since most Be stars keep their disks for long periods, either the mass supply must be continuous or the interval of mass supply events must be much shorter than the viscous timescale. Otherwise, the disk would decay before the next mass supply event occurs, like the disk of $\\mu$~Cen.  Although there is no widely-accepted mechanism for such mass supply, non-radial pulsations of stars may work. Osaki (1986) suggested that the dissipation of the non-radial pulsations gives matter the necessary angular momentum. Recently, Rivinius et al.\\ (1997) found that in $\\mu$~Cen the time interval of mass supply events is equal to a beat period of several strongest non-radial pulsation (NRP) modes, indicating that the mass supply from the star results from the resonance between strongest NRP modes. If this mechanism works in other Be stars, the frequencies of the strongest NRP modes in most Be stars will be distributed so that the beat period becomes much shorter than the viscous timescale. It is highly desirable to study this possibility further.   \\vspace{1pc}\\par  The author is grateful for H.F.\\ Henrichs, I.\\ Negueruela, and J.M.\\ Porter for stimulating discussions. He thanks the anonymous referee for useful comments. He also thanks the Astronomical Institute \\lq Anton Pannekoek', University of Amsterdam, for the warm hospitality, where during his stay most of this work was performed."
    },
    "0010/hep-ph0010214_arXiv.txt": {
        "abstract": "s#1#2#3{{ \\centering{\\begin{minipage}{4.5in}\\baselineskip=10pt\\footnotesize \\parindent=0pt #1\\par \\parindent=15pt #2\\par \\parindent=15pt #3 \\end{minipage}}\\par}}  \\def\\keywords#1{{ \\centering{\\begin{minipage}{4.5in}\\baselineskip=10pt\\footnotesize {\\footnotesize\\it Keywords}\\/: #1 \\end{minipage}}\\par}}  \\newcommand{\\bibit}{\\nineit} \\newcommand{\\bibbf}{\\ninebf} \\renewenvironment{thebibliography}[1] {\\frenchspacing \\ninerm\\baselineskip=11pt \\begin{list}{\\arabic{enumi}.} {\\usecounter{enumi}\\setlength{\\parsep}{0pt} \\setlength{\\leftmargin 12.7pt}{\\rightmargin 0pt} % \\setlength{\\itemsep}{0pt} \\settowidth {\\labelwidth}{#1.}\\sloppy}}{\\end{list}}  \\newcounter{itemlistc} \\newcounter{romanlistc} \\newcounter{alphlistc} \\newcounter{arabiclistc} \\newenvironment{itemlist} {\\setcounter{itemlistc}{0} \\begin{list}{$\\bullet$} {\\usecounter{itemlistc} \\setlength{\\parsep}{0pt} \\setlength{\\itemsep}{0pt}}}{\\end{list}}  \\newenvironment{romanlist} {\\setcounter{romanlistc}{0} \\begin{list}{$($\\roman{romanlistc}$)$} {\\usecounter{romanlistc} \\setlength{\\parsep}{0pt} \\setlength{\\itemsep}{0pt}}}{\\end{list}}  \\newenvironment{alphlist} {\\setcounter{alphlistc}{0} \\begin{list}{$($\\alph{alphlistc}$)$} {\\usecounter{alphlistc} \\setlength{\\parsep}{0pt} \\setlength{\\itemsep}{0pt}}}{\\end{list}}  \\newenvironment{arabiclist} {\\setcounter{arabiclistc}{0} \\begin{list}{\\arabic{arabiclistc}} {\\usecounter{arabiclistc} \\setlength{\\parsep}{0pt} \\setlength{\\itemsep}{0pt}}}{\\end{list}}  \\newcommand{\\fcaption}[1]{ \\refstepcounter{figure} \\setbox\\@tempboxa = \\hbox{\\footnotesize Fig.~\\thefigure. #1} \\ifdim \\wd\\@tempboxa > 5in {\\begin{center} \\parbox{5in}{\\footnotesize\\smalllineskip Fig.~\\thefigure. #1} \\end{center}} \\else {\\begin{center} {\\footnotesize Fig.~\\thefigure. #1} \\end{center}} \\fi}  \\newcommand{\\tcaption}[1]{ \\refstepcounter{table} \\setbox\\@tempboxa = \\hbox{\\footnotesize Table~\\thetable. #1} \\ifdim \\wd\\@tempboxa > 5in {\\begin{center} \\parbox{5in}{\\footnotesize\\smalllineskip Table~\\thetable. #1} \\end{center}} \\else {\\begin{center} {\\footnotesize Table~\\thetable. #1} \\end{center}} \\fi}  \\def\\@citex[#1]#2{\\if@filesw\\immediate\\write\\@auxout {\\string\\citation{#2}}\\fi \\def\\@citea{}\\@cite{\\@for\\@citeb:=#2\\do {\\@citea\\def\\@citea{,}\\@ifundefined {b@\\@citeb}{{\\bf ?}\\@warning {Citation `\\@citeb' on page \\thepage \\space undefined}} {\\csname b@\\@citeb\\endcsname}}}{#1}}  \\newif\\if@cghi \\def\\cite{\\@cghitrue\\@ifnextchar [{\\@tempswatrue \\@citex}{\\@tempswafalse\\@citex[]}} \\def\\citelow{\\@cghifalse\\@ifnextchar [{\\@tempswatrue \\@citex}{\\@tempswafalse\\@citex[]}} \\def\\@cite#1#2{{$\\null^{#1}$\\if@tempswa\\typeout {IJCGA warning: optional citation argument ignored: `#2'} \\fi}} \\newcommand{\\citeup}{\\cite}  \\def\\pmb#1{\\setbox0=\\hbox{#1} \\kern-.025em\\copy0\\kern-\\wd0 \\kern.05em\\copy0\\kern-\\wd0 \\kern-.025em\\raise.0433em\\box0} \\def\\mbi#1{{\\pmb{\\mbox{\\scriptsize ${#1}$}}}} \\def\\mbr#1{{\\pmb{\\mbox{\\scriptsize{#1}}}}}  \\def\\fnm#1{$^{\\mbox{\\scriptsize #1}}$} \\def\\fnt#1#2{\\footnotetext{\\kern-.3em {$^{\\mbox{\\scriptsize #1}}$}{#2}}}  \\def\\fpage#1{\\begingroup \\voffset=.3in \\thispagestyle{empty}\\begin{table}[b]\\centerline{\\footnotesize #1} \\end{table}\\endgroup}  \\def\\runninghead#1#2{\\pagestyle{myheadings} \\markboth{{\\protect\\footnotesize\\it{\\quad #1}}\\hfill} {\\hfill{\\protect\\footnotesize\\it{#2\\quad}}}} \\headsep=15pt  \\font\\tenrm=cmr10 \\font\\tenit=cmti10 \\font\\tenbf=cmbx10 \\font\\bfit=cmbxti10 at 10pt \\font\\ninerm=cmr9 \\font\\nineit=cmti9 \\font\\ninebf=cmbx9 \\font\\eightrm=cmr8 \\font\\eightit=cmti8 \\font\\eightbf=cmbx8 \\font\\sevenrm=cmr7 \\font\\fiverm=cmr5  \\newtheorem{theorem}{\\indent Theorem}  \\newtheorem{lemma}{Lemma}  \\newtheorem{definition}{Definition} \\newtheorem{corollary}{Corollary}  \\newcommand{\\proof}[1]{{\\tenbf Proof.} #1 $\\Box$.}     \\textwidth=5truein \\textheight=7.8truein  \\def\\qed{\\hbox{${\\vcenter{\\vbox{\t\t\t% \\hrule height 0.4pt\\hbox{\\vrule width 0.4pt height 6pt \\kern5pt\\vrule width 0.4pt}\\hrule height 0.4pt}}}$}}  \\renewcommand{\\thefootnote}{\\fnsymbol{footnote}}\t%  \\begin{document}  \\runninghead{Galaxy Dynamics and the Second Peak} {Cold Dark Matter?}  \\normalsize\\textlineskip \\thispagestyle{empty} \\setcounter{page}{1}  \\copyrightheading{}\t\t\t%  \\vspace*{0.88truein}  \\fpage{1} \\centerline{\\bf DYNAMICS AND THE SECOND PEAK: COLD DARK MATTER?} \\vspace*{0.035truein} \\centerline{\\footnotesize STACY MCGAUGH} \\vspace*{0.015truein} \\centerline{\\footnotesize\\it Department of Astronomy, University of Maryland} \\baselineskip=10pt \\centerline{\\footnotesize\\it College Park, Maryland 20742-2421, USA} \\vspace*{10pt}  \\vspace*{0.21truein} \\abstracts{The amplitude of the second peak in the angular power spectrum of the cosmic microwave background radiation is constrained to be small by recent experiments like Boomerang.  This is surprising in the context of the $\\Lambda$CDM model, which predicted a large second peak.  However, this result is expected if CDM does not exist.  The observed shape of the power spectrum was accurately predicted (before the fact) by a model motivated by the surprising recent successes of the modified dynamics (MOND) hypothesized by Milgrom.}{}{}   \\vspace*{1pt}\\textlineskip\t% ",
        "introduction": " ",
        "conclusions": ""
    },
    "0010/astro-ph0010029_arXiv.txt": {
        "abstract": "We calculate the magnification light curves due to stellar microlensing of gamma-ray burst (GRB) afterglows. A GRB source appears on the sky as a thin ring which expands faster than the speed of light and is maximally magnified as it crosses the lens caustics. While a single star lens produces a single peak in the magnification light curve, binary star lenses may produce multiple peaks.  The shape of the magnification light curve provides invaluable information on the surface brightness distribution of the afterglow photosphere on sub micro-arcsecond scales.  We find that {\\it all} afterglows are likely to show variability at the level of a few percent about a year following the explosion, due to stars which are separated by tens of Einstein radii from their line-of-sight. ",
        "introduction": "The lensing cross-section of a star at cosmological distances is defined by the Einstein angle, \\begin{equation} \\thetaE=\\left({4GM_{\\rm lens}\\over c^2 D}\\right)^{1/2}= 1.6 \\left({M_{\\rm lens}\\over 1 M_\\odot}\\right)^{1/2}\\left({D\\over 10^{28}~{\\rm cm}}\\right)^{-1/2}~{\\rm micro-arcsecond}, \\label{eq:1} \\end{equation} where $M_{\\rm lens}$ is the lens mass, $D\\equiv {D_{\\rm os}D_{\\rm ol}/ D_{\\rm ls}}$ is the ratio of the angular-diameter distances between the observer and the source, the observer and the lens, and the lens and the source (Schneider, Ehlers, \\& Falco 1992, p. 27).  Typical astrophysical sources and lenses move at speeds of $v\\la 10^3~{\\rm km~s^{-1}}$, and hence produce a characteristic event duration of $\\sim D_{\\rm os}\\thetaE/v\\ga 25$ years for solar mass lenses.  Coincidentally, the apparent angular size of gamma-ray burst (GRB) afterglows is also of order a micro-arcsecond, i.e. comparable to the Einstein angle of a solar mass lens at a cosmological distance.  Loeb \\& Perna (1998) noted that the highly relativistic expansion of these sources would shorten considerably the duration of their microlensing events. A GRB afterglow is predicted to appear on the sky as a narrow ring which expands laterally at a superluminal speed, $\\sim \\Gamma c$, where $\\Gamma\\gg1$ is the Lorentz factor of the relativistic blast wave which emits the afterglow radiation (Waxman 1997b; Sari 1998; Panaitescu \\& Meszaros 1998; Granot, Piran, \\& Sari 1999).  For $\\Gamma\\sim 10$ the event duration is shortened by 3--4 orders of magnitude relative to typical astrophysical sources, and is reduced to the convenient timescale of days.  Recently, Garnavich, Loeb, \\& Stanek (2000) have reported the possible detection of a microlensing magnification feature in the optical-infrared light curve of GRB 000301C. The achromatic bump in the light curve is well fitted by a microlensing event of a $0.5 M_\\odot$ lens separated by an Einstein angle from the source center, as long as the source is a ring with a narrow fractional width ($\\sim 10\\%$), in accordance with earlier theoretical predictions. In a case similar to GRB sources, Koopmans \\& de Bruyn (2000) have recently reported evidence for intra-day radio microlensing in the lens B1600+434 due to superluminal motions of the lensed source. On a longer-time scale, microlensing of the lensed QSO 2237+0305 had been observed over many years and more recently in real-time (Wozniak et al. 2000 and references therein).  The predicted angular size of GRB afterglows was already confirmed through the detected transition from diffractive to refractive scintillations in the radio afterglow of GRB 970508 (Goodman 1997; Waxman, Kulkarni, \\& Frail 1998).  Unfortunately, scintillations are not a sufficiently delicate probe to provide information about the ring-like surface brightness distribution of a GRB source on the sky.  On the other hand, microlensing can resolve the sub micro-arcsecond structure of GRB photospheres as a function of observed photon frequency.  In this {\\it Letter}, we expand the original discussion of Loeb \\& Perna (1998) on single star lenses, and calculate the magnification light curve for two other lens configurations, namely binary star lenses and a single star in an external shear field (due to the gravitational potential of the host galaxy or a distant binary companion). These lensing configurations are more likely to describe stars which are embedded in the luminous cores of galaxies. At cosmological distances, the simple model of single star lensing is adequate only for stars which reside in the outer regions of galaxies or in the intergalactic medium.  The probability for having an intervening lens star within a projected angular separation $\\theta$ from a source at a redshift $z\\sim 2$ is~~$\\sim 0.3 \\Omega_\\star (\\theta/\\thetaE)^2$ (Press \\& Gunn 1973; Blaes \\& Webster 1992; Nemiroff 1998; Nemiroff et al. 1998), where $\\Omega_\\star$ is the cosmological density parameter of stars.  The value of $\\Omega_\\star$ is bounded between the density of the luminous stars in galaxies and the total baryonic density as inferred from big bang nucleosynthesis, $7\\times 10^{-3} \\la \\Omega_\\star\\la 5\\times 10^{-2}$ (Fukugita, Hogan, \\& Peebles 1998). Hence, all GRB afterglows should show evidence for events with $\\theta \\la 20\\thetaE$, for which microlensing provides a small perturbation to the light curve\\footnote{This crude estimate ignores the need to subtract those stars which are embedded in the dense central regions of galaxies, where macrolensing dominates and the microlensing optical depth is of the order of unity.}. In this {\\it Letter}, we calculate the magnification light curves of such events, as well as light curves of rarer events with a smaller impact parameter -- for which the peak amplitude is larger.  In \\S 2 we describe the method of our calculation, and in \\S 3 we present our numerical results. Finally, \\S 4 summarizes the main implications of these results. ",
        "conclusions": "We have found that all GRB afterglows are likely to show a microlensing amplification bump of a few percent after about a year (Fig. 1).  The peak magnification and its timing depend strongly on the impact parameter.  A smaller impact parameter results in an earlier and a higher peak magnification. For $b\\ga 2$, the peak magnification is usually reached when the limb of the source crosses the lens. Equation (\\ref{source}) implies that the time of peak magnification is then $t_{\\rm peak}\\approx (b/R_0)^{8/5}$. Numerically, we find that the peak magnification scales roughly as $\\mu_{\\rm peak} \\approx 1+2.5 (W/0.05)^{-1/2} b^{-3/2}$ for $C=0$. The {\\it Swift} satellite \\footnote{see http://swift.sonoma.edu/}, planned for launch in 2003, will provide trigger and good localization for hundreds of GRB afterglows per year.  Identification of the statistics of such features can be used to constrain $\\Omega_\\star$.  A remaining important question is how the optical depth for microlensing (or the associated shear) is distributed along the line of sight.  While stars in the halos of galaxies are likely to be in the regime of very small optical depth, the outer disk or spheroid stars will be surrounded by a modest optical depth with a significant shear, while stars in the cores of galaxies are in the high optical depth regime where the GRBs are likely to be multiply imaged.  The relative abundance of stars in these populations can be, in principle, calibrated based on ongoing microlensing studies of the Milky Way galaxy (Alcock et al.  2000b, and references therein).  The detailed shape of the magnification light curve depends on the contrast, $C$, and fractional width, $W$, of the ring-like afterglow image (Fig. 2).  The values of these parameters should change in a predictable way across spectral breaks (e.g. when comparing radio to optical-infrared data), because of the corresponding change in the structure of the source image on the sky (Granot et al. 2000). Detailed multi-frequency monitoring of a microlensing event, similar or better than the data used by Garnavich et al. (2000), could provide invaluable information on the sub micro-arcsecond structure of GRB photospheres.  We have also found that binary lenses produce multiple peaks in the magnification light curves (Fig. 3). However, an external shear has only a modest effect on the magnification history, changing its early time amplitude from the constant value which characterizes the zero shear case, but not affecting much its late evolution. While the zero shear lens does not resolve the source when the angular size of the source is much smaller than its impact parameter relative to the lens, the star$+$shear lens provides information about the expansion of the source even at these early times. The light curves of both the star$+$shear lens and the binary lens are determined by a small number of parameters.  For example, the binary light curves are fixed by two binary parameters (separation and mass ratio) and five parameters that describe the source properties: the source center position $(\\xc, \\yc)$, $R_0$, $W$ and $C$ [see Eqs. (\\ref{source}) and (\\ref{mu})].  Similarly to the known binary microlensing events toward the Galactic and Magellanic Clouds, it should be possible to infer the lens and source parameters such as $W$, $C$ and $R_0$ from a densely-sampled light curve. The values of these  parameters at various photon frequencies would in turn provide new constraints on the energy output and ambient gas density of GRB fireballs, and hence on the central engines that power them."
    },
    "0010/astro-ph0010503_arXiv.txt": {
        "abstract": "A follow-up X-ray study was made of the west lobe of the radio galaxy Fornax~A, (NGC~1316) based on new {\\it ASCA} observations made in 1997 for 98 ks, and incorporating the previous observation in 1994 for 39 ks.  The 0.7--10 keV spectrum of the emission can be described by a power-law of energy index $0.74\\pm0.10$, which agrees with the synchrotron radio index of $0.9\\pm0.2$.  Therefore, the X-rays are reconfirmed to arise via inverse-Compton scattering of the cosmic microwave photons, as Kaneda et al. (1995) and Feigelson et al. (1995) concluded.  The surface brightness of the inverse-Compton X-rays exhibits a relatively flat distribution over the west lobe, indicative of an approximately spherical emissivity distribution with a radius of $\\sim11'$ (75 kpc).  In contrast, the 1.4 GHz radio image by Ekers et al. (1983) exhibits a rim-brightened surface brightness, consistent with a shell-like emissivity distribution whose inner and outer boundaries are $4'$ and $11'$, respectively.  These morphological differences between radio and X-rays suggest that the relativistic electrons are distributed homogeneously over the lobe volume, whereas the magnetic field is amplified toward the lobe rim region. ",
        "introduction": "The jet terminal lobes of radio galaxies are gigantic inter-galactic structures consisting of magnetic fields and relativistic particles, both of which are supposed to be supplied by the active galactic nuclei (AGNs) through the jets.  Nevertheless, we do not yet know the relative importance of the magnetic fields and the relativistic particles in the mechanism of jet formation.  To answer the issue, we need to measure spatial distributions of the field and particle energy densities along various locations of the AGN-jet-lobe system.  The relativistic electrons in the radio lobes interact with the magnetic fields and soft photons to produce synchrotron radiation and inverse-Compton (IC) emission, respectively.  Feigelson et al. (1995, hereafter FEA95) and Kaneda et al. (1995, KEA95) discovered the IC X-ray emission with {\\it ROSAT} (Tr\\\"{u}mper 1983) and {\\it ASCA} (Tanaka et al. 1994), respectively, from the lobes of the radio galaxy Fornax~A (NGC~1316, redshift $z = 0.00587$; Longhetti et al. 1998), followed by authors reporting the lobe IC X-rays from Centaurus~B (PKS~1343$-$601; Tashiro et al. 1998), 3C~219 (Brunetti et al. 1999) and NGC~612 (Tashiro et al. 2000).  These authors estimated the energy density of soft photons, and succeeded in determining the magnetic field intensity and the energy densities of electrons in the lobes. Among these pioneering results, Tashiro et al. (1998) and Brunetti et al.  (1999), respectively, showed that the lobes of Centaurus~B and 3C~219 exhibit clear dominance of particle energy, in their inner regions. Notably, Tashiro et al. (1998) found relative enhancements of the magnetic fields towards the periphery in the Centaurus~B lobes.  Our next task is to resolve spatial distributions of the IC X-ray emission to trace the dynamics of the jet-lobe system.  The lobes of Centaurus~B, however, are not in fact the best target for this investigation with {\\it ASCA}, because of the bright nucleus and the relatively small angular separation of its lobes.  In this paper, we present results from {\\it ASCA} follow up observations of the west lobe of Fornax~A.  This prototypical IC source is ideal for our purpose, because of its nearly dormant nucleus (Iyomoto et al. 1998), the large ($\\sim 10'$ in radius) angular size of its lobes, and the high integrated IC flux. ",
        "conclusions": "We measured the diffuse X-rays of the Fornax~A west lobe with an improved accuracy employing the {\\it ASCA} follow up observations. The diffuse emission spatially coincides with the radio synchrotron lobe on a large scale.  The derived 0.7--10 keV spectrum of the lobe X-rays is well described with a power-law model, whose energy index of $0.74\\pm0.26$ agrees with the synchrotron radio emission energy index, $0.9\\pm0.2$, derived from published radio fluxes observed at 408 MHz, 1.4 GHz, and 2.7 GHz (Cameron 1971; Ekers et al. 1983; Shimmins 1971, respectively).  Although the obtained spectrum formally accepts a thin thermal plasma model, the high temperature ($kT = 8.4^{+\\infty}_{-3.2}$ keV) will allow neither gravitational confinement nor cooling flow to explain the emission.  The derived emission measure requires a thermal pressure of $\\sim 3\\times10^{-13}$ J~m$^{-3}$.  If we assume a magnetic field of $\\sim 1$ nT to confine the possible thermal plasma, that field will reduce the synchrotron cooling time down to $\\sim 10^6$ yr (Iyomoto et al. 1998).  This is too short for the relativistic electrons to diffuse in the whole lobes, considering the estimated growth speeds of lobes (Scheuer 1995).  Therefore, we reconfirm, with a higher accuracy, that the diffuse X-ray emission is produced via the IC process in which the cosmic microwave background (CMB) photons are boosted by the synchrotron electrons, as FEA95 and KEA95 concluded.  With the follow up observations pointing on the west lobe, we examined the brightness distribution of the IC X-ray emission, and revealed that it indicates a center-filled emissivity distribution in contrast to the rim-brightening radio shape.  This indicates that the relativistic electrons homogeneously in the sphere of radius $\\sim 75 h_{75}^{-1}$ kpc $(= 11')$, where $h_{75}$ is the Hubble constant normalized to $75 \\:\\:{\\rm km~s}^{-1} {\\rm Mpc}^{-1}$.  We therefore regard the rim brightened radio emission as a direct indicator of the magnetic pressure distribution, represented with a shell of inner and outer boundaries of $\\sim 27 h_{75}^{-1}$kpc $(=4')$ and $\\sim 75 h_{75}^{-1}$kpc $(=11')$, respectively.  On the basis of the spectral and spatial analysis, we evaluate the energy densities of magnetic field ($u_{\\rm B}$) and electrons ($u_{\\rm e}$) according to Harris and Grindlay (1979).  Since $u_{\\rm B}$ is determined by the ratio of the synchrotron radio flux to the IC X-ray flux emitted from the same volume, $u_{\\rm B}$ is in inverse proportion to the volume ratio of the shell to the electron filled sphere, which is $\\sim 0.95$.  We estimate the 1.4 GHz flux falling inside the X-ray integration region as 50 Jy ($=5\\times10^{-25}$W~s$^{-1}$m$^{-2}$), based on the radio map by Ekers et al. (1983).  We then calculate $u_{\\rm B}$ after Harris and Grindlay (1979), considering the filling factor ($f$) to the shell region, and assuming the magnetic fields to have random directions to the line of sight. The result is $u_{\\rm B} = (4.6 \\pm 0.4) f^{-1} \\times10^{-14}$ J~m$^{-3}$.  The derived $u_{\\rm B}$ corresponds to the magnetic field strength of $(0.34\\pm0.02)$ nT at $f = 1$.  Note here we adopted an IC scattering electron density, at Lorentz factor $\\gamma \\sim 1000$, estimated through extrapolation from the observed synchrotron spectrum at $\\gamma \\sim 5000$ -- 13000.  Utilizing equation (10) of Harris and Grindlay (1979), we also derive $u_{\\rm e}$ in the sphere.  We assume that the electron Lorentz factor ranges $\\gamma = 10^{3-5}$ in calculation below, as KEA95 adopted.  Thus we obtained $u_{\\rm e}$ from the estimated IC X-ray flux (Table 1) as $u_{\\rm e} = (2.0 \\pm 0.5) (r_{\\rm out}/11')^{-3} h_{75} \\times 10^{-14}$ J~m$^{-3}$. Although the derived $u_{\\rm B}$ is nominally larger than the $u_{\\rm e}$ by a factor of 2.3 in the shell, we cannot immediately reject electon-magnetic field equipartition ($u_{\\rm B} = u_{\\rm e}$) there, without determining the electron energy spectrum below $\\gamma \\sim 5000$.  If the spectrum turns down just below $\\gamma = 5000$, the estimated $u_{\\rm B} / u_{\\rm e} \\gtrsim 8.3$. On the contrary, assuming that the spectrum extends down to $\\gamma \\sim 350$ ($\\sim 2$ MHz in synchrotron emission), the integrated $u_{\\rm e}$ becomes equivalent to $u_{\\rm B}$ in the shell region.  The results obtained here suggest a picture that the lobe interior is dominated by the particle pressure, whereas the magnetic field pressure becomes significant in the shell region defined by the lobe boundary.  Interestingly, we see a similar situation in the lobes of Centaurus~B (PKS~1343$-$601), as Tashiro et al. (1998) showed. The similarity is thought to reflect energetics and evolutions of radio lobes (e.g. Blundell \\& Rawlings 2000) and the inter-galactic environments of particle and magnetic fields.  Further investigation with recent advanced X-ray observatories are expected.  \\clearpage  \\begin{table*} \\begin{center} \\tablenum{1} \\caption{Results of the model fit to the 0.7 -- 10 keV GIS spectrum from the lobe region \\tablenotemark{a}}  \\begin{tabular}{lcccc} \\hline \\hline % component & $N_{\\rm H}$\\tablenotemark{b} & energy index &  $kT$ [keV] &$F_{\\rm 1keV}$\\tablenotemark{c}\\\\ \\hline % source 1& (8.0)\\tablenotemark{d} & (3.2)\\tablenotemark{d}&  --- & (0.008)\\tablenotemark{d}\\\\ IC emission & (2.06)\\tablenotemark{e} & $0.74\\pm0.26$ &  --- & $0.10\\pm0.01$ \\\\ ambient thermal &(2.06)\\tablenotemark{e}& --- & $0.85\\pm0.15$& $0.11\\pm0.05$\\\\ \\hline % \\end{tabular} \\tablenotetext{a}{The fit chi-square is 94.7 for 103 degree of freedom. All the errors refer to single-parameter $90 \\%$ confidence limits.} \\tablenotetext{b}{Photoelectric absorption column density in the unit of $10^{24}$H atoms~m$^{-2}$.} \\tablenotetext{c}{Flux density at 1 keV in the unit of 1 $\\mu$Jy = $1\\times 10^{-32}$W~m$^{-2}$Hz$^{-1}$.} \\tablenotetext{d}{Fixed at the best-fit values of the {\\it ROSAT}-PSPC data.} \\tablenotetext{e}{Fixed at the Galactic line-of-sight value.} \\end{center} \\end{table*}  \\clearpage"
    },
    "0010/astro-ph0010445_arXiv.txt": {
        "abstract": "\\noindent This is the second of two papers presenting basic observational data on the 6CE sample of extragalactic radio sources. It presents the results of optical spectroscopy which has yielded virtually complete redshift information for the 6CE sample: 56 of the 59 sample members have spectroscopic redshifts which, with the exception of seven cases, are secure. The redshift distribution $N(z)$ is fairly flat over the redshift range $0 \\leq z < 2$, with a median redshift of $\\approx 1.1$ and a high-redshift tail reaching to $z = 3.4$. The highest-redshift ($z > 1.75$) members of the 6CE sample have similar optical spectra and a tight (less than one dex) spread in narrow-Ly$\\alpha$ emission line luminosity. ",
        "introduction": "\\label{sec:intro}  Obtaining spectroscopic redshifts for complete samples of radio sources is a key requirement for a large number of investigations of the radio source population and its cosmic evolution. Such investigations are severely compromised if they are based on a sample selected at a single limiting flux density, for example the 3CRR sample (Laing, Riley \\& Longair 1983), because of the limited coverage they provide of the 151 MHz luminosity $L_{151}$ versus redshift $z$ plane (e.g.\\ Fig.~\\ref{fig:pz}). Studies based only on the 3CRR sample cannot distinguish between correlations with redshift $z$ and correlations with $L_{151}$, and they cannot straightforwardly measure the relative numbers of radio sources at a given redshift as a function of $L_{151}$, or conversely the relative numbers at a given $L_{151}$ as a function of $z$.  \\begin{figure*} \\begin{center} \\setlength{\\unitlength}{1mm} \\begin{picture}(150,120) \\put(-50,-40){\\special {psfile=\"fig_pz.ps\" vscale=120 hscale=120 angle=0}} \\end{picture} \\end{center} {\\caption[junk]{\\label{fig:pz} The coverage of the 151 MHz luminosity $L_{151}$ versus redshift $z$ plane for the 3CRR sample (triangles) and the 6CE sample (circles); the shaded region shows the rough location of the 131 members of the 7C Redshift Survey (7CRS; Lacy et al.\\ 1999; Blundell et al.\\ in prep.; Willott et al.\\ in prep). }} \\end{figure*}  To ameliorate this problem we have been seeking complete redshift information for low frequency (151 MHz) selected samples drawn from the 6C and 7C radio surveys (e.g.\\ Rawlings et al.\\ 1998). These samples greatly increase the coverage of the $L_{151}$-$z$ plane as illustrated in Fig.~\\ref{fig:pz}. In this paper we report on optical spectroscopy of a 6C-based sample first defined by Eales (1985a) and now, as described in this paper, slightly revised. This is the second of two papers aimed at presenting basic observational data on this revised `6CE' sample. The first paper (Eales et al.\\ 1997) presented near-IR images of many members of the sample.  A numbers of papers have made use of the 6CE dataset. Investigations of the inter-correlations between radio properties (e.g.\\ luminosity, projected linear size, spectral index) and redshift can be found in Neeser et al.\\ (1995), Best et al.\\ (1999) and Blundell, Rawlings \\& Willott (1999). Investigations of the near-IR properties of the combined 6CE/3CRR dataset can be found in Eales et al.\\ (1997) and Roche, Eales \\& Rawlings (1998). An investigation of the fraction of quasars in low frequency selected complete samples can be found in Willott et al.\\ (2000b). Measurements of the radio luminosity function using the combined 3CRR/6CE/7CRS dataset are presented in Willott et al.\\ (2000c).  Throughout this paper we adopt the following values for the cosmological parameters: $H_{\\circ}=50~ {\\rm km~s^{-1}Mpc^{-1}}$, $\\Omega_ {\\rm M}=1$ and $\\Omega_ {\\Lambda}=0$. B1950.0 coordinates are used and the convention for radio spectral index $\\alpha$ is that $S_{\\nu} \\propto \\nu^{-\\alpha}$, where $S_{\\nu}$ is the flux density at frequency $\\nu$. ",
        "conclusions": "A summary of key information on the 6CE sample; the reason for the lack of data on 6C1036+3616 is explained in Sec.~\\ref{sec:sample}. {\\bf Column 1:} `A' indicates that the radio source is also in the B2-selected `1-Jy' sample of Allington-Smith (1982). {\\bf Column 2:} name of the 6CE radio source. {\\bf Column 3:} 151-MHz flux density from the final versions of the 6C survey (Hales et al.\\ 1998; Hales et al.\\ 1993), and {\\bf column 4:} radio spectral index evaluated at rest-frame 151 MHz using a polynomial fit to the multi-frequency data. {\\bf Column 5:} classification, Q=quasar and BLRG=broad-line radio galaxy, following the prescription of Willott et al.\\ (1998); and HEG=high-excitation galaxy, LEG=low-excitation galaxy, following the prescription of Jackson \\& Rawlings (1997) when [OIII] is accessible, and using the detection of any line from an at least doubly-ionized ion as a diagnostic in higher-redshift cases; `?' means that the classification is uncertain -- object-by-object discussions are presented in Secs.~\\ref{sec:obj_by_obj} and ~\\ref{sec:obj_by_obj_lit}. {\\bf Column 6:} $K-$band magnitude of the radio source ID -- a reference to the source of this magnitude and the diameter (in arcsec) of the photometric aperture, is given in the brackets; `?' indicates that this $K-$band detection is significant at only the $2.6 \\sigma$ level; `-' signifies that no $K-$band photometry is available; references are Ea=Eales et al.\\ (1993a), Eb=Eales et al.\\ (1993b), Eea = Eales et al.\\ (1997), ER = Eales \\& Rawlings (1996), LLA=Lilly, Longair \\& Allington-Smith (1985), L89=Lilly (1989). {\\bf Column 7:} redshift, `?' means that this value is not yet an unequivocal redshift; `-' indicates that spectroscopy has yet to reveal a redshift; `*' indicates that the galaxy with this redshift is not yet certain to be the correct radio source ID. {\\bf Column 8:} Prominent emission line in the existing spectra, a `?' meaning that the line identification is uncertain. {\\bf Column 9:} log$_{10}$ of the line luminosity in units of W, a `-' means the data are inadequate to obtain a line luminosity; brackets denote that the luminosity is estimated from an observed equivalent width (EW) and an $r-$ or $R-$band magnitude (from either LLA or Eales 1985c); to estimate [OII] limits from the spectra of the low-redshift LEGs studied by Sargent (1973) or Warner et al.\\ (1983), we took $EW < 25 ~ \\rm \\AA$ and $EW < 10 ~ \\rm \\AA$ respectively, a detected [OII] line in a Sargent spectrum is assigned an $EW = 25 ~ \\rm \\AA$. {\\bf Column 10:} reference to spectroscopy/spectrophotometry: Apc, J. Allington-Smith {\\it priv. comm.}; ALL, Allington-Smith et al.\\ (1985); ASDL, Allington-Smith et al. (1988); DSpc, M. Dickinson \\& H. Spinrad {\\it priv. comm.}; EH, Eracleous \\& Halpern (1994); EHRpc, S. Eales, C. Haniff \\& S. Rawlings {\\it priv. comm.}; ERWpc, S. Eales, S. Rawlings \\& S. Warren {\\it priv. comm.}; La, Lahulla et al. (1991); L, Lilly (1988); LMRSpc, M. Lacy, S. Maddox, S. Rawlings \\& S. Serjeant {\\it priv. comm.}; REL, this paper; REW, Rawlings, Eales \\& Warren (1990); S, Sargent (1973); VT, Vermeulen \\& Taylor (1995); Wea, Warner et al. (1983); WW, Wills \\& Wills (1979). A $\\dag$ indicates that the line flux has been measured from a new spectrum presented in this paper. A $\\ddag$ means that the previously reported redshift is incorrect (see object-by-object notes in Secs.~\\ref{sec:obj_by_obj} and~\\ref{sec:obj_by_obj_lit}). }} \\end{center} \\end{table*}  \\normalsize  \\addtocounter{table}{-1}  \\clearpage   \\scriptsize  \\begin{table*} \\begin{center} \\begin{tabular}{lllrlllrlr} \\hline\\hline (1) & (2) & (3) & (4) & (5) & (6) & (7) & (8) & (9) & (10) \\\\ \\hline & name & \\mc{1}{c|}{$S_{151}$} & \\mc{1}{c|}{$\\alpha_{151}$} & \\mc{1}{c|}{Cl} & \\mc{1}{c|}{$K$} & \\mc{1}{c|}{$z$} & \\mc{1}{c|}{Line} & \\mc{1}{c|}{$\\log_{10} L_{\\rm line}$} & \\mc{1}{r|}{$z$ ref.} \\\\  \\hline  & 6C1045+3553& 2.07 & 0.76 & LEG?& $\\phantom{>}16.33$ ~ (Eea  5)& 0.851?     & [OII]?     &(  35.4  )& LMRSpc     \\\\ & 6C1045+3513& 3.03 & 0.20 & Q   & $\\phantom{>}16.77$ ~ (Eea  5)& 1.594      &           -&          & ERWpc$\\ddag$ \\\\ A& 6C1100+3505& 2.26 & 0.58 & HEG & $\\phantom{>}18.33$ ~ (Eea  5)& 1.440      & [NeIV]     &   35.73  & Apc        \\\\ A& 6C1108+3956& 2.10 & 0.74 & LEG & $\\phantom{>}16.27$ ~ (LLA 12)& 0.590      & [OII]      &(  34.9  )& ASDL       \\\\ A& 6C1113+3458& 2.33 & 0.52 & HEG?& $\\phantom{>}18.29$ ~ (L    8)& 2.406?     & Ly$\\alpha$?&   35.97  & REL        \\\\ & 6C1123+3401& 3.40 &-0.27 & LEG?& $\\phantom{>}17.92$ ~ (Eea  5)& 1.247?     &  CII]?     &   34.59  & REL        \\\\ A& 6C1125+3745& 2.07 & 0.79 & Q   & $\\phantom{>}17.46$ ~ (LLA 12)& 1.233      & [OII]      & $<35.15$ & REL        \\\\ A& 6C1129+3710& 2.36 & 0.51 & HEG?& $\\phantom{>}17.52$ ~ (Eea  5)& 1.060      & [OII]      &          & REL        \\\\ A& 6C1130+3456& 3.20 & 0.57 & LEG & $\\phantom{>}15.76$ ~ (LLA 12)& 0.512      & [OII]      &(  35.1)  & ALL        \\\\ & 6C1134+3656& 2.07 & 0.73 & HEG & $\\phantom{>}19.40$ ~ (ER   5)& 2.125      & Ly$\\alpha$ &   36.62  & REW$\\dag$  \\\\  \\hline  A& 6C1141+3525& 2.40 & 0.82 & HEG & $\\phantom{>}18.72$ ~ (Eea  5)& 1.781      & Ly$\\alpha$ &   37.00  & ASDL     \\\\ A& 6C1143+3703& 2.06 & 0.24 & LEG?& $\\phantom{>}19.83$?~ (Eea  5)&      -     &           -&         -& REL      \\\\ A& 6C1148+3638& 3.21 & 0.69 & LEG?& $\\phantom{>}14.63$ ~ (LLA 12)& 0.141      & [OII]      &($>33.8$ )& S        \\\\ A& 6C1148+3842& 3.83 & 0.64 & Q   &        -         & 1.303      &           -&          & WW       \\\\ & 6C1158+3433& 2.12 & 0.50 & LEG &        -         & 0.530?     & [OII]?     &   34.60  & REL      \\\\ A& 6C1159+3651& 2.20 & 0.50 & LEG?& $\\phantom{>}17.42$ ~ (L89  8)&      -     &           -&         -& REL      \\\\ A& 6C1204+3708& 3.92 & 0.61 & HEG & $\\phantom{>}19.29$ ~ (Eea  5)& 1.779      & Ly$\\alpha$ &   37.00  & REL      \\\\ A& 6C1204+3519& 3.43 & 0.54 & HEG & $\\phantom{>}17.97$ ~ (Eea  5)& 1.376      & CIV        &   35.73  & REL      \\\\ & 6C1205+3912& 3.83 & 0.76 & LEG?&        -         & 0.243      & [OII]      &(  34.9  )& S        \\\\ A& 6C1212+3805& 2.14 & 0.49 & LEG?& $\\phantom{>}17.74$ ~ (Eea  5)& 0.947?     & [OII]?     &   35.20  & REL      \\\\  \\hline  & 6C1213+3504& 2.39 & 0.02 & Q   &        -         & 0.857      & [OII]      &(  35.4  )& VT       \\\\ A& 6C1217+3645& 2.40 & 0.74 & HEG?& $\\phantom{>}17.27$ ~ (Eea  5)& 1.089?     & [OII]?     &   34.81  & REL      \\\\ A& 6C1220+3723& 2.52 & 0.59 & Q   & $\\phantom{>}15.30$ ~ (LLA 12)& 0.489      & [OIII]     &(  36.2  )& ALL      \\\\ A& 6C1230+3459& 2.90 & 0.42 & HEG & $\\phantom{>}18.40$ ~ (Eea  5)& 1.533      & [NeIV]     &   35.59  & ASDL     \\\\ & 6C1232+3942& 3.27 & 0.93 & HEG & $\\phantom{>}17.80$ ~ (Ea  11)& 3.221      & Ly$\\alpha$ &   36.73  & REW      \\\\ A& 6C1255+3700& 3.66 & 0.49 & Q   & $\\phantom{>}15.59$ ~ (LLA 12)& 0.710      & [OIII]     &(  36.1  )& EH$\\ddag$\\\\ A& 6C1256+3648& 2.88 & 0.60 & HEG?& $\\phantom{>}17.75$ ~ (Eea  5)& 1.127      & [OII]      &   35.37  & REL      \\\\ & 6C1257+3633& 2.40 & 0.84 & HEG & $\\phantom{>}17.51$ ~ (Eea  5)& 1.003      & [OII]      &   36.58  & REL      \\\\ A& 6C1301+3812& 3.46 & 0.63 & HEG & $\\phantom{>}14.66$ ~ (LLA 12)& 0.470*     & [OII]      &(  35.5  )& ALL      \\\\   \\hline\\hline \\end{tabular} {\\caption[Table 1]{{\\bf (cont).} }} \\end{center} \\end{table*} \\normalsize  \\clearpage   \\clearpage  \\normalsize  \\begin{figure*} \\begin{center} \\setlength{\\unitlength}{1mm} \\begin{picture}(150,180) \\put(80,133){\\special {psfile=\"6c0820.ps\" vscale=35 hscale=35 angle=90}} \\put(80,111){\\special {psfile=\"6c0901_c.ps\" vscale=35 hscale=35 angle=0}} \\put(80,78){\\special {psfile=\"6c0905.ps\" vscale=35 hscale=35 angle=90}} \\put(163,78){\\special {psfile=\"6c0919_c.ps\" vscale=35 hscale=35 angle=90}} \\put(80,23){\\special {psfile=\"6c0930.ps\" vscale=35 hscale=35 angle=90}} \\put(163,23){\\special {psfile=\"6c0943_c.ps\" vscale=35 hscale=35 angle=90}} \\put(80,-32){\\special {psfile=\"6c1011.ps\" vscale=35 hscale=35 angle=90}} \\put(163,-32){\\special {psfile=\"6c1016_c.ps\" vscale=35 hscale=35 angle=90}} \\end{picture} \\end{center} \\vspace{1.3in} {\\caption[junk]{\\label{fig:spectra} Spectra of the 6CE WHT targets with definite or possible spectral features. The synthetic aperture used to extract 1D spectra from the 2D data was typically defined by the full-width at zero-intensity of a cross-cut through the data, excepting the cases of 6C1158+3433 and 6C1217+3645 for which a full-width at half-maximum aperture was used. }} \\end{figure*}  \\addtocounter{figure}{-1}  \\clearpage   \\begin{figure*} \\begin{center} \\setlength{\\unitlength}{1mm} \\begin{picture}(150,180) \\put(80,133){\\special {psfile=\"6c1017.ps\" vscale=35 hscale=35 angle=90}} \\put(163,133){\\special {psfile=\"6c1031_c.ps\" vscale=35 hscale=35 angle=90}} \\put(80,78){\\special {psfile=\"6c1042.ps\" vscale=35 hscale=35 angle=90}} \\put(163,78){\\special {psfile=\"6c1045_c.ps\" vscale=35 hscale=35 angle=90}} \\put(80,23){\\special {psfile=\"6c1113.ps\" vscale=35 hscale=35 angle=90}} \\put(163,23){\\special {psfile=\"6c1123.ps\" vscale=35 hscale=35 angle=90}} \\put(80,-32){\\special {psfile=\"6c1125.ps\" vscale=35 hscale=35 angle=90}} \\put(163,-32){\\special {psfile=\"6c1129_c.ps\" vscale=35 hscale=35 angle=90}} \\end{picture} \\end{center} \\vspace{1.3in} {\\caption[junk]{{\\bf (cont)} }} \\end{figure*}  \\addtocounter{figure}{-1}  \\clearpage   \\begin{figure*} \\begin{center} \\setlength{\\unitlength}{1mm} \\begin{picture}(150,180) \\put(80,133){\\special {psfile=\"6c1134.ps\" vscale=35 hscale=35 angle=90}} \\put(163,133){\\special {psfile=\"6c1158_c.ps\" vscale=35 hscale=35 angle=90}} \\put(80,78){\\special {psfile=\"6c1204+37_c.ps\" vscale=35 hscale=35 angle=90}} \\put(163,78){\\special {psfile=\"6c1204_c.ps\" vscale=35 hscale=35 angle=90}} \\put(80,23){\\special {psfile=\"6c1212_c.ps\" vscale=35 hscale=35 angle=90}} \\put(163,23){\\special {psfile=\"6c1217_c.ps\" vscale=35 hscale=35 angle=90}} \\put(80,-32){\\special {psfile=\"6c1256_c.ps\" vscale=35 hscale=35 angle=90}} \\put(163,-32){\\special {psfile=\"6c1257_c.ps\" vscale=35 hscale=35 angle=90}} \\end{picture} \\end{center} \\vspace{1.3in} {\\caption[junk]{{\\bf (cont)} }} \\end{figure*}   \\clearpage   \\normalsize   \\subsection{Observations and spectroscopic redshifts} \\label{sec:spectra}  The optical spectra presented in this paper were mostly obtained during the course of two WHT observing runs in Jan.\\ 1994 and Jan.\\ 1995 using the ISIS spectrometer; some data taken with the FOS-II spectrograph in Jan.\\ 1992 are also presented (see Cotter, Rawlings \\& Saunders 1996 for details of the observational-setup used in this run). A journal of all these observations is presented in Table~2. All ISIS observations used the R158R and R158B gratings with the beam split using the 5400-$\\rm \\AA$ dichroic; the blue-arm detector was the TEK1 chip on both runs, the red-arm detector was the EEV3 chip in 1994 and the TEK2 chip in 1995. These setups produced continuous spectral coverage from the blue atmospheric cut-off to $\\approx 8750 ~ \\rm \\AA$ in 1994, and to $\\approx 8200 ~ \\rm \\AA$ in 1995. The pixel scales for the TEK and EEV chips were $0.358$ and $0.336$ arcsec pixel$^{-1}$ respectively.  Observations and data analysis followed the standard methods described by Lacy \\& Rawlings (1994). We offset from nearby bright stars to either optical or radio positions, choosing the position angle (PA) of the slit either on the basis of encompassing companion optical objects or to coincide with the radio axis. All the targets were observed below an air mass of 1.2, and all bar two were observed below an air mass of 1.1. All the data presented here were taken in photometric conditions with the seeing close to 1 arcsec. In nearly all cases spectral features, emission lines and/or spectral breaks, were detected and the reduced 1D spectra are presented in Fig.~\\ref{fig:spectra}. For cosmetic purposes, most of the spectra have been smoothed by replacing the value of each spectral bin by the average of the values in the bin and its two neighbours. All measurements from the spectra were made prior to this process, and the results of these measurements are tabulated in Table~\\ref{tab:measurements}.   \\subsubsection{Notes on the 6CE WHT targets} \\label{sec:obj_by_obj}  \\begin{description}  \\item[{\\bf 6C0820+36}] We obtain an unequivocal redshift from the spatially-extended emission lines seen in Fig.~\\ref{fig:spectra}: the alignment of the slit with the radio axis meant that we missed the companion object to the south-west of the radio galaxy in the image of Eales et al.\\ (1997).  \\item[{\\bf 6C0901+3551}] Our WHT spectrum confirms the tentative redshift obtained by Rawlings, Eales \\& Warren (1990).  \\item[{\\bf 6C0905+3955}] Our WHT spectroscopy of this giant radio galaxy has been reported previously by Law-Green et al. (1995b): the Ly$\\alpha$ flux tabulated in Table~\\ref{tab:measurements} is significantly higher than the value given in Table 2 of Law-Green et al.\\ because it is evaluated from the total-light spectrum. The Ly$\\alpha$ emission extends from the continuum-emitting galaxy towards the shorter western radio lobe with a secondary peak approximately midway between the radio galaxy and a companion object visible in the near-IR image of Law-Green et al.\\  \\item[{\\bf 6C0919+3806}] Our WHT spectrum does not yield an unequivocal redshift for this object although two plausible emission features in the spectrum are consistent with $z=1.650$: the redshift given in Table 1 of Eales \\& Rawlings (1996) is almost certainly incorrect. Continuum is detected from $3800 ~ \\rm \\AA$, and it remains roughly flat in $F_{\\lambda}$ until $\\approx 6100 ~ \\rm \\AA$, and then rises (roughly $\\propto \\lambda^{3}$) into the red: the lack of a detected Lyman-limit cut-off means the redshift is less than about $3.2$. We note broad similarities between this spectrum and those of the old, red radio galaxies studied by Dunlop (1999), and by identifying the continuum inflexion with a rest-frame wavelength $\\approx 2600 ~ \\rm \\AA$, we obtain a redshift estimate $z \\sim 1.3$.  \\item[{\\bf 6C0930+3855}] Our WHT spectrum contains a single doubled-peaked line which, since it matches well with lines detected in the near-IR by Eales \\& Rawlings (1996), is securely identified with Ly$\\alpha$. The cause of the double-peaked structure is most probably HI-absorption in the host galaxy.  \\item[{\\bf 6C0943+3958}] Our WHT spectrum contains a single bright emission line, but recent deeper spectroscopy in the optical (P. Best \\& S. Rawlings, {\\it priv. comm.}) has confirmed the redshift obtained if this feature is associated with [OII]3727. The location of the strong line (and weak continuum) emission along the slit confirms that the radio galaxy lies much closer to the eastern than the western radio hotspot in agreement with the near-IR position given by Eales et al. (1997).  \\item[{\\bf 6C1011+3632}] The spectrum of this galaxy is rich in high-ionization emission lines which are neither spatially nor spectrally resolved.  \\item[{\\bf 6C1016+3637}] We aligned the slit through two plausible optical IDs marked on the radio map presented by Law-Green et al.\\ (1995a). Our WHT spectroscopy detects continuum emission from both these objects but separate line emission from about half-way between them; both candidates IDs are therefore likely to be spurious, with the true ID having a Dec (1950.0) of $+36~37~46$. Since this ID is likely to lie along the radio axis it is possible that we have missed some fraction ($\\ltsimeq 50$ per cent) of the line emission. In Table~\\ref{tab:summary} we have placed a limit on the $K-$band magnitude of the true ID equal to the magnitude of the northern optical object, since this was detected at $\\approx 3 \\sigma$ in the near-IR image of Eales et al.\\ (1997).  \\item[{\\bf 6C1017+3712}] We targeted the optical position given by Lilly (1989) confirming the redshift he referenced to H. Spinrad ({\\it priv. comm.}).  \\item[{\\bf 6C1031+3405}] Our 1995 spectrum yields an unambiguous redshift for object `b' of Eales \\& Rawlings (1996) [the redshift given in their Table 1 is incorrect] and its strong emission lines, together with its location near the radio midpoint, mean that this is almost certainly the correct radio galaxy ID. Object `a', an $R-$ band ID proposed by Eales (1985b), was observed in 1994 and has a tentative redshift of 0.449 from a weak but definite line at $7255 ~ \\rm \\AA$, assumed to be [OIII]5007, and a possible line at $\\approx 5413 ~ \\rm \\AA$, assumed to be [OII]3727.  \\item[{\\bf 6C1042+3912}] The Ly$\\alpha$ emission line from this object is detected in the far blue, but two confirming emission lines are also evident in the 2D blue-arm spectrum.  \\item[{\\bf 6C1045+3403}] The redshift of this object is based on five emission lines.  \\item[{\\bf 6C1113+3458}] There is just one emission feature in the spectrum of this object which because of its wavelength and huge equivalent width we take to be Ly$\\alpha$. The lack of any confirming features renders this redshift insecure.  \\item[{\\bf 6C1123+3401}] There is only one definite emission line in the spectrum of this object so its redshift is not yet secure. In $F_{\\lambda}$ the spectrum rises roughly $\\propto \\lambda^{4}$ from $\\lambda \\approx 5000 ~ \\rm \\AA$ until $\\lambda \\approx 7200 ~ \\rm \\AA$ then flattens off. Together with the near-IR photometry (Eales et al.\\ 1997), this produces a spectral energy distribution that is reminiscent of the old red galaxies studied by Dunlop (1999); by identifying the continuum flattening with a rest-frame wavelength $\\approx 3250 ~ \\rm \\AA$, we estimate a redshift $\\sim 1.2$, which means that a plausible identification of the emission line is CII]2326. At this redshift [OII]3727 would lie just beyond the wavelength span of the 1995 WHT spectroscopy and the relatively poor quality of the 1992 WHT (FOS-II) data means that the lack of any detected feature is not a strong argument against the veracity of the tentative redshift.  \\item[{\\bf 6C1129+3710}] Our WHT spectrum contains a single bright emission line, but additional deeper optical spectroscopy (P. Best \\& S. Rawlings, {\\it priv. comm.}) detects additional lines which confirm the redshift obtained if this feature is associated with [OII]3727.  \\item[{\\bf 6C1134+3656}] Our WHT spectrum confirms the tentative redshift obtained by Rawlings et al. (1990), although their claimed CIII]1909 feature is not seen in the new spectrum.  \\item[{\\bf 6C1143+3703}] We observed this source in both 1994 and 1995 (with the same pointing position, but different slit PAs) but detected neither continuum nor lines: excepting the regions of the strongest night-sky emission lines, we can put a $5 \\sigma$ limit on any emission line at the $\\sim 1 \\times 10^{-19} ~ \\rm W m^{-2}$ level. Being only marginally detected at $K-$band, this object is the faintest object in the 6CE sample.  \\item[{\\bf 6C1158+3433}] The spectrum of this object is relatively poor because it was a short exposure taken near twilight. There is a tentative detection of a single emission line which, if taken as [OII]3727, is consistent with a plausible $4000$-$\\rm \\AA$ spectral break. The precise redshift of this object is therefore uncertain but both the spectral shape and magnitude are those expected for a giant elliptical galaxy at $z \\sim 0.5$.  \\item[{\\bf 6C1159+3651}] In 1994 we observed at a PA selected to include a companion object about 6 arcsec to the east (Lilly 1989) and in 1995 at the PA of the radio source. In all the red-arm spectra we see continuum from the ID amounting to $R \\sim 24$, in reasonable agreement with the photometry of Lilly, but there are no emission lines in the coadded data to a $5 \\sigma$ limit of $\\sim 1 \\times 10^{-19} ~ \\rm W m^{-2}$ (excepting the regions of the strong night-sky emission lines). The continuum is roughly flat in $F_{\\lambda}$, although at the signal-to-noise ratio of the detected continuum it is impossible to search for any spectral breaks. Note the following: (i) the redshift given for this source in Table 1 of Eales \\& Rawlings (1996) is almost certainly incorrect; and (ii) the near-IR photometry (Lilly 1989) may include some contribution from the companion object (from which we detected featureless continuum in 1994).  \\item[{\\bf 6C1204+3708}] Since this is a large (51.5 arcsec) radio source, we began our blind exposure with a 5-arcsec wide slit, and on seeing no emission lines in the first red exposure, we widened the slit to 7 arcsec for the second red exposure, and more importantly (given the much higher ratio of read noise to sky noise in the blue) for the remainder of the blue exposure. We detected strong emission lines in the blue-arm spectrum from a location along the slit consistent with the position of the near-IR ID of Eales et al. (1997). The line parameters for Ly$\\alpha$ are rather uncertain because a cosmic ray event effects two pixels near to its location. We also detected the object near the western hotspot in the Eales et al.\\ image: if the single emission line detected from this object is [OII]3727, its redshift is $z=0.864$ making it foreground to the radio galaxy.   \\item[{\\bf 6C1204+3519}] This object provides a salutary lesson on the dangers of assuming that a bright, highly-extended emission line in the blue is always Ly$\\alpha$: the detection of three clear emission lines leaves no doubt that the brightest line, which is extended over 6 arcsec, is CIV not Ly$\\alpha$. The HeII and CIII] lines peak at the location of the continuum, and are also detected in a continuum-free blob $\\approx 2$ arcsec (approximately) to the north; the line fluxes of these two components are roughly equal, although there is additional diffuse material, at least in CIV. In the red-arm spectrum there is also evidence for MgII emission from the blob (although none is clearly detected in the total-light spectrum of Fig.~\\ref{fig:spectra}). There are hints in both the HeII and CIII] emission that the blob is blue-shifted by $\\approx 750 ~ \\rm km ~ s^{-1}$ with respect to the continuum-emitting object, which might also help explain the broad width of the CIV feature in the total-light spectrum. These observations might explain two unusual features of this object: (i) relatively unusually for the 6CE sample, the near-IR emission is well aligned with the radio axis (Eales et al. 1997), perhaps caused by two near-IR objects coinciding with the spatial peaks in the optical line emission; and (ii) the relatively large offset between the putative radio central component (Law-Green et al. 1995a) and the position of the optical ID --- either the `radio core' is actually a knot in the jet, coincident with the emission-line blob, or the active nucleus lies within the blob approximately 2-arcsec to the north of an optically-brighter galaxy.  \\item[{\\bf 6C1212+3805}] The spectrum of this object shows only a single strong emission line which we associate with [OII]3727 since the spectrum has red continuum detected on both sides of the line, and the spectral energy distribution rises rapidly into the near-IR (Eales et al. 1997; see also Maxfield et al. 1995) where the ID is a bright as a typical 6CE galaxy at $z \\sim 1$. We eliminated the possibility that the single bright line is Ly$\\alpha$ by detecting the ID in a deep ($3 \\times 600 ~ \\rm s$) $B-$band image with the WHT Prime Focus Camera on 19 Apr.\\ 1996 since this pass-band would lie below the Lyman limit if the line were Ly$\\alpha$. The line emission is spatially resolved but not in an obviously asymmetric way with respect to the continuum. Our slit also encompassed the companion galaxy to the north-west in the image of Eales et al.\\ from which we detect spatially-extended red continuum but no emission lines.  \\item[{\\bf 6C1217+3645}] Three highly tentative emission features are detected which line up well with a redshift supported by a far-from-convincing $4000$-$\\rm \\AA$ break. The emission from this object is spatially unresolved in seeing of about 1 arcsec, but although the signal-to-noise ratio is too low to be certain, there are no hints that the putative CIII] and MgII features are broad as would be the case if, as suggested by Ben\\'{i}tiz et al. (1995), this object should be re-classified as a quasar. Note, further, that the 1.24-arcsec seeing, near-IR image of this galaxy is resolved (Roche et al. 1998). The radio properties of this source are highly unusual (Best et al. 1999), and one might speculate that the compact nature of the ID, and its relatively blue colour are further indications that this source is anomalous within the sample, one possibility being that it is a radio halo source in a rich cluster rather than a classical double.  \\item[{\\bf 6C1256+3648}] Our WHT spectrum contains only a single definite emission line, but additional deeper optical spectroscopy (P. Best \\& S. Rawlings, {\\it priv. comm.}) detects additional lines which confirm the redshift obtained if this feature is associated with [OII]3727. The emission line region extends from the galaxy over the entire southern radio lobe; no line emission is seen towards the northern lobe. The slit also encompassed a galaxy $\\approx 17$ arcsec to the north-east of the radio galaxy, visible as a similar magnitude galaxy in the image of Eales et al. (1997): we detect an [OII] line in this galaxy at a similar redshift, so this galaxy is a companion to the radio galaxy in line with the suggestion of Lilly (1989) that this field is reminiscent of the core of a rich cluster.  \\item[{\\bf 6C1257+3633}] The extension of the [OII]3727 emission is towards the south eastern radio lobe which shows very little radio polarization (Best et al. 1999). The extended emission shows a distinct blob $\\sim 3$ arcsec in this direction which, following the near-IR/radio registration of Best et al., puts it slightly beyond a prominent knot in the radio emission. Although there is no clear MgII emission in the total-light spectrum (Fig.~\\ref{fig:spectra}), there are hints of MgII emission in the 2D spectrum at the position of the blob, but no hints of high-ionization lines suggesting a lower effective ionization parameter for this emission.   \\end{description}  \\subsubsection{Notes on other 6CE sources} \\label{sec:obj_by_obj_lit}  \\begin{description}   \\item[{\\bf 6C0825+3452}] Inspection of the NVSS image (Condon et al. 1998) of this radio source shows that there is a confusion problem which means that the 151-MHz flux density $S_{151}$ of the main source is likely to lie uncomfortably close to the 6CE sample flux density limit. In the absence of evidence to the contrary we assume that the main source still lies above the $S_{151}$ lower limit. There are two faint near-IR objects close to the position of this radio source: Eales et al. (1997) favoured the brighter south-eastern object as the ID, but the subsequent detection of an inverted radio central component at the position of the fainter north-western object (Best et al. 1999) means that this is more likely to be the correct near-IR counterpart. The redshift was obtained by blind spectroscopy.  \\item[{\\bf 6C1045+3553}] The redshift is based on a single emission line presumed to be [OII].  \\item[{\\bf 6C1045+3513}] The nature and redshift of the ID of this source reported by Naundorf et al. (1992) is incorrect: spectroscopy shows that as originally suspected by Eales (1985b) the object is a quasar and lies at $z = 1.594$, not at $z \\sim 0.7$. The cause of the confusion is the optical/near-IR colour of the object which is similar to that of an elliptical galaxy at $z \\sim 0.7$: the compact near-IR image of Eales et al. (1997) shows that the red colour is the result of light reddening of a quasar nucleus just like 3C318, the nature and redshift of which has recently similarly been revised (Willott, Rawlings \\& Jarvis 2000a).  \\item[{\\bf 6C1255+3700}] The redshift reported for this source by Warner et al. (1983) is incorrect -- we have estimated an observed equivalent width of $\\approx 50 ~ \\rm \\AA$ for the [OIII]5007 line from the spectrum of Eracleous \\& Halpern (1994) which also shows clear broad H$\\beta$.  \\item[{\\bf 6C1301+3812}] No radio central component has been detected for this object, so, as discussed by Allington-Smith, Lilly \\& Longair (1985), identification of the radio source with their galaxy `B' remains an unproven assumption.  \\end{description}   \\begin{figure*} \\begin{center} \\setlength{\\unitlength}{1mm} \\begin{picture}(150,120) \\put(-20, -10){\\special {psfile=\"fig_nz.ps\" vscale=80 hscale=80 angle=0}} \\end{picture} \\end{center} {\\caption[junk]{\\label{fig:nz} Histograms of the redshift distributions $N(z)$ of the 6CE sample (dotted line and cross-hatched) and the 3CRR sample (full line and open). The bin width on each histogram is $\\Delta z = 0.2$. }} \\end{figure*}  \\subsection{Objects without secure spectroscopic redshifts} \\label{sec:noz}  Inspection of Table~\\ref{tab:summary} shows that, excluding the object obscured by a bright star, only two 6CE objects (6C1143+3703 and 6C1159+3651) lack any sort of spectroscopic redshifts. This remarkable (97 per cent) level of redshift completeness is unprecedented at these flux density levels (e.g.\\ McCarthy et al. 1996), and provides compensation for the relatively modest sample size. One of these troublesome objects (6C1159+3651) is relatively bright in the near-IR, but faint in the optical, properties that are characteristic of a galaxy in the redshift range $0.8 \\leq z \\leq 2$ and we take $z=1.4$ here. The other object (6C1143+3703) is faint at $K$ so, rather arbitrarily given the scatter in the $K-z$ relation at faint magnitudes (e.g.\\ Eales et al. 1997), we take $z = 2$.  The 56 spectroscopic redshifts given in Table~\\ref{tab:summary} are not yet all secure. In four cases (6C1045+3553, 6C1113+3458, 6C1123+3401, 6C1212+3805) this is because the redshift depends on the correct identification of a single emission line; in a further three cases (6C0919+3806, 6C1158+3433, 6C1217+3645) this is because there are no clear-cut spectral features in the observed spectrum, but typically several tentative features consistent with a single redshift. Longer spectroscopic exposures of all these seven objects would be useful. In one further case (6C1301+3812) there is some doubt about whether the measured redshift is for the host galaxy of the 6C radio source.   \\subsection{The redshift distribution} \\label{sec:nz}   The redshift distribution of the 6CE sample is compared with that of the 3CRR sample in Fig.~\\ref{fig:nz}. Note the very different redshift distributions resulting from the nature and redshift dependence of the radio luminosity function and the observable comoving volume on our light cone (see Willott et al. 2000c). The median redshift of the 6CE sample is $\\approx 1.1$ with a maximum redshift $\\approx 3.4$, while the 3CRR sample has a median redshift $\\approx 0.55$ and a maximum redshift $\\approx 2$.    \\scriptsize \\begin{table*}  \\begin{center}   \\begin{tabular}{llllllr}  \\hline \\hline  Object   & Date & \\multicolumn{2}{c}{Position (1950.0)} & Exposures & \\multicolumn{2}{c}{Slit Width and PA} \\\\  & & & & & (arcsec) & ($^{\\circ}$) \\\\   \\hline  6C0820+3642 & 27/01/95 & 08 20 33.89 & +36 42 28.3 (r) & ($1 \\times 900$B,  $1 \\times 900$R) & 3 & 155 (r) \\\\  \\hline  6C0901+3551 & 08/01/94 & 09 01 25.10& +35 51 00.8 (r) & ($1 \\times 1800$B, $2 \\times 900$R) & 3 & 120 (r) \\\\  \\hline  6C0905+3955 & 06/01/94 & 09 05 04.95 & +39 55 34.9 (c) & ($1 \\times 1800$B, $2 \\times 900$R) & 3 & 103 (i) \\\\ & 08/01/94 &  09 05 04.95  & +39 55 34.9 (c) & ($1 \\times 1800$B, $2 \\times 900$R) & 3 & 103 (i) \\\\  \\hline  6C0919+3806 & 28/01/95 & 09 19 08.00 & +38 06 51.7 (c?) & ($2 \\times 1800$B, $4 \\times 900$R) & 2 & 151 (r)  \\\\  \\hline  6C0930+3855 & 07/01/94 & 09 30 00.76  & +38 55 07.4 (r) & ($1 \\times 1800$B, $2 \\times 900$R) & 3 & 0 (r)  \\\\  \\hline  6C0943+3958 & 27/01/95 & 09 43 13.67  & +39 58 11.2 (r) & ($1 \\times 600$B, $1 \\times 600$R) & 3 & 100 (r)  \\\\  \\hline  6C1011+3632 & 07/01/94 & 10 11 16.91  & +36 32 12.4 (c) & ($1 \\times 1500$B, $1 \\times 900+ 1 \\times 600$R) & 2 & 0 (r)  \\\\  \\hline  6C1016+3637 & 07/01/94 & 10 16 58.45  & +36 37 49.7 (o) & ($1 \\times 1800$B, $2 \\times 900$R) & 3 & 167 (o)  \\\\  \\hline  6C1017+3712 & 27/01/95 & 10 17 44.70  & +37 12 08.0 (o) & ($1 \\times 600$B, $1 \\times 600$R) & 3 & 45 (r)  \\\\  \\hline  6C1031+3405 & 06/01/94 & 10 31 44.21 & +34 04 54.7 (o) & ($1 \\times 1800$B, $2 \\times 900 $R) & 3 & 135 (r) \\\\ & 27/01/95 & 10 31 44.29  & +34 04 57.9 (r) & ($1 \\times 1800$B, $1 \\times 1800$R) & 3 & 135 (r)  \\\\  \\hline  6C1042+3912 & 07/01/94 & 10 42 23.73 & +39 12 25.2 (r) & ($1 \\times 1800$B, $2 \\times 900 $R) & 3 & 110 (r) \\\\  \\hline  6C1045+3403 & 06/01/94 & 10 45 24.18 & +34 03 37.9 (r) & ($1 \\times 1800$B, $2 \\times 900 $R) & 3 & 110 (r) \\\\  \\hline  6C1113+3458 & 06/01/94 & 11 13 47.65 & +34 58 46.5 (r) & ($1 \\times 1800$B, $2 \\times 900 $R) & 3 & 18 (r) \\\\ & 08/01/94 & 11 13 47.65 & +34 58 46.5 (r) & ($1 \\times 1200$B, $1 \\times 1200 $R) & 3 & 18 (r) \\\\  \\hline  6C1123+3401 & 29/01/92 & 11 23 43.09 & +34 01 57.0 (r) & ($2 \\times 900$B, $2 \\times 900 $R) & 4 &  93  (v) \\\\ & 28/01/95 & 11 23 43.09 & +34 01 57.0 (r) & ($1 \\times 1800$B, $2 \\times 900 $R) & 2 & 176 (v) \\\\ & 31/01/95 & 11 23 43.09 & +34 01 57.0 (r) & ($1 \\times 1800$B, $2 \\times 900 $R) & 2 & 115 (v) \\\\  \\hline  6C1125+3745 & 29/01/92 & 11 25 49.77 & +37 45 25.3 (o) & ($2 \\times 900$B, $2 \\times 900 $R) & 4 &  87  (r) \\\\  \\hline  6C1129+3710 & 06/01/94 & 11 29 55.45 & +37 10 51.9 (r) & ($1 \\times 1000$B, $1 \\times 900 $R) & 3 & 138 (v) \\\\  \\hline  6C1134+3656 & 08/01/94 & 11 34 28.60 & +36 56 30.0 (r) & ($1 \\times 1800$B, $2 \\times 900 $R) & 3 & 130 (r) \\\\  \\hline  6C1143+3703 & 07/01/94 & 11 43 01.23& +37 03 07.4 (r) & ($1 \\times 1800$B, $1 \\times 1800 $R) & 2 &  68 (v) \\\\ & 28/01/95 & 11 43 01.23& +37 03 07.4 (r) & ($1 \\times 1550$B, $1 \\times 1550 $R) & 2 &  90 (v) \\\\  \\hline  6C1158+3433 & 08/01/94 & 11 58 19.50 & +34 33 28.9 (o) & ($1 \\times 600$B, $1 \\times 600 $R) & 3 & 90 (v) \\\\  \\hline  6C1159+3651 & 09/01/94 & 11 59 20.94& +36 51 36.2 (r) & ($1 \\times 1800$B, $2 \\times 900 $R) & 3 &  100 (o) \\\\ & 27/01/95 & 11 59 20.94& +36 51 36.2 (r) & ($1 \\times 1800$B, $1 \\times 1800 $R) & 2 &  32 (r) \\\\   \\hline  6C1204+3708 & 09/01/94 & 12 04 22.08 & +37 08 17.7 (r) & ($1 \\times 1200$B, $2 \\times 600 $R) & 5-7 & 107 (r) \\\\  \\hline  6C1204+3519 & 07/01/94 & 12 04 59.32 & +35 19 46.8 (o) & ($1 \\times 1800$B, $2 \\times 900 $R) & 2 & 12 (r) \\\\  \\hline   6C1212+3805 & 27/01/95 & 12 12 26.04 & +38 05 31.0 (r) & ($1 \\times 1800$B, $1 \\times 1800 $R) & 2.7 & 130 (r) \\\\  \\hline   6C1217+3635 & 09/01/94 & 12 17 40.16 & +36 45 45.4 (o) & ($1 \\times 1200$B, $2 \\times 600 $R) & 2 & 95 (o) \\\\ & 27/01/95 & 12 17 40.16 & +36 45 45.4 (r) & ($1 \\times 1800$B, $1 \\times 1800 $R) & 1.8 & 40 (r) \\\\  \\hline  6C1256+3648 & 06/01/94 & 12 56 44.75 & +36 48 08.4 (o) & ($1 \\times 1800$B, $2 \\times 900 $R) & 3 & 50 (r) \\\\   \\hline  6C1257+3633 & 09/01/94 & 12 57 08.78 & +36 33 11.5 (c) & ($1 \\times 1200$B, $2 \\times 600 $R) & 3 & 138 (r) \\\\  \\hline \\hline  \\end{tabular}  \\end{center} \\caption{\\label{tab:journal} Journal of optical spectroscopy of the 6CE targets. The letter in brackets after the target position denotes its origin: `c' denotes a radio central component; `r', the centre of an extended radio structure; `o', the position of the optical ID; `i', the position of an infrared ID. Exposures gives the time in s of each exposure in the blue (B) and red (R) arms. The letter in brackets after the PA denotes whether the slit alignment was chosen to match the radio (`r') axis, or the major axis of infrared (`i') or optical (`o') structures, or whether the choice was made irrespective of the object properties (`v'), normally at an angle close to the parallactic angle at the time of observation. } \\end{table*}   \\begin{table*}  \\begin{center} \\begin{tabular}{llccrrrl}   \\hline \\hline  Object     & Line & $\\lambda_{rest}$ &  $\\lambda_{meas}$ & Continuum & Flux (\\% Error) & Extent & Comments \\\\ &      & ($\\rm \\AA$)          &  ($\\rm \\AA$)          & (10$^{-22}$ W m$^{-2}$ $\\rm \\AA^{-1}$) & (10$^{-19}$ W m$^{-2}$) & (arcsec) & \\\\  \\hline  6C0820+3642 & Ly$\\alpha$ & 1216 & 3464 &$\\leq10.0$ &15.0 (30) & 7  & FWHM unreliable \\\\ & CIV        & 1549 & 4420 & 20.0   & 1.2 (50) &    & \\\\ & HeII       & 1640 & 4691 & 25.0   & 3.3 (50) &    & near CR removed \\\\ & CIII]      & 1909 & 5463 & 25.0   & 1.6 (50) &    & \\\\ \\hline  6C0901+3551 & Ly$\\alpha$ & 1216 & 3533 &$\\leq10.0$&18.8 (15) & 1  & FWHM = 0-1300\\\\ & CIV        & 1549 & 4498 & 10.0   & 2.6 (50) &    & \\\\ & HeII       & 1640 & 4764 & 10.0   & 2.6 (30) &    & \\\\ & CIII]      & 1909 & 5538 & 25.0   & 1.0 (50) &    & \\\\ \\hline   6C0905+3955 & Ly$\\alpha$ & 1216 & 3505 &$\\leq10.0$& 8.4 (30) & 5  & FWHM = 0-1300\\\\ & CIV        & 1549 & 4457 & 15.0   & 2.0 (40) &    & \\\\ & HeII       & 1640 & 4719 & 20.0   & 0.5 (50) &    & \\\\   \\hline  6C0919+3806 & CIII]?!    & 1909 & 5050 &  5.0   & 0.2 (50) &    & \\\\ & CII]?!     & 2326 & 6174 &  5.0   & 0.5 (50) &    & \\\\  \\hline  6C0930+3855 & Ly$\\alpha$ & 1216 & 4116/4140 & $\\leq5.0$ & 3.0 (30) &$<1$& Double-peaked \\\\   \\hline  6C0943+3958 &[OII]       & 3727 & 7593 & 10.0   & 6.9 (15) & 4 & FWHM = 400-900 \\\\   \\hline  6C1011+3632 &CIII]       & 1909 & 3899 & 20.0   & 1.3 (30) &      & \\\\ &[NeIV]      & 2424 & 4955 & 20.0   & 1.0 (20) &      & \\\\ &MgII        & 2798 & 5711 & 25.0   & 0.6 (50) &      & \\\\ &[NeV]       & 3346 & 6835 & 20.0   & 0.5 (30) &      & \\\\ &[NeV]       & 3426 & 6998 & 20.0   & 1.7 (15) &      & \\\\ &[OII]       & 3727 & 7617 & 20.0   & 1.5 (20) & $<1$ & FWHM=0-400 \\\\ &[NeIII]     & 3869 & 7904 & 20.0   & 2.0 (20) &      & \\\\   \\hline  6C1016+3637 & Ly$\\alpha$ & 1216 & 3516 & $\\leq10.0$& 4.2 (20) & 4  & FWHM =0-1200\\\\ & CIV        & 1549 & 4488 & $\\leq10.0$& 1.0 (30) &    &            \\\\  \\hline   6C1017+3712 &[NeV]       & 3426 & 7025 & 10.0   & 1.4 (50) &      & \\\\ &[OII]       & 3727 & 7651 & 10.0   & 4.5 (30) &  5   & FWHM=750-1100 \\\\  \\hline  6C1031+3405 & Ly$\\alpha$ & 1216 & 3440 & 10.0   & 7.0 (30) & $<1$ & FWHM=0-800 \\\\ & CIV        & 1549 & 4386 & 30.0   & 2.4 (30) &    & \\\\ & HeII       & 1640 & 4646 & 25.0   & 1.0 (30) &    & \\\\ & CIII]      & 1909 & 5404 & 20.0   & 1.7 (40) &    & \\\\ \\hline   6C1042+3912 & Ly$\\alpha$ & 1216 & 3368 &$\\leq 10.0$& 8.5 (30) & $<1$ & FWHM=0-1100\\\\ & CIV        & 1549 & 4291 &$\\leq 10.0$& 1.0 (30) &      & \\\\ & HeII       & 1640 & 4544 &$\\leq 10.0$& 0.6 (50) &      & \\\\  \\hline   6C1045+3403 & Ly$\\alpha$ & 1216 & 3435 &$\\leq 10.0$&25.8 (15) &  5   & FWHM=800-1800\\\\ & CIV        & 1549 & 4382 & 15.0      & 1.5 (30) &      & \\\\ & HeII       & 1640 & 4635 & 15.0      & 1.2 (50) &      & \\\\ & CII]       & 2326 & 6567 & 10.0      & 0.6 (60) &      & \\\\ &[NeIV]      & 2424 & 6835 & 20.0      & 0.6 (50) &      & \\\\   \\hline \\hline  \\end{tabular}  \\end{center} \\caption{\\label{tab:measurements} Measurements obtained from the WHT spectra of 6CE radio sources. The `?' denotes an uncertain line identification and the `!' denotes a feature which is plausibly spurious. Errors on the line fluxes represent $\\approx$90\\% confidence intervals expressed as a percentage of the best line flux estimate; for the strongest lines these are dominated by roughly equal contributions from uncertainties in fixing the local continuum level, and from the absolute flux calibration (and plausible slit losses). The spatial extents of the emission lines were estimated by evaluating the full-width zero-intensity of a cross-cut through the emission line, deconvolved from the seeing. Line widths (given in units of km s$^{-1}$) were estimated from the FWHM of the best Gaussian fit to each line, the lower value of a range assumes the line-emitting region fills the slit, and the higher value assumes that it is broadened only by the seeing. `CR contam' is short-hand for cosmic-ray contamination. } \\end{table*} \\normalsize  \\clearpage   \\addtocounter{table}{-1}   \\scriptsize  \\begin{table*}  \\begin{center} \\begin{tabular}{llccrrrl}   \\hline \\hline  Object     & Line & $\\lambda_{rest}$ &  $\\lambda_{meas}$ & Continuum & Flux (\\% Error) & Extent & Comments \\\\ &      & ($\\rm \\AA$)          &  ($\\rm \\AA$)          & (10$^{-22}$ W m$^{-2}$ $\\rm \\AA^{-1}$) & (10$^{-19}$ W m$^{-2}$) & (arcsec) & \\\\  \\hline  6C1113+3458 & Ly$\\alpha$?& 1216 & 4142 & $\\leq10.0$& 2.2 (20) & 3  & FWHM = 0-900\\\\  \\hline  6C1123+3401 & CII]?      & 2326 & 5227 &  6.0 &  0.4 (50) &   &           \\\\  \\hline  6C1125+3745 & CIV        & 1549 & 3471 & 77.0 & 48.0 (30) & $<1$ & FWHM $\\sim$ 10000\\\\ & CIII]      & 1909 & 4237 & 80.0 &  9.0 (50) &   &           \\\\ & MgII       & 2798 & 6249 & 65.0 & 15.0 (25) &   &           \\\\  \\hline  6C1129+3710 & [OII]      & 3727 & 7678 & 45.0 &  8.2 (30) & 7 & FWHM = 0-700\\\\ & [NeIII]?!  & 3869 & 7974 & 45.0 &  1.2 (60) &   &           \\\\  \\hline   6C1134+3656 & Ly$\\alpha$ & 1216 & 3800 &$\\leq 20.0$&13.0 (30) &  15   & FWHM=1000-1700\\\\ & CIV        & 1549 & 4839 & 25.0      & 2.0 (40) &       & \\\\ & HeII       & 1640 & 5121 & 30.0      & 2.6 (40) &       & \\\\   \\hline   6C1158+3433 & [OII]?!    & 3727 & 5704 & 65.0 &  2.5 (50) & $<1$ & FWHM unreliable\\\\  \\hline   6C1204+3708 & Ly$\\alpha$ & 1216 & 3379 & 20.0      &47.0 (40) &  4   & CR contam\\\\ & CIV        & 1549 & 4305 & 35.0      &11.5 (25) &      & FWHM=1200-1400\\\\ & HeII       & 1640 & 4635 & 20.0      & 3.7 (30) &      & \\\\ & CIII]      & 1909 & 5308 & 25.0      & 2.1 (40) &      & \\\\  \\hline  6C1204+3519 & CIV        & 1549 & 3681 & 20.0   &  4.5 (25) &  6 & FHWM=2000-2500  \\\\ & HeII       & 1640 & 3894 & 25.0   &  1.5 (50) &    & CR contam? \\\\ & CIII]      & 1909 & 5404 & 20.0   &  1.6 (40) &    &            \\\\  \\hline  6C1212+3805 & [OII]?     & 3727 & 7256 & 10.0   &  3.0 (20) &  2 & FWHM=550-950\\\\  \\hline  6C1217+3645 & CIII]?!    & 1909 & 3965 & 30.0   &  0.7 (60) &$<1$& CR contam  \\\\ & MgII?!     & 2798 & 5830 & 25.0   &  0.7 (60) &    &            \\\\ & [OII]?!    & 3727 & 7793 & 30.0   &  0.9 (60) &    &            \\\\ \\hline   6C1256+3648 & [OII]      & 3727 & 7926 & 20.0   &  3.0 (40) &  6 & FWHM=0-500\\\\ & [NeIII]?!  & 3869 & 8232 & 25.0   &  1.0 (60) &    &           \\\\  \\hline  6C1257+3633 & CIII]      & 1909 & 3825 & 20.0   &  2.9 (40) &    &           \\\\ & [NeIV]     & 2424 & 4855 & 15.0   &  1.3 (50) &    &           \\\\ & [NeV]      & 3346 & 6705 & 40.0   &  0.5 (60) &    &           \\\\ & [NeV]      & 3426 & 6862 & 35.0   &  1.2 (50) &    &           \\\\ & [OII]      & 3727 & 7465 & 35.0   &  6.4 (20) &  5 & FWHM=0-500\\\\ & [NeIII]    & 3869 & 7748 & 40.0   &  3.2 (30) &    &           \\\\   \\hline \\hline  \\end{tabular}  \\end{center} \\caption{{\\bf (cont).}} \\end{table*} \\normalsize  \\clearpage   \\subsection{Emission line properties} \\label{sec:disco}  At redshifts $> 1.75$ the only 3CRR objects known are, in terms of their radio luminosity, amongst the most extreme objects on our light cone: the 16 $z > 1.75$ 6CE objects are typically $\\sim 1$ dex less radio luminous. Thirteen of these 16 objects lie in very restricted bands of luminosity and and redshift, $1.75 \\leq z \\leq 2.45$ and $27.3 \\leq \\log_{10} (L_{\\rm 151}) < 27.8$, and we will take their optical properties to be characteristic of `high-redshift 6CE sources'. The two extreme luminosity (and redshift) 6CE sources have been studied in detail elsewhere (6C0902+3419 by, for example, Lilly 1988, Eales et al. 1993b and Carilli 1995; and 6C1232+3942 by Eales et al. 1993a), and the lowest luminosity $z > 1.75$ object is distinct in its radio properties (its projected linear size is more than 1 dex lower than all other 6CE objects st $z > 1.75$) and as discussed in Sec.~\\ref{sec:noz}, is one of the two 6CE sources without a spectroscopic redshift.  As well as spanning narrow ranges of $L_{151}$ and $z$, the thirteen high-redshift 6CE sources also have strikingly similar optical properties. Excepting 6C0824+3535, the objects are narrow-line radio galaxies rather than quasars, and for the radio galaxies the distribution in Ly$\\alpha$ luminosities $L_{\\rm Ly \\alpha}$ is very narrow: from Table~\\ref{tab:summary} we find $\\log_{10} L_{\\rm Ly \\alpha} = 36.45 \\pm 0.36$. The tight spread in $\\log_{10} L_{\\rm Ly \\alpha}$ is presumably a manifestation of the well-known correlation between narrow-line and radio luminosities (e.g.\\ Willott et al. 1999). Eleven of the twelve relevant 6CE radio galaxy spectra are presented in this paper (Sec.~\\ref{sec:spectra}) and their apparent homogeneity encouraged us to create a composite spectrum which is hopefully representative of a high-$z$ 6CE radio source. This composite was constructed by: (i) changing the wavelength scale of each spectrum to place each object at redshift zero, ensuring each had the same binning in wavelength; (ii) scaling each de-redshifted spectrum to equalize the Ly$\\alpha$ fluxes; and (iii) taking a bin-by-bin average with an iterative sigma-clipping rejection algorithm (with the clip set at $\\pm 1.5 \\sigma$). The composite is shown in Fig.~\\ref{fig:compo}; the line fluxes relative to Ly$\\alpha$ of CIV and HeII are 0.13 and 0.08 respectively, and there are marginally significant detections of the NV and CIII] lines. The CIII] line is clearly detected in at least four of the objects making up the composite, and its low signal-to-noise ratio in Fig.~\\ref{fig:compo} is probably partially due to its location (at high redshift) within the spectral region of the dichroic cross-over for the WHT observations. The rest-frame equivalent widths of Ly$\\alpha$, CIV and HeII are within a factor of two of $450$, $70$ and $30 ~ \\rm \\AA$ respectively. These results are in good quantitative agreement with the composite spectrum presented by McCarthy (1993) which, since it was calculated from an independent but similar dataset (3C galaxies plus radio galaxies with similar low frequency flux densities to the 6CE sample), shows that there does appear to be such a thing as a generic UV spectrum of radio-luminous high-redshift galaxies.   \\begin{figure*} \\begin{center} \\setlength{\\unitlength}{1mm} \\begin{picture}(150,100) \\put(160,0){\\special {psfile=\"composite_final.ps\" vscale=70 hscale=70 angle=90}} \\end{picture} \\end{center} {\\caption[junk]{\\label{fig:compo} Composite spectrum resulting from the combination of the 11 spectra of $z > 1.75$ 6CE galaxies discussed in Sec.~\\ref{sec:spectra} with $z > 1.75$; the method of combination is described in Sec.~\\ref{sec:disco}, and the resultant spectrum has been smoothed with a 3-bin boxcar filter. The flux density is measured in units of $\\rm W ~ m^{-2} ~ \\AA^{-1}$ with an arbitrary normalization. }} \\end{figure*}   \\subsection{Quasar fraction} \\label{sec:qf}  The low fraction of quasars in the 6CE sample was first noticed and discussed by Eales (1985b,c). There are only three quasars out of the 18 6CE objects in the $27.3 \\leq \\log_{10} (L_{\\rm 151}) < 27.8$ luminosity bin whereas amongst the 41 similarly radio-luminous 3CRR sources, between 14 and 16 (depending on how one classifies 3C22 and 3C41; see Simpson, Rawlings \\& Lacy 1999) are quasars, i.e.\\ a fraction $\\approx 0.4$. Using a combined 3CRR/6CE/7CRS dataset, Willott et al. (2000b) have argued that this fraction is virtually independent of $L_{\\rm 151}$ and $z$ above a critical value of $\\log_{10} (L_{\\rm 151}) \\approx 26.5$. Taking 0.4 as the probability that a given source is viewed as a quasar, the Binomial probability of finding three or fewer 6CE quasars out of 18 6CE objects is about 3 per cent.  One possible reason for this (marginally significant) deficit of quasars is that at the higher redshifts probed by the 6CE sample (see Fig.~\\ref{fig:pz}) quasar nuclei may be more easily hidden by dust: perhaps partly because optical spectra probe shorter rest-frame wavelengths, and perhaps partly because of an intrinsically larger fraction of reddened lines-of-sight at high redshift (e.g.\\ Willott et al. 2000a). Just one of the luminous 6CE radio galaxies needs to be a reddened quasar for the binomial probability of the low quasar fraction in 6CE to rise to $\\sim 10$ per cent. Note also that, at the highest radio luminosities and redshifts, one of the two 6CE galaxies is 6C0902+3419 which, as pointed out by Carilli (1995), has radio properties which are consistent with a jet orientation as close to the line-of-sight as a lower-redshift 3CRR quasar. We conclude that the low quasar fraction observed in the 6CE sample, although still plausibly an artefact of small number statistics, hints at one or more redshift-dependent effects that warrant further investigation in a larger sample of high-$z$ radio sources."
    },
    "0010/astro-ph0010359_arXiv.txt": {
        "abstract": "We present the results of a search for variability in and near the core of the metal-rich, obscured globular cluster Terzan 5, using NICMOS on {\\it HST}. This extreme cluster has approximately solar metallicity and a central density that places it in the upper few percent of all clusters. It is estimated to have the highest interaction rate of any galactic globular cluster. The large extinction towards Terzan 5 and the severe stellar crowding near the cluster center present a substantial observational challenge. Using time series analysis we discovered two variable stars in this cluster. The first is a RRab Lyrae variable with a period of $\\sim$0.61 days, a longer period than that of field stars with similar high metallicities. This period is, however, shorter than the average periods of RR Lyraes found in the metal-rich globular clusters NGC 6441, NGC 6388 and 47 Tuc. The second variable is a blue star with a 7--hour period sinusoidal variation and a likely orbital period of 14 hours. This star is probably an eclipsing blue straggler, or (less likely) the infrared counterpart to the low mass X-ray binary known in Terzan 5. Due to the extreme crowding and overlapping Airy profile of the IR PSF, we fall short of our original goal of detecting CVs via P$\\alpha$ emission and detecting variable infrared emission from the location of the binary MSP in Terzan 5.\\\\ ",
        "introduction": "The study of variable stars in globular clusters has a long, rich history. Studies of RR Lyrae variables, long period variables and Population II cepheids have addressed fundamental questions about stellar evolution, globular cluster evolution and the distance scale. More recently, observations of variable stars such as millisecond pulsars (MSPs), cataclysmic variables (CVs) and eclipsing binaries have been motivated by studies of stellar interactions and globular cluster dynamics. For example, during stellar encounters binaries are expected to become more tightly bound and give the energy difference up to passing stars, a process that can support a cluster against core collapse.  Observations with {\\it HST} have been crucial in overcoming the significant crowding present in such dense stellar environments.  For example, {\\it HST} observations of the center of the massive globular cluster 47 Tucanae have discovered two likely CVs (Paresce and de Marchi 1994 and Shara et al. 1996), nine eclipsing binaries (Edmonds et al. 1996) and 6 pulsating blue stragglers (BSs) (Gilliland et al. 1998). Observations of the collapsed--core cluster NGC 6397 have discovered 5 CVs (Cool et al. 1995, Grindlay et al. 1995; Edmonds et al. 1999; Grindlay 1999) and 3 likely low-mass WDs (Cool et al. 1998 and Edmonds et al. 1999) that may also be MSPs.  {\\it HST} has also been used to obtain the first high quality images and color magnitude diagrams (CMDs) of NGC 6441 and NGC 6388, two partially obscured clusters with surprisingly rich blue horizontal branches (HBs) when their high metallicity is taken into account (Rich et al. 1997). Most metal--rich clusters have red, stubby HBs, in agreement with standard theories of stellar evolution. As metallicity decreases HBs usually become increasingly blue, but some clusters, such as NGC 6441 and NGC 6388, contradict this pattern and a second parameter is assumed to affect HB morphology.  The cause of this ``2nd parameter effect'' is a subject of vigorous debate, but some intriguing clues have appeared. Both NGC 6441 and NGC 6388 appear to have HBs that are brighter than the HBs in metal--poor clusters. Variability searches in these clusters have subsequently shown that the RR Lyrae stars have significantly longer periods than expected for metal-rich stars (Layden et al. 1999 and Pritzl et al. 2000), as does the RR Lyrae V9 in 47 Tuc (Carney et al. 1993). This apparently new subclass of long period, metal--rich RR Lyraes may show the lack of universality of the luminosity--metallicity relation for RR Lyraes and may require the establishment of a third Oosterhoff group (Pritzl et al. 2000).  The globular cluster Terzan 5, the subject of this paper, represents an excellent laboratory for the studies introduced above. Because its metallicity is close to solar (Ortolani, Barbuy and Bica 1996; Cohn et al. 2000) it is ideal for studies of HB morphology and RR Lyrae behavior in a metal--rich globular cluster. It also has the highest predicted rate of stellar collisions of any galactic globular cluster (Verbunt \\& Hut 1987) making it a laboratory for stellar dynamics. It is already known to contain a LMXB (Johnston, Verbunt and Hasinger 1995), a binary MSP (Lyne et al. 1990; Nice \\& Thorsett 1992), one other pulsar and a large number of unresolved pulsars (Fruchter \\& Goss 2000). Unfortunately Terzan 5 is highly reddened, with E(B-V) = 2.49 (Ortolani et al. 1996) and because of its large mass the crowding near cluster center, where the binaries are expected to be concentrated, is a formidable problem.  Here, we describe an attempt to overcome the high reddening and crowding in Terzan 5 by observing it in the IR with NICMOS on {\\it HST}. Our aims were (1) to search the center of the cluster for CVs and the IR counterpart of the LMXB using wide (F187W) and narrow-band (F187N) NICMOS filters centered on Paschen-$\\alpha$ (P$\\alpha$), and (2) to search for variability from the binary MSP. The general photometric results are described in Cohn et al. (2000). In this paper we describe a detailed search for variability in two fields, one covering the center of the cluster and another containing the binary MSP. Although we fall short of detecting CVs because of crowding and no evidence is found for variability of any of the objects lying in the error circle of the MSP, we do detect an interesting RR Lyrae variable and a second variable that is likely to be an eclipsing BS, or (less likely) the IR counterpart of the LMXB. ",
        "conclusions": "\\subsection{RR Lyrae variable}  The identification of V1 as a RR Lyrae variable and a cluster member is secure, based on the light curve, the position of the star in the CMD and the small chance of field star contamination.  As with 47 Tuc, only one RR Lyrae has been discovered in Terzan 5 and the bulk of the HB stars are red and lie near the giant branch (see Figure \\ref{fig.hpcmd}), as they do in most metal-rich globular clusters. However, Cohn et al. (2000) report the possible detection of a small number of blue HB stars in Terzan 5. The high reddening and crowding in this cluster makes such detections much more difficult than in NGC 6388 and NGC 6441 where many blue HB stars have been found and there are multiple RR Lyrae variables.  Rich et al. (1997) have explored two possibilities to explain the blue HB star populations in NGC 6388 and NGC 6441, the effects of age and dynamics. At high metallicity an increase in age can populate a blue HB tail with stars at the low end of the mass distribution, while dynamical encounters in the cores of dense clusters can result in extra mass loss via tidal stripping. Rich et al. (1997) conclude that although neither effect alone appears sufficient to explain the cluster blue HBs, a combination of them may have an effect.  Our discovery of a RR Lyrae variable in Terzan 5 is therefore useful for the study of the influence of metallicity, age and interaction rate on the production of HB stars in the instability strip. For example, Terzan 5 has a higher metallicity than any of the above clusters which should result in fewer RR Lyraes, but it also has a higher interaction rate which may result in a higher number of RR Lyraes (see Rich et al. 1997). It is possible that these two factors combine. For example, maybe a large number of stars in Terzan 5 lose mass via tidal stripping but the cluster metallicity is high enough that only the stars with the most extreme mass loss show up as RR Lyraes.  Sweigart and Catelan (1998) note that the HBs in NGC 6388 and NGC 6441 have a pronounced upward slope with decreasing $B-V$, a result which is not predicted by canonical HB models and which cannot be reproduced by either a greater cluster age or enhanced mass loss along the red giant branch. They explore three possibilities to explain these sloped HBs: (1) a high helium abundance, (2) a rotation scenario in which internal rotation during the RGB phase increases the HB core mass, and (3) a helium-mixing scenario in which deep mixing on the RGB enhances the the envelope helium abundance.  One prediction of the high-Y scenario is that the number ratio $R$ of HB stars to RGB stars brighter than the mean RR Lyrae luminosity should be large. We calculated $R$ for Terzan 5 by defining the HB as those stars having 15.6 $<$ F110W $<$ 16.1 and correcting for the large number of RGB stars found in this region of the CMD by averaging and subtracting the number of stars present in adjacent magnitude bins (15.1 $<$ F110W $<$ 15.6 and 16.1 $<$ F110W $<$ 16.6). Our value for $R$ is 1.03 $\\pm$ 0.11, where the error is based on Poisson statistics alone. Although systematics will increase the error, this value is clearly inconsistent with the high-Y scenarios presented in Sweigart and Catelan (1998), where $R=3-4$. Layden et al. (1999) came to a similar conclusion for their study of NGC 6441. Sweigart and Catelan (1998) suggest a variety of other observational tests for discriminating among the various scenarios, all requiring further observations.  Figure \\ref{fig.hpcmd} shows that V1 may be somewhat brighter ($\\sim$ 0.3 mag) than expected for a blue HB star, consistent with the sloped HB slopes mentioned above, although uncertainties in the large reddening for this cluster and the limitations of our simple HB model make detailed comparisons difficult.  As Sweigart and Catelan (1998) point out, unusually bright RR Lyraes should have unusually long periods, as observed for the RR Lyraes in NGC 6388, NGC 6441 and 47 Tuc. This self-consistency between the RR Lyrae luminosities and periods is reassuring.  The obvious question arises: is the period of V1 unusually long?  First, the metallicity has to be considered. V1 is likely to be the most metal-rich RR Lyrae yet found in a globular cluster, since Terzan 5 is thought to have roughly solar metallicity (Ortolani et al. 1996) or even higher (Cohn et al. 2000 estimate [Fe/H]=0.25). NGC 6388, NGC 6441 and 47 Tuc have [Fe/H] values of --0.60, --0.53 and --0.76 respectively (Harris 1999)\\footnote{The Catalog of Parameters for Milky Way Globular Clusters, compiled by William E. Harris, McMaster University, revised June 22, 1999 is available at http://physun.mcmaster.ca/$\\sim$harris/mwgc.dat}. It may even be the highest metallicity RR Lyrae ever found.  Examination of the metallicity--period relationship for field RR Lyraes (Layden 1995) shows that V1 has a significantly longer period than field stars with similar metallicity. The period-metallicity distribution of Pritzl et al. (2000) shows that V1 lies well outside the two Oosterhoff classes, with a longer than expected period given the metallicity of Terzan 5.  Therefore V1 joins the rapidly growing subclass (suggested by Layden et al. 1999) of long period RR Lyraes in metal rich globular clusters.  Also plotted on Figure 1 of Pritzl et al. (2000) are the periods and metallicities of the RR Lyraes in NGC 6388 and NGC 6441. These two clusters are clearly separate from the two Oosterhoff classes, with higher metallicity and longer periods. Interestingly, while the average period for these two clusters is longer than {\\it all} of the OoI and OoII globulars, V1 has a period lying almost exactly between the 2 Oosterhoff groups. Further, a line drawn between V1 and the average of NGC 6388 and NGC 6441 in the period--metallicity plot lies almost parallel to the line between the 2 Oosterhoff classes, suggesting two period--metallicity relations with similar slope may exist.  This interesting but speculative result clearly requires verification by observing a larger sample of RR Lyraes in clusters with solar metallicity or above. It could be that V1 has an unusually short period compared to other (undetected) RR Lyraes in this cluster. For example, the periods in NGC 6441 alone range from $\\sim$0.6 days to $\\sim$ 0.9 days, with a reasonably flat distribution between these two extremes.  \\subsection{The faint variable V2: LMXB or eclipsing BS?}  To summarize our results, V2 is blue, relatively faint and variable, with a near sinusoidal light curve, and a period of $\\sim$7 hr. We consider two possibilities for this variable, that it is a LMXB or that it is an eclipsing BS.  First, we consider the LMXB hypothesis. In this case the variability could be caused by at least three different phenomena: (1) Ellipsoidal variations from distortion of the secondary. These variations typically have a distinctive double-lobed pattern, with a difference in depth between successive minima of $\\sim$0.05--0.10 mag and an expected difference in amplitude between F110W and F187W of $\\sim$0.05 mag, using reasonable assumptions regarding mass ratios, temperatures and inclinations (van Hamme, private communication). These amplitude differences are not clearly detected in the data, though the relatively large error for the F187W amplitude means we cannot rule out this possibility.  (2) Another possible explanation of variability, under the LMXB hypothesis, is the effects of heating of the secondary by the disk and neutron star. Here, the expected light curve is also double-lobed, but with the positions of the deep and shallow minima reversed (see van Paradijs \\& McClintock 1995). A large contribution from heating would be obvious in the light curve because of the significant expected difference in eclipse depth, but a small contribution would plausibly cancel out the difference between the depths of adjacent minima, when combined with an ellipsoidal variation. However, this requires some fine-tuning of the binary parameters and a difference between the light curve amplitudes remains (Van Hamme, private communication). Note that if either case (1) or (2) applies to V2 then the period would be twice the 7-hour sinusoidal variability period, ie $\\sim$14 hours. It would thus be similar to the field LMXB, and recurrent transient, Aql X-1 (Chevalier \\& Ilovaisky 1991), an attractive possibility given the transient, or highly variable nature of the Terzan 5 LMXB.  (3) Flickering from the disk may also be expected to show up in the light curve of a LMXB, as it does in field systems. If V2 is a LMXB, the blue color of V2 would imply a reasonably prominent disk (the source of flickering).  However, no evidence for flickering is seen as the residuals after subtracting the sinusoidal model are consistent with the noise properties of the data.  Besides being blue, variable objects, LMXBs also typically have weak emission lines in the optical and infrared.  Shahbaz et al. (1996) show that the average H$\\beta$ equivalent width (EW) of 15 LMXBs, is 1.9 \\AA. Scaling by the average EW ratio between H$\\beta$ and P$\\alpha$ for the two CVs PQ Gem and EX Hya (using values from Dhillon et al. 1997; no published P$\\alpha$ EWs for LMXBs were found) we expect an average P$\\alpha$ EW of 3.8 \\AA. This is equivalent to F187N-F187W = --0.07, well below the 1$\\sigma$ error for the faint end of our F187N vs F187N-F187W CMD (containing V2) of 0.11 mag. Therefore the P$\\alpha$ data is inconclusive in determining whether V2 is an LMXB.  An obvious constraint on the identify of V2 is its position relative to the X-ray source identified by Johnston et al. (1995). Our measured positional offset of 5.5$''$ is just outside the 5$''$ error circle quoted by Johnston et al. (1995). We hope to significantly reduce the size of this error circle with the use of our Cycle 1 Chandra observations. With these deep, 50ks observations the LMXB should have an error circle $< 1''$ and should be distinguishable from the large expected population of CVs and MSPs in this cluster, unless it is in deep quiescence.  We may also compare the period and luminosity of V2 with UV or optically identified globular cluster LMXBs.  Three of the six globular cluster LMXBs with optical counterparts have $M_V \\gtrsim$4.5, according to Deutsch, Margon and Anderson (2000).  Figure \\ref{fig.simul1} shows that this is beyond our detection limits, for stars on the main sequence or the giant branch (for hotter objects our detection limits are even shallower). To make matters worse the Terzan 5 LMXB, being a transient should be fainter than most other cluster LMXBs.  We can predict $M_V$ using the dependence of optical luminosity upon X-ray luminosity and period (van Paradijs and McClintock 1994) a relation that holds for both cluster and field systems as shown by Deutsch et al. (2000). For a period of $\\sim$14 hrs, we estimate $M_V=4.2$, beyond our detection limits. Finally, we note that 4 of the 5 systems with optical counterparts and measured periods have period $\\lesssim$ 5.7 hours and three have periods $\\lesssim$ 0.73 hours.  To summarize, although we find no {\\it individual} argument gives compelling evidence against V2 being the infrared counterpart of the Terzan 5 LMXB, the accumulated weight of evidence argues against it. An alternative hypothesis is that V2 is an eclipsing BS. This is consistent with: (1) The position of V2 in the F110W vs F110W-F187W CMD (see figure \\ref{fig.hpcmd}). V2 has about the same luminosity as the brightest BS in 47 Tuc, and its luminosity may also be significantly enhanced by its binary companion. (2) The light curve amplitudes of V2. Sinusoidal light curves with our measured amplitudes are easy to reproduce using the light curve models of Wilson for W UMa systems (van Hamme, private communication). A growing number of binary BSs are being discovered in globular clusters (eg Edmonds et al. 1996 and Kaluzny et al. 1998), although the fraction of BSs that are binaries is typically quite low.  (3) The inferred orbital period of 14 hours is consistent with the periods of other binary BSs in globular clusters such as 47 Tuc (Edmonds et al. 1996). As with the LMXB hypothesis, the P$\\alpha$ data is inconclusive, since for a model BS with $T_{eff}$ = 8000 K (at the top of our artificial BS sequence), F187N-F187W = 0.04, well within our 0.11 mag scatter for this line discriminant.  Using our simulations we can ask how many eclipsing binaries we should expect to see with our data set.  Since most eclipsing binaries discovered in clusters are found near the MSTO (eg Edmonds et al. 1996) and we have very little sensitivity to such variables, the simple answer is none. Therefore, a naive calculation may predict a very large number of eclipsing systems (and BSs) should be found in Terzan 5, if V2 is indeed an eclipsing BS. While clearly small number statistics apply since we have detected only one system, it is striking that evidence exists for a bright eclipsing BS at the edge of our detection limits. A relatively large number of BSs and binaries would have interesting consequences for the dynamics of this extreme globular cluster. For example, a high BS number may be caused by a high stellar interaction rate, which itself may be enhanced by a large binary fraction. A high interaction rate is expected for Terzan 5 because of its moderately sized core and extremely high stellar density (Cohn et al. 2000). Terzan 5 also may be the globular cluster with the largest population of MSPs in the galaxy (Fruchter \\& Goss 2000), further evidence for possible large interaction rates.  It is useful to discuss the potential for detailed studies of other interaction products or binaries in Terzan 5. Our upcoming GO Chandra observations of Terzan 5 should allow a deep search for CVs in this cluster, unless the LMXB is in outburst. Unfortunately optical or IR identifications of such sources will be extremely difficult or impossible. We have already seen the problems with IR observations, where crowding prevented us from reaching even close to the MSTO, but with their smaller PSFs, Gemini and NGST will offer substantial improvements over {\\it HST} for studying this cluster.  In optical passbands such as $V$, the PSF will be substantially reduced in size but the enormous reddening for this cluster implies that the MSTO will be found near $V \\sim$25.5, and therefore CVs at $V\\gtrsim 28$.  Studies of the BS sequence will be difficult but may be feasible depending on the results of detailed crowding simulations."
    },
    "0010/astro-ph0010390_arXiv.txt": {
        "abstract": "Assuming LTE, we synthesise solar G band spectra from the semiempirical flux-tube model of Briand \\& Solanki (1995). The results agree with observed G-band bright-point contrasts within the uncertainty set by the amount of scattered light.  We find that it is the weakening of spectral lines within the flux tube that makes the bright-point contrast in the G band exceed the continuum contrast. We also synthesise flux-tube spectra assuming LTE for the full wavelength range from UV to IR, and identify other promising passbands for flux-tube observations. ",
        "introduction": "Filtergrams taken in the G-band around 430.5~nm are often used for proxy magnetometry since photospheric bright points associated with magnetic structures show up very well in them (e.g.\\ Berger et al.\\ 1995). We have performed LTE spectrum synthesis of the G band (mostly formed by CH) using the semi-empirical magnetic flux-tube model NCHROM7 of Briand \\& Solanki (1995) which is a refinement of the models of Bruls \\& Solanki (1993) and Solanki (1986). It is embedded in a standard model for the quiet photosphere.  Figure~1 shows results. The hot bright walls of the (rotationally symmetric) flux tube create a bright ring seen as a double peak in the radiation temperature. Figure~1 % also shows the temperature structure along three lines of sight: through quiet sun, through the tube centre, and through the tube wall where the emergent intensity peaks considerably because it samples the quiet-sun model at larger depth than the quiet-sun line of sight does. The peak contrast with respect to the quiet sun is larger in $G$ than in $C$ mainly because the photosphere appears darker in the G band.  Detailed inspection of the synthetic spectra shows how most spectral lines weaken in the flux tube.  We interpret this as due to the shallower temperature gradient and the lower density inside the flux tube. Molecular lines weaken the most due to the pressure dependence of the molecular association equilibria. This is the reason that bright points should show up particularly well in molecular lines. ",
        "conclusions": "The Briand \\& Solanki model produces bright-point contrasts similar to those observed at the SVST. However, the spectrum synthesis was made assuming LTE and the photospheric part of the flux tube model was primarily constructed from LTE inversions of spectropolarimetric data. Do we believe these LTE assumptions? Not necessarily. A contrasting scenario would be one where UV radiation from the hot walls photodissociates CH in the flux tube and also causes overionisation of metals, thus affecting our G-band synthesis as well as the semi-empirical modelling. If there is any place in the solar photosphere where 3D--NLTE effects are important, it should be in these flux tubes."
    },
    "0010/nucl-th0010084_arXiv.txt": {
        "abstract": "We introduce in this work a  class of relativistic models for nuclear matter and neutron stars which exhibits a parameterization, through mathematical constants, of the non-linear meson-baryon couplings. For appropriate choices of the parameters,   it recovers current QHD models found in the literature: Walecka and the Zimanyi and Moszkowski models (ZM and ZM3). For other choices of the parameters, the new models give very interesting and new physical results.We have obtained numerical values for the maximum mass and radius of a neutron star sequence, red-shift, hyperon populations, radial distribution of particles, among other relevant static properties of these stellar objects. The phenomenology of neutron stars in ZM models is presented and compared to the phenomenology obtained in other versions of the Walecka model. We have found that the ZM3 model is too soft and predicts a very small maximum neutron star mass, $\\sim0.72 M_\\odot$. A strong similarity between the results of ZM-like models and those with exponential couplings is noted. Sensibility of the results to the specific choice of the values for the binding energy and saturation density is pointed out. Finally, we discuss the very intense scalar condensates found in the interior of neutron stars which may lead to negative effective masses. ",
        "introduction": "Since the dis\\-co\\-ve\\-ry of the first pul\\-sar in 1968 \\cite{hewish} and its i\\-den\\-ti\\-fi\\-ca\\-tion as a ro\\-ta\\-ting neu\\-tron star, the struc\\-tu\\-re, com\\-po\\-si\\-tion, dy\\-na\\-mics and e\\-vo\\-lu\\-tion of these as\\-tro\\-phy\\-si\\-cal o\\-bjects be\\-ca\\-me im\\-por\\-tant the\\-mes of theo\\-re\\-ti\\-cal and phe\\-no\\-me\\-no\\-lo\\-gi\\-cal re\\-search. According to an early suggestion \\cite{baade}, neutron stars evolve from an initially hot protoneutron star that forms in the collapse of a massive star in the supernova phenomenon. At densities exceeding that of nuclear matter, important static properties of a neutron star as the mass-radius relation, the crust extent, the distribution of the stellar moments of inertia and the central density, may be determined by its equation of state (EOS) \\cite{shapiro}.     During this period, there has been a continuous enhancement concerning relativistic microscopic calculations of the EOS of neutron stars, improving our understanding of the structure of these stellar objects. In particular, any theory must at least account for a neutron star as massive as the most massive observed pulsar; the knowledge of pulsar masses provides a very important constraint on the theory.  More recent calculations based on relativistic properties of nuclear matter at high densities indicate that the equations of state are considerable stiffer than those predicted by non-relativistic approaches. As a result, the mass of a neutron star is believed to be at least as large as $1.6-2.1M_{\\odot}$.   From the theoretical point of view, quantum chromodynamics (QCD) represents the most profound description of the strong interaction and would be the ideal tool for neutron star applications. However, the highly non-linear behavior of QCD at the hadronic energy scales inhibits any practical calculations leading most theorists to search for phenomenological descriptions of the structure of nuclear matter. One of these alternative approachs is quantum hadrodynamics (QHD) \\cite{walecka 86}, a relativistic quantum field theory based on a local Lagrangian density which uses ba\\-ry\\-on and me\\-son fields as the re\\-le\\-vant degrees of freedom. This model provides consistent theoretical framework for describing such a relativistic interacting many-body system and, based on it, Glendenning presented \\cite{glendenning 85} a very comprehensive treatment of the matter in neutron stars using an extended version which included leptons and the fundamental baryonic octet.  Alternative versions  of the Walecka model, namely the Boguta-Bodmer (BB)  \\cite{boguta} and the Zimanyi-Moszkowski (ZM) models \\cite{zimanyi,delfino}, were developed to improve the description of the nucleon effective mass, $M^\\star$, (too low) and compression modulus of nuclear matter, $K$, (too high) as attained with the original approach. Boguta and Bodmer introduced cubic and quartic scalar self-interactions in the Lagrangian while the ZM models have heightened the delineation of these quantities by replacing the Yukawa scalar coupling term by a {\\em derivative coupling} contribution. This derivative coupling may be interpreted alternatively as a phenomenological coupling between the scalar neutral meson and the nucleon fields through the introduction of a baryon density dependence in the scalar, vector and isovector coupling constants of the theory \\cite{Chiapparini}.  In this work we analyze the structure of neutron stars by introducing a QHD Lagrangian with a parameterized meson coupling contribution \\cite{taurines}. This phenomenological approach contains high order self-coupling contributions of the meson fields and permits in particular to restore the results obtained with Walecka, ZM  and ZM3 models by making suitable choices of the values of the mathematical parameters of the theory. The control on the analytical form of the couplings allows us to investigate other values of these parameters which give new  physical results. By extending the formalism to include hyperons and leptons, we investigate several static bulk properties of neutron stars using the Walecka, ZM and ZM3 models, the two later being applied to this problem for the first time. As we have a class of models we are able to relate nuclear matter and neutron star properties. In particular, we have found that some of the studied models describe very strong scalar fields in the interior of the neutron stars, leading to a negative nucleon effective mass. ",
        "conclusions": "We have done an analysis on the influence of nuclear matter properties on the structure of neutron stars using a new class of RMF models with parameterized couplings among mesons and baryons. These couplings allow to reproduce, trough a suitable choice of mathematical parameters ($\\lambda,\\beta,\\gamma$), results of the Walecka and derivative coupling models such as ZM and ZM3 models. As we have shown above, this new class of relativistic models permits some control on the intensity of the scalar, vector and, isovector mesons mean-field potentials. In particular, we have scanned values of nuclear matter properties which correspond to intermediate region of predictions of the Walecka, ZM  and ZM3 models.  In this new class of models we considered two cases, the Scalar (case S) with variations of $\\lambda$ and $\\beta = \\gamma = 0$ ( this case contains the results of the Walecka and  ZM models) and the Scalar-Vector (case S-V) with variations of $\\lambda$ and $\\beta = \\gamma  = \\lambda$ (ZM3 model belongs to this category). Taking the limit $\\lambda\\rightarrow\\infty$ for the S case, we have a model with exponential couplings (EXP/S model). Similarly, for $\\lambda = \\beta = \\gamma \\rightarrow\\infty$ we have the asymptotic model EXP/S-V.   An extended RMF which includes hyperon and lepton degrees of freedom was presented. For the sake of simplicity, we have assumed an universal coupling for the meson-baryon couplings. We have investigated the effects of the non-linear coupling in the nucleon effective mass and compression modulus of nuclear matter and in neutron star properties.  Having this in mind let us summarize the main findings of our work:  $\\bullet$ For the nuclear matter properties we have found values of parameters that reproduce reasonably the best range of phenomenological results, namely $0.6 < M^*/M < 0.7$ and $200 < K < 300 $ MeV. We have also found  that to higher values of the nucleon effective mass it corresponds lower values of the compression modulus. The Scalar case leads to a narrow interval of values of $\\lambda$ with different descriptions of nuclear matter properties  while the Scalar-Vector case has a broader range. To be more specific, case S gives reasonable results in the range $\\lambda \\sim 0.05 - 0.07$ and case S-V better results for $\\lambda \\sim 0.16 - 0.4$. We have also pointed out the strong similarity between the results of ZM-like models and those with exponential couplings. We have also obtained  an analytical expression for the compression modulus of nuclear matter as a function of the nucleon effective mass (Appendix \\ref{appendixb}).  $\\bullet$ We have pre\\-sen\\-ted ex\\-pli\\-ci\\-tly the neu\\-tron star phe\\-no\\-me\\-no\\-lo\\-gy for the Wa\\-le\\-cka, ZM and ZM3 models. To the best of our knowledge, we display in this work, for the first time, the results of these two later models. The ZM model predicts a maximum mass of approximately $1.6 M_\\odot$ for a neutron star, while the ZM3, being too soft, leads to  $\\sim0.72 M_\\odot$ as the limiting mass. We have obtained Walecka and ZM fermionic distributions. The corresponding particle populations exhibit expressive differences essentially due to the strength of the scalar potential in these two models.  $\\bullet$ Sensibility of this approach to the specific choice of $B$ and $\\rho_0$ was noticed with differences of the order of a half solar mass.  For the Walecka model, the neutron star mass with hyperons has varied from $2.77M_\\odot$ to $2.33M_\\odot$ with a very small change in those nuclear matter bulk properties. Thus, extrapolation for neutron star densities from the fitting of $B$ and $\\rho_0$ at saturation needs more precision on the choice of the values for these quantities.  $\\bullet$  We analyzed our new class of models allowing the variation of the $\\lambda$ parameter and  showed results for the radius, redshift (z) and hyperon/baryon ratio for the maximum mass of a neutron star sequence, considering hyperon, nucleon and lepton degrees of freedom. In figure \\ref{4mstlam} we showed the dependence of the maximum mass with this parameter. We found that some models corresponding to the S-V case (including the ZM3 model) predicted very small neutron star masses. We observed again the similarity between  the predictions of neutron star phenomenology associated to the ZM and ZM3 models compared to those with exponential couplings, EXP/S and EXP/S-V.   $\\bullet$ The dependence of the maximum neutron star mass of a sequence with the nuclear matter bulk properties $K$ and $M^\\star$ at saturation density was also analyzed. In both S and S-V cases, to a higher maximum neutron star mass it corresponds, in general,  a less compressible matter (higher $K$). We also concluded that the maximum neutron star mass decreases with the nucleon effective mass $M^\\star$.   $\\bullet$ Finally, we found that for $\\lambda<0.5$ in the S-V case, we have obtained negative values for the nucleon effective mass, which corresponds to a density region for which the strength of the  scalar condensate exceeds the free mass of the nucleon, $| S| > M$. We have shown how the differences in the baryonic bare masses explain this result; as explained in the text, in practice the inclusion of hyperons is responsible for this behaviour. These results indicate the existence of very intense scalar condensates in the interior region of neutron stars. Concerning  mathematics, the theory continues to work due to the symmetry of negative and positive values of $M^\\star$ in the Fermi integrals."
    },
    "0010/astro-ph0010115_arXiv.txt": {
        "abstract": "The unprecedented quality of recent colour-magnitude diagrams of resolved stellar populations in nearby galaxies requires state-of-the-art techniques to infer the star formation histories which gave rise to the observed distributions. We have developed a maximum likelihood technique, which coupled to a variational calculus allows us to make a robust, non-parametric reconstruction of the evolution of the star formation rate. A full Bayesian analysis is also applied to assess whether the best solutions found are also good fits to the data. Applying this new method to WFPC2 observations of five dSph galaxies of the Local Group, we find a wide variety of star formation histories, with no particular epoch being dominant. In the case of the solar neighbourhood observed by Hipparcos  we infer, with an unprecedented resolution of 50 Myr, its star formation history over the past 3 Gyr. The surprising regularity of star formation episodes separated by some 0.5 Gyr could be interpreted as the result of interactions with two spiral arms or the Galactic bar. These bursts possibly trigger the formation of massive star clusters which slowly dissolve into the galactic field. ",
        "introduction": "\\label{sec_intro} The history of the star formation rate in galaxies is one of the main ingredients required to understand their formation and evolution. Whilst the Roberts' time scale for gas consumption by star formation gives a rough idea of the evolution of the stellar and gas content, more elaborate models (as summarised, e.g. by Sandage, 1986) can reproduce a wide variety of observables, from the evolution of the bulge-to-disc ratio in the Hubble sequence to the integrated colours as a function of redshift. However, in sharp contrast with these global views where the star formation rate SFR(t) is a monotone function of time, current bottom-up hierarchical scenarios of structure formation predict far more complicated histories, where star formation episodes are related to, if not driven by, halo mergers. The star formation history is, in this context, a more or less faithful reproduction of the merger history, rich in both strong and minor events at redshifts $z < 2$ or so.  While it is possible to understand some of the observed properties of high redshift galaxies in this framework, a far more direct test can be achieved with the fossil record of the star formation history itself: the colour-magnitude diagrams (CMDs) of the resolved stellar populations in nearby galaxies. Progress in this field has been steady, driven both by ground-based wide-field surveys and by deep, high-resolution, narrow-field HST studies, reaching well below the main sequence turn off point in many satellites of the Local Group (see Aparicio 1998, Mateo 1998 for recent reviews). ",
        "conclusions": "We put forward a maximum likelihood technique, coupled to a variational calculus, which allows the robust, non-parametric reconstruction of the evolution of the star formation rate from the information contained in colour-magnitude diagrams. A full Bayesian analysis is also applied to assess whether the best solutions found are also good fits to the data. Its main limitation, at the moment, is the prior knowledge of the metallicity of the ensemble of stars in a CMD.  For the first time, an objective reconstruction of star formation histories without any a priori or model-dependent information is applied to an homogeneous sample of dwarf spheroidals of the Local Group. We find a wide variety of SFHs, with bursts of activity uncorrelated to any special epoch or event, like perigalacticon passages. Among many other things, this also implies that late accretion was not important in the formation of the Galactic halo (Gilmore et al. 2000).  In the solar neighbourhood observed by Hipparcos,  we infer --with an unprecedented resolution of 50 Myr-- its star formation history over the past 3 Gyr, finding a surprising regularity of star formation episodes separated by some 0.5 Gyr. A possible explanation  is that the solar neighbourhood interacted with two spiral arms or the Galactic bar,  triggering star formation at each interaction. These bursts are likely to induce the formation of massive star clusters which slowly dissolve into the galactic disc."
    },
    "0010/astro-ph0010323_arXiv.txt": {
        "abstract": "Recent analysis of the helioseismic observations indicate that the previously observed surface torsional oscillations extend significantly  downwards into the solar convection zone.  In an attempt to understand these oscillations, we study the nonlinear coupling between the magnetic field and the solar differential rotation in the context of a mean field dynamo model, in which the nonlinearity is due to the action of the azimuthal component of the Lorentz force of the dynamo generated magnetic field on the solar angular velocity. The underlying zero order angular velocity is chosen to be consistent with the most recent helioseismic data.  The model produces butterfly diagrams which are in qualitative agreement with the observations. It displays torsional oscillations that penetrate into the convection zone, and which with time  migrate towards the equator. The period of these oscillations is found to be half that of the period of the global magnetic fields. This is compatible with the observed period of the surface torsional oscillations. Inside the convection zone, this is a testable prediction that is not ruled out by the observations so far available. ",
        "introduction": "An important feature of the solar convection zone is the presence of differential rotation in the form of a decrease in angular velocity from equator to the pole. This has been observed both in the surface layers (e.g.\\ Snodgrass 1984) and deeper in the convection zone, as inferred from helioseismic measurements (e.g.\\ Thompson {\\em et al.} 1996). Furthermore, the differential rotation on the surface has been observed to vary with time (e.g.\\ Howard \\& LaBonte 1980; Snodgrass, Howard, \\& Webster 1985). These so called torsional oscillations, which have periods of about 11 years, manifest themselves in the form of four alternating latitudinal bands of slightly faster and slower than average zonal flows which migrate towards the equator in about 22 years. These oscillations have also been confirmed by the analysis of the helioseismic data from the Solar and Heliospheric Observatory (SOHO) spacecraft for the present solar cycle (Kosovichev \\& Schou 1997; Schou {\\em et al.} 1998).  Recent analysis of the helioseismic data, from both the Michelson Doppler Imager (MDI) instrument on board the SOHO spacecraft and the Global Oscillation Network Group (GONG) project, has also produced evidence that this banding signature is not merely a surface feature, but extends into the convection zone, to a depth of at least 8 percent in radius (Howe {\\em et al.} 2000). These authors present data on departures of the reconstructed rotation rate from its temporal averages - the residuals - as a function of latitude at several target depths, which behave in a manner similar to the migration of sunspots during the solar cycle (the `butterfly diagram'). This finding is also supported by the analysis of Antia \\& Basu (2000), who use different data sets from GONG and independent inversion techniques. The time-base of these observations is only a few years, less than a complete solar cycle.  These torsional oscillations are thought to be produced as a consequence of the nonlinear interactions between the magnetic fields and the solar differential rotation. A zero order `mean' differential rotation is assumed to be maintained by the Reynolds stresses of the turbulence; this can be included as a constant `background' angular velocity, or explicitly parametrized, e.g.\\ by the so called $\\Lambda$--effect (e.g.\\ R\\\"udiger 1989). Attempts have been made to explain the surface oscillations in terms of the effects of the Lorentz force exerted by the large scale magnetic field on the azimuthal velocity field (e.g.\\ Brandenburg \\& Tuominen 1988), or as a consequence of the `quenching' of the turbulence-dependent quantities by the magnetic field (Kitchatinov {\\em et al.} 1999).  Here we study this nonlinear coupling in the context of a two dimensional axisymmetric mean field dynamo model, in a spherical shell, in which the only nonlinearity  is the action of the azimuthal component of the Lorentz force of the dynamo generated magnetic field on the solar angular velocity. Obtaining torsional oscillations in this way is also of interest in view of the fact that the Lorentz force is of second order in the magnetic field, thus naturally leading to the excitation of hydrodynamical oscillations of about half the period of the magnetic oscillations.  In the next section we introduce our model. Section 3 contains our results and section 4 gives a brief discussion. ",
        "conclusions": "We have studied a solar dynamo model, calibrated to have the correct cycle period, with a mean rotation law given by recent helioseismic observations. This model produces butterfly diagrams in qualitative agreements with those observed.  We have shown that a nonlinear coupling between the magnetic field and the solar differential rotation, where the nonlinearity is due to the action of the azimuthal component of the Lorentz force, is capable of producing torsional oscillations, with period of about 11 years, which penetrate into the convection zone and which migrate towards the equator in about 22 years. The period of these oscillations is about half that of the period of the global magnetic fields. This is in agreement with the observed period of the torsional oscillations at the surface. For the oscillations inside the convection zone, this is a testable prediction, not contradicted by the current helioseismic observations, which so far extend over an interval less than a solar cycle. We note also that solutions with even parity ($P =+1$) show slightly larger amplitudes without the equatorial feature.  The current inversions of the helioseismological data seem to suggest that the torsional oscillations largely disappear below about $R =0.9 R_{\\odot}$, in contrast to our model oscillations. However, there are uncertainties in these inversions, specially at the deeper levels. At the same time our dynamical model is oversimplified and substantial improvements can be made. In particular, the predicted amplitudes for torsional oscillations in our model are likely to be affected by our assumption of uniform density in Eq.\\ (\\ref{NS}). Nevertheless, we find it interesting that some of the major features in the torsional oscillations can be readily reproduced. A more detailed study of these oscillations, including more realistic density profiles, is in progress."
    },
    "0010/astro-ph0010609_arXiv.txt": {
        "abstract": "We have obtained spectral and timing data on GX 3+1 and Ser X-1. Both sources were observed simultaneously with \\sax\\ and RXTE. The RXTE data is used to provide power spectra and colour-colour diagrams in order to constrain the state (and thus track $\\dot M$) the sources are in. The \\sax\\ data provide the broad-band spectra. The spectra of both sources are reasonably well-fit using a model consisting of a disk-blackbody, a comptonized component and a Fe line, absorbed by interstellar absorption. The electron temperature (kT$_{\\rm e}$) of the Comptonizing plasma is in both cases $\\sim$2.5 keV. This implies that no strong high-energy tail from the Comptonized component is present in either of the sources. We discuss the similarities between these burst sources and the luminous X-ray sources located in globular clusters.  We find that the spectral parameters of the comptonized component provide information about the mass-accretion rate, which agrees well with estimates from the timing and spectral variations. ",
        "introduction": "\\label{sect:intro} The sources GX 3+1 and Ser X-1 are both bright X-ray burst sources. Ser X-1 was discovered in 1965 (Friedmann et al.\\ 1967), and X-ray bursts from this source were first detected by Swank et al.\\ (1976). GX 3+1 was discovered by Bradt et al.\\ (1968), and the first X-ray bursts were detected by Makishima et al.\\ (\\cite{m:83}).  These sources, containing neutron stars, are studied with \\sax\\ in order to quantify their spectrum in the $\\sim$0.1--30 keV energy range. It is now well established that hard X-ray emission (above $\\sim$30 keV) is not an unique feature of black hole systems as indicated by detections of several X-ray binaries at high ($\\sim$100 keV) energies (e.g., Barret \\& Vedrenne 1994). Following the detections of neutron star systems at high energies van Paradijs \\& van der Klis (\\cite{vpvdk:94}) concluded from data collected from the HEAO 1 A-4 catalogue that the 10--80 keV spectra of accreting neutron stars with low magnetic fields get progressively harder when the source gets dimmer.  Historically observations with Ginga and nearly simultaneous with BATSE of 4U 1608-522 (Yoshida et al. \\cite{y:93}, Zhang et al. \\cite{z:96}) have shown hard spectra from a neutron star when it is in a low state.  A recent example of a neutron star showing a high energy-tail in its spectrum is 4U 0614+091 (Piraino et al.\\ \\cite{psfk:99})  However, quantative differences might exist between the two types of compact objects. For example Churazov et al.\\ (\\cite{c:95}) suggest that the power-law components of black-hole candidates are systematically harder than those of low-luminosity neutron star accretors.  The two sources are observed in a program aimed at determining and quantifying the spectral properties in the 0.1--30 keV band. Results on data obtained with RXTE on four different X-ray bursters have been reported by Barret et al.\\ (2000). They find that the spectra of sources in a low state ($\\sim$0.05--0.1 L$_{\\rm Edd}$) can be described by thermal Comptonization with an electron temperature of 25--30 keV and (in general) an underlying soft component. When the sources are in a higher state ($\\sim$0.35 L$_{\\rm Edd}$) there is no significant hard tail, while the Comptonized component can be described by a much lower electron temperature, kT$_{\\rm e}$ ($\\sim$3 keV).  We use (quasi-)simultaneous RXTE data to constrain the state of the sources. According to Hasinger \\& van der Klis (\\cite{hk:89}) ``Atoll'' sources trace out a distinct pattern in a colour-colour diagram, while their power spectra show distinct changes when moving through these states. This will allow us to obtain an estimate of the immediate mass-accretion rate of these sources. Using the simultaneous data obtained by two different satellites allow us to use the complementary properties of both satellites. ",
        "conclusions": "\\label{sect:discussion}  We have observed GX 3+1 and Ser X-1 in the 0.1--30~keV energy range using \\sax.  We find that the spectra are best described by a combination of a {\\sc compTT} model and blackbody-like model (either a blackbody or a multi-colour blackbody model {\\sc diskBB}). From the RXTE timing and colour-colour diagram we infer that GX 3+1 is in the lower banana state, implying a $\\dot M$ of around 0.15 L$_{\\rm Edd}$, while Ser X-1 is in the upper banana state, implying a slightly higher mass accretion rate. The absence of kilohertz QPO is consistent with the fact that kilohertz QPO are not normally seen in the banana-type state in other sources (for a list of similar upper limits see van Straaten et al.\\ 2000).     Comparison of the spectra of the two sources discussed here with the spectra of various low-mass X-ray binaries (see Barret et al.\\ \\cite{b:00}), reveals a striking resemblance in terms of fit-parameters with KS1731-260. This is not suprising, since this source was also in the banana state during the RXTE observations. The obtained fit-parameters are very similar.  Comparing our two sources with another well-studied atoll-source like 4U\\,1820--30 we see that the spectral parameters are compatible when 4U\\,1820--30 is rather bright (e.g.\\ the observation reported in Piraino et al. \\cite{p:98}). Interestingly 4U\\,1820--30 is not detected at high ($\\sim$100 keV) X-ray energies (see e.g.\\ Bloser et al.\\ \\cite{bloser:00}), even at its lowest accretion rates.    The observations presented here provide additional evidence for a strong relation between the luminosity and the presence of a hard tail which can be modeled by thermal Comptonization with an electron temperature ranging from $\\sim$30 keV at the lowest accretion rates to $\\sim$ 2.5 keV at higher accretion rates ($\\sim$0.15 L$_{\\rm Edd}$). Thus the broad-band spectral properties of X-ray burst sources seem to be governed, to first order, by the mass accretion rate and appear to be very similar between sources.     The observed optical depth ($\\tau$), electron temperature and the diskblackbody spectral parameters are rather similar to X1820-303, the brightest source in the sample of globular cluster sources as discussed in Guainazzi et al.\\ (2000). This is consistent with the observed relation between luminosity and e.g. $\\tau$ (Guainazzi et al. 2000) and the fact that these sources are expected to accrete at a $\\sim$10--30\\% of the Eddington limit. Using the correlation between the observed $\\tau$ and accretion rate expressed in Eddington luminosity (Fig.\\ 4 in Guainazzi et al. 2000) we can estimate the Eddington luminosity of Ser X-1 and hence its distance. The observed $\\tau$ would put it a luminosity of $\\sim$0.15 L$_{\\rm Edd}$, which would place Ser X-1 at a distance of $\\sim$6 kpc.  A similar exercise, using the observed $\\tau$, would place GX 3+1 at a distance of $\\sim$5 kpc, in remarkable agreement with the distance of 4.5 kpc determined by Kuulkers \\& van der Klis (2000) using a unique radius-expansion burst from this source. However, this remarkable agreement may be fortuitous, since a more complete sample of globular cluster sources (Sidoli et al.\\ \\cite{s:00}) shows a less clear correlation between luminosity and $\\tau$ than in Guainazzi et al.\\ (2000).  Alternatively the observed $\\tau$ from the spectrum of GX 3+1 can be used to estimate the luminosity in terms of L$_{\\rm Edd}$, using the correlation in Guainazzi et al. (2000). We found a luminosity of $\\sim$0.1 L$_{\\rm Edd}$. This value, when the source is in the lower banana state is in line with current thoughts about the mass-accretion rate in this state (van der Klis \\cite{vdk:95}).  For Ser X-1 we infer a slightly higher $\\dot M$, since it probably is higher ``up'' the banana-curve, which is consistent with the inferred luminosity of $\\sim$0.15 L$_{\\rm Edd}$ from $\\tau$.  The energy dependence of the rms variability of the high-frequency noise (Fig.\\ \\ref{fig:hfn}) shows that the HFN is ``hard'', i.e.\\ the variability increases towards higher energies. This has been observed before in e.g.\\ 1E1742.5--3045 (Olive et al. \\cite{o:98}, in the island state), GX 3+1 (Makishima et al.\\ \\cite{m:89}), and 4U 1820--30 (Dotani et al.\\ \\cite{d:89}). The energy dependence is roughly consistent with a $\\sim$8\\% variability in the Comptonized component, which, at lower energies, progressively gets diluted by the soft component (the (disk-)blackbody). However the spectral decomposition is not unique and the statistical quality of the rms variability as a function of energy is not good enough to associate the variability uniquely with one spectral component. e.g.\\ it is conceivable that (an energy-dependent) part of the disk-blackbody component also contributes to the variability.   It is interesting to interpret the spectrum in terms of the boundary layer model (e.g.\\ Popham \\& Sunyaev \\cite{ps:00}; earlier work by e.g.\\ Kluzniak \\& Wilson \\cite{k:91}). In this model the dominant source of high-energy photons is expected to be the boundary layer where the accretion disk meets the (weakly magnetized) neutron star. Roughly speaking this model predicts a Comptonized component together with a thermal component. Our results (and those of e.g.\\ Barret et al.\\ \\cite{b:00}) are in global agreement with the predictions of this model. Also the fact that the observed variability of the high-frequency noise is consistent with most of the variability originating in the Comptonized component, argues in favour of the model.   The observed Fe line energy is consistent with neutral or moderately ionized iron.  However, care should be taken when interpreting the energy of the Fe line, since the addition of an edge significantly shifts the energy (to a value consistent with higly ionized Fe i.e.\\ 6.7 keV). Following the discussion in Barret et al.\\ (\\cite{b:00}) the most likely location of the material emitting the Fe line is in the accretion disk.  A strong Fe line should then be accompanied by a Compton reflected spectrum. It is indeed the case that a fit with a reflected component gives a better fit (although it is not significantly better than the disk blackbody + compTT fit) to the Ser X-1 spectrum (where the Fe line is strong, $\\sim$275 eV).  However, the quality of our data does not allow us to draw any definitive conclusions about the presence or absence of reflection.  The obtained width ($\\sigma$) of the Fe line is significantly higher than 0. This might be caused by Comptonization of the Fe line photons, or by other effects which cause broadening of the Fe line.  Summarizing we have observed two burst sources (GX 3+1 and Ser X-1) with similar spectral parameters (temperature of Comptonizing plasma $\\sim$2.5 keV with a large (6--10) optical depth). This is in contrast with low state spectra of burst sources which have a much higher temperature ($\\sim$30 keV) and a smaller optical depth."
    },
    "0010/astro-ph0010053_arXiv.txt": {
        "abstract": " ",
        "introduction": "The recent identification of several groups of young stars within 100 parsecs of the Sun has generated widespread interest. Given their proximity and possible age differences, these systems are ideally suited for detailed studies of disk evolution and planet formation. Here I discuss recent results and prospects for the near-future. ",
        "conclusions": ""
    },
    "0010/astro-ph0010579_arXiv.txt": {
        "abstract": "In this paper, we present results from a simulation of stellar core collapse, bounce, and postbounce evolution with Boltzmann neutrino transport. We motivate the development of our Boltzmann solver in light of the sensitivity of the neutrino-heating core collapse supernova paradigm to details in the neutrino transport, particularly near the neutrinospheres, where the neutrinos are neither diffusing nor free streaming and a kinetic description is necessary, and in light of the mixed outcomes and transport approximations used in all prior supernova models in both one and two dimensions. We discuss the implications of our findings for the supernova mechanism and future supernova research. We also present the results of a Boltzmann transport prediction of the early neutrino light curves in the model included here. ",
        "introduction": "Beginning with the first numerical simulations conducted by Colgate and White\\cite{cw66}, three decades of supernova modeling have established a basic supernova paradigm. The supernova shock wave---formed when the iron core of a massive star collapses gravitationally and rebounds as the core matter exceeds nuclear densities and becomes incompressible---stalls in the iron core as a result of enervating losses to nuclear dissociation and neutrinos. The failure of this ``prompt'' supernova mechanism sets the stage for a ``delayed'' mechanism, whereby the shock is reenergized by the intense neutrino flux emerging from the neutrinospheres carrying off the binding energy of the proto-neutron star\\cite{w85,bw85}. The heating is mediated primarily by the absorption of electron neutrinos and antineutrinos on the dissociation-liberated nucleons behind the shock. This past decade has also seen the emergence of multidimensional supernova models, which have investigated the role convection, rotation, and magnetic fields may play in the explosion \\cite{hbhfc94,bhf95,jm96,mcbbgsu98a,mcbbgsu98b,fh99,khowc99}, in some cases invoking new explosion paradigms.  Although a plausible framework is now in place, fundamental questions about the explosion mechanism remain: Is the neutrino heating sufficient, or are multidimensional effects such as convection and rotation necessary? Can the basic supernova observable, explosion, be reproduced by detailed spherically symmetric models, or are multidimensional models required? Without a doubt, core collapse supernovae are not spherically symmetric. For example, neutron star kicks\\cite{fbb98} and the polarization of supernova emitted light\\cite{w99} cannot arise in spherical symmetry. Nonetheless, ascertaining the explosion mechanism and understanding every explosion observable are two different goals. To achieve both, simulations in one, two, and three dimensions must be coordinated. ",
        "conclusions": ""
    },
    "0010/astro-ph0010170.txt": {
        "abstract": "Quel est, ou pourrait \\^etre, la topologie globale de la partie spatiale de l'Univers ? L'Univers entier (pr\\'ecis\\'ement, l'hypersurface spatiale de celui-ci) est-il observable ?  Les math\\'ematiciens, les physiciens et les cosmologistes observationnels ont des approches diff\\'erentes pour aborder ces questions qui restent ouvertes.  {\\em What is, or could be, the global topology of spatial sections of the Universe? Is the entire Universe (spatial hypersurface thereof) observable ? Mathematicians, physicists and observational cosmologists have different strategies to approaching these questions which are not yet fully answered.}  In this paper, we classify all closed flat 4-manifolds that have a reflective symmetry along a separating totally geodesic hypersurface. We also give examples of small volume hyperbolic 4-manifolds that have a reflective symmetry along a separating totally geodesic hypersurface. Our examples are constructed by gluing together polytopes in hyperbolic 4-space.  If  the topology of space is multi-connected, rather than simply connected as it is most often assumed, this would cause  a major revolution in cosmology, and  a huge progress in the knowledge of our world (see the review paper by   Lachi\\`eze-Rey \\& Luminet, 1995, hereafter LaLu). This  would  set  new constraints and ask new questions  on the physics of the primordial universe.  Why space is multi-connected~? What has determined its  principal directions    and the values of its spatial dimensions~? The links between  cosmology and  quantum physics would be modified, in particular    the question  of the vacuum energy and of the cosmological constant.  All methods of constraining or detecting candidates for the global topology of the Universe share the same common principle: an object or a region of space should be observed several times at, in general, different angular positions and redshifts. In practice, whether using the cosmic microwave background (CMB) or collapsed astrophysical objects, the practical details of how an object or a region of space emits electromagnetic radiation (degree of local isotropy, evolution) imply that different strategies have to be adopted, depending on which ``standard candles'' are used.  Contrary to popular opinion, it should be noted that the claimed CMB ``constraints'' against small flat multiply connected models are weak: these are statements about the rarity of simulated perturbation statistical properties required to match COBE data, rather than about the consistency of the data with multiple topological imaging on the surface of last scattering.  \\noindent One of the possibilities to constrain the topology of the Universe to the observational data is to search for topological images of our own Galaxy. This method is based on the idea that in a multi-connected Universe we would, in principle, be able to see the light emitted by our own Galaxy in the early stages of its evolution. The significant identification of these images would give us strong evidence that the topology of the Universe may be non-trivial.  Although it has been argued that ``small universes'' being multiply connected on scales smaller than the particle horizon are ruled out, it is found that constraints on compact multiply connected models with low matter-density (flat or hyperbolic) are not stringent. Furthermore, compact hyperbolic models ($\\Omega_0=0.1\\sim0.2$) with volume comparable to the cube of the present curvature radius are much favored compared to the infinite counterparts due to the mild suppression on large-angle power.  We provide an informal discussion of pattern formation in a finite universe. The global size and shape of the universe is revealed in the pattern of hot and cold spots in the cosmic microwave background. Topological pattern formation can be used to reconstruct the geometry of space, just as gravitational lensing is used to reconstruct the geometry of a lens. ",
        "introduction": " ",
        "conclusions": ""
    },
    "0010/astro-ph0010265_arXiv.txt": {
        "abstract": "The CELESTE experiment uses the heliostats of an old solar farm in the French Pyrenees to detect gamma ray air showers by the atmospheric Cerenkov technique. CELESTE has been operating since November 1999 with an array of 40 heliostats fully instrumented with 1GHz flash ADCs. Significant advances have been made in the detector simulations and in the data analysis techniques. We report here on results from recent observations of the Crab nebula above an energy threshold of 50GeV. The results and simulations illustrate the current sensitivity of the experiment. ",
        "introduction": " ",
        "conclusions": ""
    },
    "0010/astro-ph0010522_arXiv.txt": {
        "abstract": "We present high-resolution near-infrared imaging obtained using adaptive optics and {\\it HST}/NICMOS, and ground-based spectroscopy of the hot-spot galaxy NGC~2903. Our near-infrared resolution imaging enables us to resolve the infrared hot spots into individual young stellar clusters or groups of these. The spatial distribution of the stellar clusters is not coincident with that of the bright H\\,{\\sc ii} regions, as revealed by the {\\it HST}/NICMOS Pa$\\alpha$ image. Overall, the circumnuclear star formation in NGC~2903 shows a ring-like morphology with an approximate diameter of 625\\,pc.  The SF properties of the stellar clusters and H\\,{\\sc ii} regions have been studied using the photometric and spectroscopic information in conjunction with evolutionary synthesis models. The population of bright stellar clusters shows a very narrow range of ages, $4-7 \\times 10^6\\,$yr after the peak of star formation, or absolute ages $6.5-9.5 \\times 10^6\\,$yr (for the assumed short-duration Gaussian bursts), and luminosities similar to the clusters found in the Antennae interacting galaxy. This population of young stellar clusters accounts for some $7-12\\%$ of the total stellar mass in the central 625\\,pc of NGC~2903. The H\\,{\\sc ii} regions in the ring of star formation have luminosities close to that of the super-giant H\\,{\\sc ii} region 30 Doradus, they are younger than the stellar clusters, and will probably evolve into  bright infrared stellar clusters similar to those observed today. We find that the star formation efficiency in the central regions of NGC~2903 is higher than in normal galaxies,  approaching the lower end of infrared luminous galaxies. ",
        "introduction": "Numerical simulations of  bars have shown that they may provide an efficient mechanism for transporting interstellar gas into the central regions of galaxies. Thus, bars are often invoked as a possible mechanism to fuel nuclear/circumnuclear star formation (SF; Heller \\& Shlosman 1994), as well as active galactic nuclei (Shlosman, Frank, \\& Begelman 1989). It is now well established that circumnuclear regions (CNRs) of enhanced SF are commonly associated with the presence of stellar bars (see the recent review by Knapen 1999, and references therein). In particular, a significant percentage of barred galaxies show nuclear rings within 1\\,kpc from the nucleus of the galaxy; these are believed to be formed by gas accumulation near galactic resonances. In these nuclear rings, strong density enhancements of gas and increased SF are often observed (Piner, Stone, \\& Teuben 1995; Knapen et al. 1995b; Buta \\& Combes 1996; Shlosman 1999).  Knapen (1999) pointed out that CNRs are excellent laboratories for the study of the effects of inflow processes on the triggering of SF. A detailed study of CNRs has been hampered by the lack of imaging at sufficiently high spatial resolution. The main difficulty arises when trying to disentangle the effects of young and old stars, cold gas and dust, and possibly emission from dust heated by the SF activity (e.g. Knapen et al. 1995a, b). High resolution imaging allows one to resolve individual SF sites, to determine their location with respect to dust lanes, and to discover dynamically important morphological features that remain hidden in imaging of CNRs at lower resolution. These features include nested bars (e.g. Knapen et al. 1995a; Phillips et al 1996; Elmegreen et al. 1998; Carollo, Stiavelli, \\& Mack 1998; Colina \\& Wada 2000), leading spiral arms (Knapen et al. 1995a, 2000); nuclear flocculent (e.g., Elmegreen et al. 1999) or even grand-design spirals (Laine et al. 1999).   High resolution near-infrared  (NIR) observations of CNR are of particular importance because they are less affected by extinction, and because the NIR emission is a better tracer of old stellar population (and hence of the stellar mass distribution) than optical or ultraviolet observations. However, in sites of strong SF there can be a significant contribution from young red supergiants to the NIR emission, and therefore both NIR spectroscopy and imaging are an absolute need to study the SF processes in CNR  (e.g., Engelbracht et al. 1998; Ryder \\& Knapen 1999; Puxley \\& Brand 1999; Alonso-Herrero et al. 2000; 2001). Quantifying the contribution of young stars to the NIR emission so the emission from old stars can be isolated is essential for the interpretation of the dynamics of CNRs and of inflow in barred galaxies in general, through the use of dynamical modeling. Such modeling is generally based upon an estimate of the galactic potential as derived from red or NIR imaging, under the assumption of a constant mass-to-light ratio (e.g., Quillen, Frogel, \\& Gonz\\'alez 1994; Knapen et al. 1995a). It is important that this assumption be tested, and the circumstances under which the mass-to-light ratio may vary be understood, to allow for more accurate modelling.  In this paper we present high angular resolution  NIR adaptive optics (AO) and {\\it HST}/NICMOS imaging, and  spectroscopy of the hot-spot galaxy NGC~2903. Although NGC~2903 is classified as an SAB(rs)bc (de Vaucouleurs et al. 1991), the presence of a large scale bar with a position angle (P.A.) of $\\simeq 20\\deg$ \\ is clearly inferred from $K$-band (Regan \\& Elmegreen 1997) and H$\\alpha$ (Jackson et al. 1991; P\\'erez-Ram\\'{\\i}rez \\& Knapen 2000) images, as well as from CO observations (Regan, Sheth, \\& Vogel 1999). Early NIR  studies of this galaxy have shown that considerable SF is occurring within the complex hot spot morphology in the nuclear region (e.g., Wynn-Williams \\& Becklin 1985; Simons et al. 1988). However these studies lacked the necessary spatial resolution to determine whether the SF was confined to these hot spots or not. The infrared luminosity ($8-1000\\,\\mu$m) of NGC~2903 is $9.1 \\times 10^9\\,{\\rm L}_\\odot$ (Sakamoto et al. 1999) which is similar to (although slightly smaller than) those of M82 and NGC~253, both also well known hot-spot galaxies. Throughout the paper we will assume a distance to NGC~2903 of $d = 8.6\\,$Mpc (Telesco \\& Harper 1980), for which $1\\,{\\rm arcsec} \\simeq 42\\,$pc. This paper is organized as follows. Section~2 describes the observations. Section~3 discusses the morphology, and Section~4 the extinction. The properties of the stellar clusters and H\\,{\\sc ii} regions are studied in Sections~5 and 6, while conclusions are given in Section~7. ",
        "conclusions": "We have presented  high resolution NIR imaging and spectroscopy of the nuclear region of NGC~2903 and conducted a detailed study of its morphology and SF properties. Our main conclusions are:  \\begin{itemize} \\item The bright infrared hot spots (known from ground-based observations) are resolved into individual stellar clusters or groups of these in our AO and {\\it HST}/NICMOS images.  \\item The {\\it HST}/NICMOS Pa$\\alpha$ (at $1.87\\,\\mu$m) image reveals a ring of star formation (diameter $\\simeq 625\\,$pc) with bright H\\,{\\sc ii} regions. The overall morphology of the Pa$\\alpha$ emission is  similar to the 2\\,cm and 20\\,cm radio morphologies, which seems to indicate a significant contribution of thermal emission to the total radio emission.  \\item The high resolution images show a complex distribution of the extinction in the CNR of this galaxy. However the derived values of the extinction are relatively modest (with values of up to $A_V = 9.5\\,$mag or infrared extinction at the $H$-band of up to $A_H = 1.6\\,$mag for a foreground dust screen model).  \\item We combined spectroscopic and photometric information available for the stellar clusters with evolutionary synthesis models to constrain their SF properties.  Short-lived bursts of SF (Gaussian bursts with ${\\rm FWHM} = 1, \\, 5\\,$Myr) produced a good fit to the data. The stellar clusters have luminosities similar to the young stellar clusters in the Antennae galaxy. They are relatively young, with  a narrow range of ages, $4-7\\times 10^6\\,$yr after the peak of SF  (the absolute ages depend upon the specific duration of the burst of SF), and contribute to some $7-12\\%$ of the stellar mass in the central regions of NGC~2903.  \\item The H\\,{\\sc ii} regions in the star forming ring exhibit luminosities approaching that of 30 Doradus, and are younger ($1-4 \\times 10^6\\,$yr after the peak of SF) than the stellar clusters.   \\item The ages and masses derived for the bright stellar clusters and H\\,{\\sc ii} regions suggest that the H\\,{\\sc ii} regions are the progenitors of the bright infrared clusters. This may offer an explanation for the lack of a detailed spatial correspondence between the clusters and the H\\,{\\sc ii} regions.   \\item Finally, we show that the SF efficiency in the central kpc of NGC~2903 is more elevated than in normal galaxies, similar to (although slightly smaller than) those for infrared luminous galaxies.  \\end{itemize}"
    },
    "0010/astro-ph0010008_arXiv.txt": {
        "abstract": "s{The first detections of cosmic shear signal reported recently by 4 independent groups cover angular scales between one and 10 arcmin. On those scales, the cosmic shear is a signature of non-linear perturbations, like groups and clusters of galaxies. I present the results obtained by our team on CFHT and on the VLT and discuss its impact for the analysis of cluster abundances and cosmology} ",
        "introduction": "The effects of gravitational lensing induced by large-scale structures of the universe accumulate along the lines of sight and manifest as a weak distortion on lensed galaxies (the cosmic shear). Theoretical predictions done over the last decade reveal that the statistical properties of cosmic shear contain invaluable clues on the cosmological parameters, the power spectrum of density fluctuation and the biasing. The cosmic shear is therefore an interesting way to probe the properties of the universe on large-scale.  \\\\ The amplitude of the weak cosmological distortion only reaches  few percents and is contaminated by artificial distortions of similar amplitude making its measurement a challenging task.  In contrast to arc(let)s analysis, its observational signature can only be recovered statistically from a large sample of galaxies covering several degrees of the sky.  Despite difficulties, four teams challenged in order to measure the cosmic shear and get very first constraints on cosmological models from this new technique. \\\\ Regarding the topic of this conference, the cosmic shear looks somewhat marginal.  However, on small scales the cosmic shear signal is dominated by distortion produced by non-linear gravitational systems, like groups and clusters of galaxies. It is therefore interesting to present and discuss these important results in a broader context including cluster formation. ",
        "conclusions": ""
    },
    "0010/astro-ph0010652_arXiv.txt": {
        "abstract": "We study the spectrum and average mass composition of cosmic rays with primary energies between $10^{17}$eV and $10^{18}$eV using a hybrid detector consisting of the High Resolution Fly's Eye (HiRes) prototype and the MIA muon array.  Measurements have been made of the change in the depth of shower maximum as a function of energy. A complete Monte Carlo simulation of the detector response and comparisons with shower simulations leads to the conclusion that the cosmic ray intensity is changing from a heavier to a lighter composition in this energy range. The spectrum is consistent with earlier Fly's Eye measurements and supports the previously found steepening near $4\\times 10^{17}$eV . ",
        "introduction": "The source of cosmic rays with particle energies above $10^{14}$ eV is still unknown. Models of origin, acceleration, and propagation must be evaluated in light of the observed energy spectrum and chemical composition of the cosmic rays. Several experiments have attempted to determine the mean cosmic ray composition through the ``knee'' region of the spectrum, up to $3\\times10^{16}$eV\\cite{BLANCA}.  While the results are not in complete agreement, there is some consensus for a composition becoming heavier at energies above the knee, a result consistent with charge-dependent acceleration theories or rigidity-dependent escape models.  In the region above the knee, the Fly's Eye experiment has reported a changing composition from a heavy mix around $10^{17}$eV to a proton dominated flux around $10^{19}$eV \\cite{FEcomp}. This result makes this particular energy region much more interesting than the expectation from a naive rigidity model. This changing composition may imply that there may be multiple sources of cosmic rays.   The  AGASA experiment shows broad agreement with this trend if the data are interpreted  using the same hadronic interaction model as used in the Fly's Eye analysis \\cite{AGASA,DMS}.  The recently reported HiRes/MIA\\cite{prl} hybrid observation on the cosmic ray composition in a narrower energy region, $10^{17} \\sim 10^{18}$ eV, shows a general agreement with Fly's Eye experimental result. The new result gives a somewhat more rapid change in the composition. The reliability of these experimental results depends on how well we understand the development of all components of extensive air showers (EAS) produced by cosmic rays, how well we understand the response of our detector to the EAS, and how well we have done in the EAS reconstruction. In this paper, we  address all those issues in  detail.  We  start with a general description of the techniques of cosmic ray composition measurement in the high energy region.  \\subsection{Existing Techniques of Compostion Measurement above 10$^{17}$eV}  The Fly's Eye and AGASA experiments use different techniques to study composition. The Fly's Eye experiment technique is based on the assumption that the speed of the development of EAS depends on the mass of the primary particle: a heavier nucleus induces earlier EAS development in the atmosphere. The rapid break-up of a heavy nucleus at the early stage of the cascade in the air leads to an effectively higher multiplicity than that produced by a light nucleus or a proton at the same depth in the atmosphere. The consequence is that the EAS is built up by a superposition of smaller subshowers induced by nuclear fragments. In this case, since the subshowers have lower energies, the EAS will have a shower maximum higher in the atmosphere than in the case for a proton primary. The Fly's Eye experiment is designed to measure the size, or total number of charged particles, of an EAS as a function of atmospheric depth. It is thus an ideal detector to measure the depth of maxima of showers directly.  In practice, the intrinsic fluctuations in the depth of shower maximum and the detector resolution effects imply that one can not directly resolve the type of primary nucleus on an event by event basis. What one can do is to extract an average EAS primary composition by comparing the data to a Monte Carlo simulation with a given primary composition. Since the shower development is somewhat dependent on the choice of a hadronic model, this method leads to results which have some model dependence.  The other method for studying composition depends on the assumption that the muon content of EAS produced by a superposition of sub-showers ( as in the case of heavy nuclei ) is larger than those with fewer sub-showers.  This is due to the fact that the dissociation of a heavy nucleus produces a relatively higher multiplicity in its interaction with atmospheric nuclei. The resultant sharing of the primary energy between the nuclear fragments make the secondary pions less energetic. As a consequence, those pions have a greater decay probability into muons than those produced by a lighter nucleus in the early stage of shower development. This leads to a difference in the muon content between the EAS's induced by heavier and lighter nuclei, e.g. iron and protons.  This difference shrinks with energy because the available path for the decay of the high energy pions decreases as the shower develops deeper in the atmosphere. According to the simulations the difference is still resolvable in the energy region 10$^{17}$ $\\sim$ 10$^{18}$ eV.  In principle , one can obtain the information about the composition of the primary particle by either measuring the total number of muons in EAS or a local density of shower muons at a specific distance to the core of the shower. However, as in the case of the previous method, the fluctuations are large compared with the separation between the different types of EAS, so that the resolvability is not strong. The hadronic model dependence of the predicted $\\mu$-content from a particular composition is also significant for this method.   \\subsection{Advantages and Challenges for the HiRes/MIA Experiment}  The HiRes/MIA hybrid experiment is designed to combine the two methods together using two independently developed co-sited experiments. The two experiments share a trigger to simultaneously record both the longitudinal development information and the EAS muon density.  This results in a unique data set useful for the investigation of cosmic ray composition. Results on comparison of the flourescence and muon methods have previously been published\\cite{prl}. In this paper, we use the MIA array to improve the geometrical reconstruction and concentrate on the fluorescence technique for composition and spectrum studies.  The hybrid timing information enhances the accuracy in the determination of the geometry of the EAS. This accurate shower geometry plays a crucial role in the subtraction of the Cerenkov light component of the EAS and in the corrections for detector acceptance. These turn out to be the two key issues in the shower longitudinal development profile determination. This profile provides directly the size at shower maximum and its location while the integral of the profile yields the shower energy.  The improvement in shower geometry determination is the main advantage of a hybrid experiment.  The other advantage resulting from the coincident measurement with the surface muon array is the existence of a fully efficient triggering region in the central detector volume. This is very helpful for the cosmic ray energy spectrum measurement.  One of the challenges for this experiment is that the two devices are separated by only 3.3 km.  Because the MIA detector can not be triggered by remote showers, most of the triggered events are located $\\sim$ 4 km away from HiRes detector.  This short shower-detector distance gives rise to difficulties for reconstruction of those EAS's. The lateral distribution of shower electrons is no longer a negligible effect.  The broadened source of light increases the uncertainty in the shower geometry determination and in the acceptance correction of the signals.  The limited effective trigger area of MIA largely suppresses the aperture of HiRes detector. The energy coverage of our data is  one decade of magnitude over 3 years of observation, much smaller than that from the Fly's Eye experiment.  The geometry of the triggered events is such that the fraction of Cerenkov light is often large. This  implies the need for tight criteria for event selection to reduce Cerenkov contamination, which in turn may cause a bias. We must search for a balance between the tightness of the selection criteria and the minimization of bias.  We describe our resolution of these problems in the following sections. ",
        "conclusions": "The HiRes/MIA hybrid experiment has measured  the cosmic ray energy spectrum between $10^{17}$ and $3\\times 10^{18}$. The spectral index and intensity are given in Table.\\,\\ref{spectab}. The result is in agreement with the Fly's Eye experiment. This result marginally supports the Fly's Eye stereo observation of a break in the energy spectrum at $4\\times 10^{17}$ eV.   The HiRes/MIA hybrid experiment confirms the Fly's Eye experimental result that the elongation rate is different from simulation with an unchanging composition. Modern hadronic interaction models and improved detector resolution in energy and $X_{max}$ do not change the original conclusion. Within errors, the elongation rate observed in this experiment, $93.0\\pm 8.5\\pm (10.5)$\\,g/cm$^2$/decade, is consistent with previous experiments, such as Fly's Eye and Yakutsk\\cite{Yukutsk}.  While the conclusion regarding the absolute value of $\\overline{lnA}$ of the primary composition depends on the interaction model used, this study shows that the elongation rate is stable with respect to choice of models. In the light of this, the amount of the change in the average composition, i.e. $\\Delta\\overline{lnA} = -1.5\\pm 0.6$, is largely model independent, no matter what value of $\\overline{lnA}$ the change starts from.  Putting all experimental results together from $3\\times 10^{14}$ to $3\\times 10^{18}$ eV, we note that there seems to be a correlated patterns in the energy spectrum and elongation rate, $X_{max}$ vs. energy, plot. Both measurements in energy spectrum and $X_{max}$ imply a continuity from lower to higher energies, with a flat bridge between $10^{16}$  and $10^{17}$ eV.  We note that following the break, the Fly's Eye experiment \\cite{FEcomp} reports a hardening of the spectrum near $5\\times10^{18}$ eV. This has been interpreted as evidence for the emergence of an extragalactic component above a softer galactic component \\cite{FEcomp}.  A change from a heavy to a light composition in this energy region also gives support to a changing origin for those cosmic rays.  The lack of a strong galactic anisotropy at the highest energies would also rule out galactic sources for energetic protons \\cite{Alfred}. A number of new experiments, such as HiRes, the Pierre Auger Project, the Telescope Array, EUSO and OWL will address this issue."
    },
    "0010/astro-ph0010378_arXiv.txt": {
        "abstract": "Hot gas dominates the emission in X-ray luminous early-type galaxies, but in relatively X-ray faint systems, integrated X-ray emission from discrete stellar-like sources is thought to be considerable, although the amount of the contribution is controversial. To help resolve this issue, we examine the radial X-ray surface brightness distribution of 17 X-ray faint galaxies observed with the ROSAT HRI and PSPC.  We assume that the stellar contribution follows a de Vaucouleurs law while the hot gas component follows a King $\\beta$ model.  For some galaxies, both models fit equally well, but for a number of systems, a dual component model yields the best fit, from which, upper bounds are placed on the stellar contribution.  Best-fit values for the stellar contribution are inconsistent with (lower than) that suggested by Fabbiano, Gioia, \\& Trinchieri (1989) and estimated from the bulge of M31, but are consistent with the Forman, Jones, \\& Tucker (1985) estimate of the stellar fraction in X-ray faint elliptical and S0 galaxies. Our results indicate an upper limit to discrete sources of $L_X/L_B = 1.6\\times10^{29} \\, {\\rm ergs \\, s}^{-1}/L_{\\odot}$. ",
        "introduction": "\\label{sec:intro}  One key concern in the study of early-type galaxies, is the amount of discrete X-ray sources relative to hot gas in X-ray faint ellipticals.  It is generally agreed that integrated X-ray emission from stellar sources (such as accreting X-ray binaries) is likely to be present at some fraction in all elliptical galaxies. This contribution is low in X-ray--luminous galaxies (i.e. $L_X/L_B \\gtrsim 3.2\\times10^{30} \\, {\\rm ergs \\, s}^{-1}/L_{\\odot}$), where hot, diffuse gas dominates the total X-ray emission (e.g. Forman, Jones \\& Tucker 1985; Canizares, Fabbiano \\& Trinchieri  1987; Davis \\& White 1996; Brown \\& Bregman 1998; Buote \\& Fabian 1998). The fraction of X-ray emission from stellar sources is expected to increase with decreasing $L_X$, but the $L_X$ at which the stellar component ($L_{X,\\star}$) dominates the observed emission is still debated.  Using {\\em Einstein Observatory}, {\\em ROSAT}, and {\\em ASCA} data, researchers have derived estimates (or upper limits) to the stellar X-ray emission in E and S0 galaxies using various methods. Forman, Jones \\& Tucker (1985) determined an upper limit of $L_{X,\\star}/L_B \\simeq 4\\times10^{28} \\, {\\rm ergs \\, s}^{-1}/L_{\\odot}$ ({\\em Einstein} IPC band) by assuming discrete sources make a 50\\% contribution to the total diffuse X-ray emission of Cen A.  In this limit, only the faintest galaxies ($M_B > -19$ in their sample) can have their X-ray emission dominated by discrete-source emission. Irwin \\& Sarazin (1998) concluded that the X-ray emission of faint galaxies ($L_X/L_B < 5\\times10^{29} \\, {\\rm ergs \\, s}^{-1}/L_{\\odot}$) is a combination of a hard stellar component and a very soft component, which is most likely stellar in origin as well. This was inferred by comparing the X-ray ``colors\" of a sample of early-type galaxies to the bulge of M31 ($L_X/L_B = 3.2\\times10^{29} \\, {\\rm ergs \\, s}^{-1}/L_{\\odot}$; Irwin 2000, private communication). Fabbiano, Gioia \\& Trinchieri (1989) used the average $L_X/L_B$ of early-type spirals ($L_X/L_B = 4\\times10^{29} \\, {\\rm ergs \\, s}^{-1}/L_{\\odot}$, for $L_X = 10^{38} - 10^{42} \\, {\\rm ergs \\, s}^{-1}$) as the benchmark for the stellar contribution. By comparing the $L_X$ and $L_B$ of elliptical and S0 galaxies to spirals (in the {\\em Einstein} 0.5--4.5 keV), they concluded that X-ray emission from hot gas is not significant in systems for which $L_X < 10^{41} \\, {\\rm ergs \\, s}^{-1}$. In the various estimates of the stellar emission fraction, there is an order of magnitude difference, making an accurate determination even more necessary.  The primary mechanisms for the X-ray emission can be determined through the examination of X-ray spectra (e.g., Trinchieri, Kim, Fabbiano, \\& Canizares 1994; Loewenstein \\& Mushotzky 1997; Buote \\& Fabian 1998). A hard spectrum ($T_X \\sim$ 10 keV) would indicate the presence of evolved stellar sources, while soft X-rays ($T_X \\sim$ 1 keV) would be reflective of a hot gas component.  Recent work incorporating the use of multi-temperature spectral models indicate the presence of a very soft component in addition to the hard and soft components described above, the nature of which is suspected by most to be stellar in origin but which may be due in part to a warm ISM (Pellegrini \\& Fabbiano 1994; Fabbiano, Kim, \\& Trinchieri 1994; Kim et al. 1996). {\\em ROSAT}, although it has better angular resolution than the {\\em Einstein Observatory}, does not provide high enough spectral resolution to show signatures unique to a stellar population, especially in fainter systems.  Another problem with spectral modeling is that acceptable fits can be obtained with both single temperature models and multi-temperature models. Even with the better resolution of {\\em ASCA}, calibration problems at low energies may affect spectral modeling.  The use of X-ray radial surface brightness profiles offers another way to determine the relative fractions of gas and discrete sources to the observed X-ray emission. Previously, these profiles existed only for X-ray-bright galaxies for which adequate signal-to-noise data was achieved, and were used to obtain gas densities and masses, test predictions of the cooling flow model, or model the dynamical evolution of gas flows (Sarazin \\& White 1988; Pellegrini \\& Ciotti 1998; David, Forman \\& Jones 1991).  The use of surface brightness profiles to infer the contributions to the X-ray emission has been limited, and few conclusions have been reached with regard to faint galaxies (Forman, Jones \\& Tucker 1985; Trinchieri, G., Fabbiano, G., \\& Canizares, C. R. 1986; Pellegrini \\& Fabbiano 1994).  Data has become available through {\\em ROSAT} that better allow for the examination of radial surface brightness profiles of fainter galaxies. Brown \\& Bregman (1998, 2000) provide X-ray information on a {\\em ROSAT} survey of early-type galaxies extending to fainter X-ray luminosities than available from {\\em Einstein Observatory} data.  In this paper, radial surface brightness profiles are used to determine or set upper limits to the stellar contribution in the less luminous galaxies of this survey. ",
        "conclusions": "\\label{sec:disc}  We compare the modeled estimates of the stellar fraction to the estimates of Forman, Jones \\& Tucker (1985, FJT) and Fabbiano, Gioia \\& Trinchieri (1989, FGT), and that estimated from M31 (Irwin 2000, private communication). The linear curves are superimposed upon a plot of the X-ray luminosities of the targeted galaxies against their optical blue luminosities (Figure \\ref{fig:answer}). For log($L_B/ L_{\\odot}) = 10.4-11$, the total $L_X$ of our sample spans the region between these estimates. The seven galaxies, whose radial surface brightness profiles could be fitted with combined King and de Vaucouleurs models, all have total $L_X/L_B$ values in the brighter half of the region of interest. The best fit and 90\\% upper limit $L_{X,\\star}$ values are plotted for those galaxies as well.  \\begin{figure}[tbp] \\begin{center} \\plotone{fig5.ps} \\caption{\\small X-ray vs. optical luminosities for the target sample. Errors in $L_X$ are due to photon statistics. $L_B$ for NGC 2768 has been artificially lowered by 0.02 in the log for clarity, and NGC 5102 lies beyond the plot boundaries. Stars are the ``best-fit\" $L_{X,\\star}$ values determined from a two-component fit while downward arrows are 90\\% upper limits. The linear lines are the stellar estimates of Forman, Jones \\& Tucker (1985, FJT) and Fabbiano, Gioia \\& Trinchieri (1989, FGT), and estimated from M31 (Irwin 2000, private communication) with log($L_{X,\\star}/L_B$) = 28.56, 29.56, and 29.51 respectively ($L_{X,\\star}$ in ergs  s$^{-1}$, $L_B$ in $L_{\\odot}$). \\label{fig:answer}} \\end{center} \\end{figure}  Best-fit modeling indicates a hot gas-dominated emission in each of the seven galaxies fit with the combined profile. The median for the best fits is log($L_{X,\\star}/L_B$) = 28.61 ($L_X$ in ergs s$^{-1}$ and $L_B$ in $L_{\\odot}$), in reasonable agreement with the Forman, Jones \\& Tucker (1985) upper limit of log($L_{X,\\star}/L_B$) = 28.6. In the 90-99\\% upper limit, however, all but two of the seven galaxies can be modeled with a combination profile indicative of a stellar-dominated emission.  The median for the 99\\% confidence fits is log($L_{X,\\star}/L_B$) = 29.31, which is a factor of $\\sim$1.8 and 1.6 below the limits of Fabbiano, Gioia \\& Trinchieri (1989) and M31 respectively. We further find $3\\sigma$ upper limits of log($L_{X,\\star}/L_B$) = 29.15 - 29.26 for the X-ray faintest galaxies not modeled, assuming that all of the observed emission is stellar.  We suggest, then, an upper limit to log($L_{X,\\star}/L_B$) of 29.2 for the integrated X-ray emission from discrete sources in X-ray faint early-type galaxies. This supports the Irwin \\& Sarazin (1998) suggestion that the brightest of X-ray faint galaxies (log $L_{X,\\star}/L_B < 29.7$) may retain a significant amount of gas with a temperature of 0.3--0.6 keV.  We explore how our results may relate to recent spectral studies of faint E and S0 galaxies. {\\em ASCA} studies of early-type galaxies confirm a hard spectral component attributed to the integrated emission from LMXBs (e.g., Matsumoto et al. 1997), and a very soft component (VSC), initially detected in {\\em ROSAT} data (Kim, Fabbiano, \\& Trinchieri 1992; Kim et al. 1996). For galaxies with log $L_X/L_B \\geq 30.0$, spectral signatures harden as $L_X/L_B$ decreases, until the hard component dominates the emission. The Kim et al. (1996) study of an X-ray faint S0 galaxy, NGC 4382, reports that the hard spectral component contributes one-half to three-fourths of the total X-ray luminosity, with the remainder due to the VSC. This is consistent with the earlier {\\em ROSAT} results which indicate that the VCS and hard components contribute near equally to the total X-ray emission in galaxies with log $L_X/L_B \\leq 30.0$ ($L_X$ in ergs s$^{-1}$ and $L_B$ in $L_{\\odot}$). The VSC may be due to a warm ISM, arise from stellar sources, or be a combination of both (Irwin \\& Sarazin 1998; Fabbiano, Kim, \\& Trinchieri 1994). Our results may be indicative of the VSC having an ISM origin, considering that the best-fit modeling is consistent with a substantial ISM component.  In this initial analysis of X-ray surface brightness profiles, we have assumed low-mass X-ray binaries (LMXBs) are the primary constituents of a discrete X-ray source population. LMXBs contribute to the bulk of X-ray emission in large spiral bulges such as M31 (Trinchieri \\& Fabbiano 1985). Because spiral bulges are very similar to elliptical galaxies in their properties and stellar populations, it is reasonable to presuppose that LMXBs are also the main contributors to the discrete source population in early-type galaxies (David, Forman, \\& Jones 1991; Irwin \\& Sarazin 1998). However, it is possible that globular clusters may be a significant source of X-ray emission in X-ray faint E and S0 galaxies (Trinchieri \\& Fabbiano 1985), in which case the X-ray radial brightness distribution due to non-gaseous sources may be more appropriately described by a model other than a pure de Vaucouleurs $r^{1/4}$ law.  The goal of finding the stellar contribution in X-ray faint galaxies has been hampered by low photon counts and lack of data from both {\\em ROSAT} detectors. Long observations with Chandra will be especially helpful in constraining the core radii in faint galaxies, fixing the shape of the radial surface brightness profiles, and resolving bright point sources. It is also hoped that Chandra will provide the spectral resolution needed to determine whether the ``very soft component\" modeled by various authors (see \\S\\ref{sec:intro}) is due to a warm ISM of to discrete sources.  Additionally, techniques that circumvent the problem of low counts, such as defining X-ray ``colors\" (Irwin \\& Sarazin 1998), will be helpful in obtaining a definite answer to the long-standing question of dominant emission mechanisms in X-ray faint early-type galaxies."
    },
    "0010/astro-ph0010464_arXiv.txt": {
        "abstract": "The collapse and fragmentation of filamentary primordial gas clouds are explored using one-dimensional and two-dimensional hydrodynamical simulations coupled with the nonequilibrium processes of hydrogen molecule formation. The cloud evolution is computed from the initial central density $n_c=10 - 10^6$ cm$^{-3}$. The simulations show that depending upon the initial density, there are two occasions for the fragmentation of primordial filaments. If a filament has relatively low initial density such as $n_c \\lesssim 10^5$ cm$^{-3}$, the radial contraction is slow due to less effective H$_2$ cooling and appreciably decelerates at densities higher than a critical density, where LTE populations are achieved for the rotational levels of H$_2$ molecules and the cooling timescale becomes  accordingly longer than the free-fall timescale. This filament tends to fragment into dense clumps before the central density reaches $10^{8-9}$ cm$^{-3}$, where H$_2$ cooling by three-body reactions is effective and the fragment mass is more massive than some tens $M_\\odot$. In contrast, if a filament is initially as dense as $n_c \\gtrsim 10^5$ cm$^{-3}$, the more effective H$_2$ cooling with the help of three-body reactions allows the filament to contract up to $n\\sim 10^{12}$ cm$^{-3}$. After the density reaches $n\\sim 10^{12}$ cm$^{-3}$, the filament becomes optically thick to H$_2$ lines and the radial contraction subsequently almost stops. At this final hydrostatic stage, the fragment mass is lowered down to $\\approx 1M_\\odot$ because of the high density of the filament. The dependence of the fragment mass upon the initial density could be translated into the dependence on the local amplitude of random Gaussian density fields or the epoch of the collapse of a parent cloud. Hence, it is predicted that the initial mass function of Population III stars is likely to be bimodal with peaks of $\\approx 10^2 M_\\odot$ and $\\approx 1M_\\odot$, where the relative heights could be a function of the collapse epoch. Implications for the metal enrichment by Population III stars at high redshifts and baryonic dark matter are briefly discussed. ",
        "introduction": "The existence of a very first generation of objects, namely Population III, has been originally postulated by the presence of a noticeable amount of heavy elements in Population II stars (see, e.g., Carr, Bond, \\& Arnett 1984 and references therein) and has recently gained increasing importance owing to the discovery of intergalactic metals in the Ly$\\alpha$ forest (Cowie et al. 1995; Songaila \\& Cowie 1996). In those respects, Population III objects need to be such massive stars that they can produce metals at the end of their evolution. Recent theoretical analyses on the evolution of metal-free stars predict that the fate of the massive metal-free stars can be classified as follows (e.g., Heger, Woosley, \\& Waters 2000; Chiosi 2000; see also Portinari, Chiosi, \\& Bressan 1998 for the effects of mass loss.): (1) A star with a mass of $m \\gtrsim 250 M_\\odot$ collapses completely to a black hole (BH) without ejecting any heavy elements (Bond, Carr \\& Arnett 1984). (2) A star of $100M_\\odot \\lesssim m \\lesssim 250 M_\\odot$ is partly or completely disrupted by electron-positron pair instability. For $m \\gtrsim 150 M_\\odot$, the core completely disrupts, and the whole core involving heavy elements is injected in the intergalactic medium. If it is an extremely energetic event (a hypernova), it might lead to a gamma ray burst (GRB). (3) A star of $35M_\\odot \\lesssim m \\lesssim 100 M_\\odot$ probably collapses into a black hole. (4) A star of $10M_\\odot \\lesssim m \\lesssim 35 M_\\odot$ results in a type II supernova.  Population III stars are related to various issues that are currently the object of considerable attention. The luminous Population III stars could cause the reionization of the universe at redshifts $\\gtrsim 5$ (Couchman \\& Rees 1986; Fukugita \\& Kawasaki 1994; Ostriker \\& Gnedin 1996; Gnedin \\& Ostriker 1997; Haiman \\& Loeb 1997; Miralda-Escud\\'{e} \\& Rees 1998; Gnedin 2000). Alternatively, moderately massive BHs as the end products of massive stars might coagulate into a super-massive BH, evolving to primordial AGNs (Larson 2000). The accreting super-massive BHs may be more responsible for cosmic reionization (Tegmark \\& Silk 1994; Sasaki \\& Umemura 1996; Haiman \\& Loeb 1998; Valageas \\& Silk 1999). In addition, Population III stars may play an important role in the early evolution of galaxies (e.g., Tegmark, Silk, \\& Blanchard 1994; Ostriker \\& Gnedin 1996) or the early formation of massive BHs of $\\sim 10^5 M_\\odot$ (Umemura, Loeb, \\& Turner 1993). They may be responsible for the observed abundance patterns of extremely metal-deficient stars (McWilliam et al. 1995; Ryan, Norris, \\& Beers 1996; Audouze \\& Silk 1995; Shigeyama \\& Tsujimoto 1998). Finally, if a significant number of MACHOs are ancient white dwarfs (M\\'{e}ndez \\& Minniti 2000), they may stem from low-mass Population III stars (Carr 1994; Larson 1998; Rees 1999). In the light of all of such possible significant consequences, the initial mass function (IMF) of Population III stars is an issue of undoubted importance.  Many authors have studied the collapse of primordial clouds to estimate the masses of Population III stars (e.g., Matsuda, Sato, \\& Takeda 1969; Yoneyama 1972; Hutchins 1976; Silk 1977, 1983; Carlberg 1981; Palla, Salpeter, \\& Stahler 1983; Yoshii \\& Saio 1986; Nishi et al. 1998). These studies have emphasized the importance of radiative cooling by H$_2$ molecules because the primordial gas is deficient in heavy elements, which are the most efficient coolants in present-day star formation. In the first collapsed objects, baryonic gas is heated to temperatures above $10^3$ K during the contraction. This enhances the H$_2$ formation rate and causes the H$_2$ abundance to rise from its initial value of $\\sim 10^{-6}$ to a quasi-equilibrium value of $10^{-4}\\sim 10^{-3}$ (e.g., Palla et al. 1983; Lepp \\& Shull 1984; Haiman, Thoul \\& Loeb 1996). Thus, H$_2$ molecules cool the gas to a temperature of $\\sim 500$ K. As a result, the Jeans mass descends to a stellar mass and Population III stars can form in this way through fragmentation of the first collapsed objects. Although many elaborate analyses have been made, the estimated masses of Population III stars have not been well converged. Several authors have suggested that Population III stars were low mass, whereas others have suggested that they were massive or very massive. This discrepancy seems to come from most studies being restricted to highly simplified models such as homogeneous, pressureless, and/or spherical collapses.  In the bottom-up scenarios like cold dark matter (CDM) models, the first collapsed pregalactic objects should form at redshifts of $z\\sim 10-10^2$ and have masses of $10^5 \\sim 10^8M_\\odot$ (Tegmark et al. 1997). Because of the asymptotic scale-invariance of the CDM density fluctuations, pregalactic clouds with this mass range could undergo a run-away collapse, forming cores of $\\sim 10^2 M_\\odot$ (Abel et al. 1998; Abel, Norman, \\& Bryan 2000), or form mini-pancakes, fragmenting into pieces of $10^2 M_\\odot-10^3 M_\\odot$ (Bromm, Coppi, \\& Larson 1999). In these calculations, one of the common features is the formation of a filamentary structure. Filamentary clouds are gravitationally unstable and likely to fragment into dense clumps. Such dense clumps are expected to evolve into Population III stars. The physics of the fragmentation of filamentary primordial clouds has been studied analytically (Uehara et al. 1996) or by a one-dimensional numerical simulation (Nakamura \\& Umemura 1999, hereafter Paper I). These studies suggest that the minimum mass of Population III stars is of the order of $M_\\odot$. However, the physical processes are still unclear to determine whether Population III stars can be far above $M_\\odot$ or be eventually reduced to a few $M_\\odot$. Thus, as a further step following Paper I, we here perform two-dimensional hydrodynamical simulations. Attention is focused on elucidating the physical process of the fragmentation of primordial gas filaments to assess the mass of Population III stars.  In Paper I, we pursued the radial contraction of primordial filaments and showed that the filaments continue to contract quasi-statically, the temperatures staying nearly constant at $T\\sim 500$ K. When the cloud becomes optically thick to the H$_2$ lines, the radial contraction essentially stops. Applying a linear stability analysis, the fragmentation was expected to take place at that stage and the minimum masses of Population III stars were estimated as a few $M_\\odot$. Thus, Population III stars were anticipated to be low-mass deficient compared to the present-day stars. In this paper, we pursue the fragmentation processes of the filaments by means of two-dimensional axisymmetric simulations. The present model is an improved version of Paper I. The numerical model and method are described in \\S 2. Numerical results are given in \\S 3 and \\S 4. We show that filaments with low initial density can fragment into dense clumps before the cloud becomes optically thick to the H$_2$ lines. Then, the masses of the clumps could be much more massive than a few $M_\\odot$. However, relatively dense filaments result in clumps of a few $M_\\odot$. Hence, in \\S 5, it is predicted that the initial mass function of Population III stars is likely to be bimodal. In \\S 6, we discuss some implications for the first metal enrichment and baryonic dark matter. ",
        "conclusions": "We have explored the collapse and fragmentation of filamentary primordial gas clouds numerically, including the nonequilibrium processes for hydrogen molecule formation. The simulations have shown that, depending upon the initial density, the evolution of the filaments can be classified into two types. If a filament has relatively low initial density such as $n_c \\lesssim 10^5$ cm$^{-3}$, the radial contraction is slow due to less effective H$_2$ cooling, and it appreciably decelerates at density higher than a critical density, where LTE populations are achieved for the rotational levels of H$_2$ molecules and the cooling timescale becomes  accordingly longer than the free-fall timescale. This filament tends to fragment into dense clumps before the central density reaches $10^{8-9}$ cm$^{-3}$ where H$_2$ cooling by three-body reaction is effective, and the clump mass is more massive than some tens $M_\\odot$. In contrast, if a filament is initially as dense as $n_c \\gtrsim 10^5$ cm$^{-3}$, more effective H$_2$ cooling with the help of three-body reaction allows the filament to contract up to $n\\sim 10^{12}$ cm$^{-3}$, for which the filament becomes optically thick to H$_2$ lines and then the radial contraction almost stops. At this final hydrostatic stage, the clump mass is lowered down to $\\approx 1 - 2 M_\\odot$ because of the high density of the filament. The dependence of clump mass upon the initial density could be translated into the dependence of the local amplitude of random Gaussian density fields or the epoch of collapse of a parent cloud. Hence, the distribution of the clump mass predicts that the initial mass function of Population III stars is likely to be bimodal with peaks of $\\approx 1 - 2 M_\\odot$ and $\\approx 10^2 M_\\odot$, where the relative heights could be a function of the collapse epoch. At higher redshifts or for higher $\\sigma$ density fluctuations, the contribution of the lower-mass component is likely to be more significant in the IMF and the relative height of the first peak probably decreases with time because the initial densities of the filaments are likely to descend with time.   In our model, we do not take the effects of external radiation into account. As suggested by Abel et al. (2000), a single star may form at the central high-density region of the first collapsed low-mass objects. The radiative feedback from the very first stars might be significant for subsequent star formation in the surrounding medium (e.g., Ferrara 1998; Ciardi, Ferrara, \\& Abel 2000).    We modeled the Population III IMF as a superposition of two power-law distributions with different peaks and estimated the metallicity produced by Population III stars in the first collapsed objects at high redshifts. If the star formation efficiency is of the same order as the present-day value $\\epsilon \\sim 10^{-2}$, then the metallicity is estimated as $10^{-3}-10^{-2}Z_\\odot$. This metallicity is consistent with that observed in the intergalactic medium at high redshifts.   If a significant amount of stars with masses between a few $M_\\odot$ and 8$M_\\odot$ were formed, they might have evolved into white dwarfs until the present epoch. The old white dwarfs might presently reside in the galactic halo and may be related to dark component observed in microlensing experiments, i.e., massive compact halo objects (MACHOs). If the star formation efficiency is of the same order as the present-day value, then the ancient white dwarfs are likely to contribute a few tens \\% of the dark mass in the galactic halo, which seems to be consistent with some constraints discussed by several authors (e.g., Charlot \\& Silk 1995; Alcock et al. 2000; Lasserre et al. 2000; M\\'endez \\& Minniti 2000; Hodgkin et al. 2000; Ibata et al. 2000)."
    },
    "0010/astro-ph0010187_arXiv.txt": {
        "abstract": "We report Keck LRIS observations of an arc-like background galaxy near the center of Abell~2152 ($z=0.043$), one of the three clusters comprising the Hercules supercluster. The background object has a redshift $z=0.1423$ and is situated 25\\arcsec\\ north of the primary component of the A2152 brightest cluster galaxy (BCG). The object is about 15\\arcsec\\ in total length and has a reddening-corrected $R$-band magnitude of $m_R = 18.55\\pm0.03$.  Its spectrum shows numerous strong emission lines, as well as absorption features. The strength of the H$\\alpha$ emission would imply a star formation rate $\\hbox{SFR}\\approx3\\;h^{-2}\\,\\msun\\,$yr$^{-1}$ in the absence of any lensing. However, the curved shaped of this object and its tangential orientation along the major axis of the BCG suggest lensing.  We model the A2152 core mass distribution including the two BCG components and the cluster potential. We present velocity and velocity dispersion profile measurements for the two BCG components and use these to help constrain the potential. The lens modeling indicates a likely magnification factor of $\\sim1.9$ for the lensed galaxy, making A2152 the nearest cluster in which such significant lensing of a background source has been observed. Finally, we see evidence for a concentration of early-type galaxies at $z=0.13$ near the centroid of the X-ray emission previously attributed to A2152.  We suggest that emission from this background concentration is the cause of the offset of the X-ray center from the A2152 BCG.  The background concentration and the dispersed mass of the Hercules supercluster could add further to the lensing strength of the A2152 cluster. ",
        "introduction": "Strong gravitational lensing of background sources by clusters of galaxies provides some of the most unambiguous evidence for the presence of large amounts of dark matter in clusters.  Lensing studies suggest that the mass in clusters is greater and more centrally concentrated than implied, for instance, by the X-ray properties (e.g., Mellier, Fort, \\& Kneib 1993; LeF\\`evre \\etal\\ 1994; Waxman \\& Miralda-Escude 1995; Smail \\etal\\ 1995). Moreover, the mass appears to clump around the luminous galaxies, so that consideration of substructure and the effect of the massive central galaxy are often important for understanding the observations (e.g., Miralda-Escud\\'e 1995; see also Tyson, Kochanski, \\& Dell Antonio 1998). This is particularly true in more moderate mass clusters with velocity dispersions $\\scl \\lta 1000$ \\kms, for which the proportionally larger gravitational effects of a massive cD must be included to properly account for the lensing (Williams, Navarro, \\& Bartelmann 1999).  Although there have been great advances in our knowledge of the form of the density distributions of massive dissipationless particles in simulations of various resolution (e.g., Dubinski \\& Carlberg 1991; Navarro, Frenk, \\& White 1996, 1997; Moore \\etal\\ 1998, 1999b), our knowledge of the central mass distributions of actual galaxy clusters remains fragmentary. Low redshift clusters, because of their larger angular sizes, can in principle provide a better opportunity for studying these mass distributions. Their lensing properties are acutely sensitive to the shape of the inner cluster potential, although because such clusters tend to be of modest mass (very high-mass clusters being few and far between), the individual galaxies will produce relatively larger perturbations on this potential, complicating the interpretation.  Until recently, moderate- to high-redshift clusters have been the exclusive purview of strong lensing studies, but lensing by low-redshift clusters has been gaining increased attention. For instance, Allen, Fabian, \\& Kneib (1996) discovered a redshift $z=0.43$ lensed arc in the massive cooling flow cluster PKS\\,0745--191 at $z=0.103$. Campusano, Kneib, \\& Hardy (1998) explored lensing models for a $z=0.073$ spiral galaxy located near the central elliptical in the richness class~0 cluster A3408 at $z=0.042$; they obtained an upper limit to the magnification factor of $\\sim$1.7 from their ``maximum mass'' model.  This object turned out to be the only viable lensing candidate found in an imaging survey of 33 southern galaxy clusters with $z\\le0.076$ by Cypriano \\etal\\ (2000), who also presented an estimate of the expected number of arcs and arclets in low-redshift clusters. Blakeslee \\& Metzger (1999) discovered a lensed arc at $z=0.573$ in the nearby cD cluster A2124 at $z=0.066$ and found magnification factors near 10 from their best-fitting models of this system.  Despite the recent progress, the lensing properties of low-redshift clusters remain unclear.  For instance, A2124 had by far the lowest redshift in the sample studied by Williams \\etal\\ (1999) of 24 lensing clusters with measured velocity dispersions, the next nearest being at $z=0.171$.  In this paper, we present imaging and spectroscopic observations and give a lensing analysis of the redshift $z{\\,=\\,}0.043$ cluster Abell 2152 in the Hercules supercluster. ",
        "conclusions": "\\label{sec:disc}  Deep $R$-band Keck imaging taken under excellent seeing conditions has revealed a curved object 25\\farcs2 (15 \\hkpc) from the center of the A2152 BCG, along the major axis. The reddening-corrected $R$~magnitude of this ``arc'' is $m_R=18.55\\pm0.03$, and follow-up spectroscopy shows that it lies at a redshift $z=0.1423\\pm0.0001$. Its position, orientation, and redshift all suggest a lensing explanation for the curiously bowed structure. The object has numerous strong emission lines, and the equivalent width of H$\\alpha$ is $117\\pm6$ \\AA, from which we infer a star formation rate of $(3\\pm1)\\;h^{-2}\\,\\msun\\,$yr$^{-1}$.  We have presented spatially resolved spectroscopy of the A2152 BCG and SRG, which are separated by 47\\arcsec\\ and are together catalogued as UGC\\,10187. The relative velocity of the BCG and SRG is $331\\pm7$ \\kms, and both galaxies show significant rotation. The rotational velocities along the observed position angle are $\\sim50$ \\kms\\ and $\\sim120$ \\kms\\ for the BCG and SRG, respectively. Our velocity dispersion profile for the BCG extends to a radius of $\\sim16\\arcsec$ (9.5\\,\\hMpc), about 60\\% of the way to the radial position of the lensed galaxy.  We have modeled the lensing potential of A2152 using constraints from the measured cluster velocity dispersion of 715 \\kms\\ and our own stellar velocity dispersion measurements for the BCG and SRG. The models cannot produce a significant lensing effect on galaxy A if the  A2152 potential center is taken to coincide with the peak of the overall X-ray emission. However, if the potential center is taken to be the A2152 BCG, then the models {\\it can} reproduce the observed position and apparent bowed shape of the galaxy, and the implied magnification is a factor of about 1.9.  Conversely, we can say that if the object is significantly lensed, then the 2\\farcm1 offset of the X-ray centroid from the BCG is due to contamination from another source of X-ray emission. We note that both the cluster velocity dispersion and the core softening radius in our lensing model is degenerate with the source position. However, the higher values that have been reported for the dispersion (see \\S\\ref{sec:clusprops}), if centered on the BCG, produce lens models with critical curves that lie outside of the lensed galaxy, thus multiple images would be expected of any background galaxy at the position and redshift of A.  We can therefore rule out this case from a lensing argument; even a large core radius model would produce significantly more distortion in a $z=0.142$ galaxy.  The available redshift data provide evidence for a background association of early-type galaxies at $z\\approx0.134$, projected near the center of A2152. Two of the galaxies, which we call G1/G2, have redshifts from the Barmby \\& Huchra (1998) catalogue. They are separated by 30\\arcsec\\ and are each about 70\\% more luminous than the A2152 BCG. The third galaxy, called G3, is 35\\arcsec\\ southeast of the A2152 BCG and was picked up by the LRIS slit during our observations.  It is 116\\arcsec\\ from G1 and 132\\arcsec\\ from G2, or metric distances of 188-215 \\hkpc. The velocity range for these three galaxies is $\\sim830$ \\kms. While three galaxies do not a cluster make, the luminosities, velocities, and early-type morphologies of the galaxies are highly suggestive. Moreover, there is a strong concentration of small galaxies of unknown redshift centered near the bright G1/G2 pair. It is highly likely that some of the X-ray emission, which is centered just 20\\arcsec\\ from G2, could be due to a grouping of galaxies associated with G1/G2. Therefore, the center of the true A2152 X-ray emission, and the center of the A2152 cluster potential, might well coincide with the A2152 BCG, as required by our lensing models.  We note that the centers of the outermost X-ray contours in the Jones \\& Forman map of this field drift away from G1/G2 and closer to the A2152 BCG; the optical H$\\beta$ emission we find in the BCG spectrum may be additional circumstantial evidence for central location in the cluster potential.  However, a high-resolution X-ray map is needed to confirm this hypothesis.  If there is a background cluster, galaxy~A may well be a member, in which case it would be falling away from us through the cluster with a velocity of $\\sim2500$ \\kms.  Simulations show that the tidal effects of clusters on infalling galaxies can warp spiral disks into arc-like shapes (Moore \\etal\\ 1999a) and even form giant arcs of tidal debris (Calc{\\'a}neo-Rold{\\'a}n \\etal\\ 2000). Thus, tidal interaction with a local cluster potential, rather than lensing by the foreground A2152 potential, could be responsible for the distorted appearance of the galaxy, as well as for the observed strong star formation.  Its location and orientation with respect to the A2152 BCG would then be merely fortuitous. However, the galaxy in question would have to be caught during its initial pass through the cluster, since gas in cluster galaxies tends to be lost on relatively short time-scales, truncating the star formation (e.g., Balogh \\etal\\ 1998; Kodama \\& Bower 2000). The probability that galaxy A is a member of a cluster at $z=0.13$ depends of course on the velocity dispersion of the cluster.  A deep redshift survey of this field is needed to discern whether or not there is a very substantial background cluster (as suggested by the luminosities of G1/G2 and the position of the X-ray center in this field), and how likely galaxy A's membership is.  Such a redshift survey would need to go at least $\\sim2$~mag fainter than the Barmby \\& Huchra (1998) limit in order to pick up the multitude of small galaxies that lie between the A2152 BCG/SRG pair and the G1/G2 background pair.  Although superficially A2152 might have seemed an unremarkable low-redshift cluster (leaving aside the unlikely velocity dispersions in excess 1300 \\kms\\ found in the literature), our photometric and spectroscopic observations of its central field have raised a host of intriguing new questions about this cluster and the arc-shaped galaxy projected near its BCG. How much mass is there along this line of sight and how is it distributed?  Is the A2152 BCG actually at the center of the cluster potential and the observed offset with respect to the X-ray center due to the superposed emission of a background cluster at $z=0.13$?  If so, is the distorted galaxy a member which is being distorted by the hypothetical background cluster, or is it a field galaxy being lensed by A2152? Along with high-resolution X-ray observations and a deep redshift survey, a weak lensing analysis of this field would help in unraveling the projected mass distribution and provide insight into these questions.  We believe that the weight of the evidence favors the view that the A2152 BCG is at the center of its cluster's potential and is significantly lensing galaxy A.  Further lensed objects, if present near critical curves, could provide useful future constraints on the central structure of rich clusters.  For now, however, a complete understanding of this system and definitive answers to the questions raised by this study must await an infusion of new data."
    },
    "0010/astro-ph0010302_arXiv.txt": {
        "abstract": "We consider a sample of 51 distant galaxy clusters at $0.15\\lesssim z \\lesssim 0.9$ ($<z>\\sim0.3$), each cluster having at least 10 galaxies with available redshift in the literature.  We select member galaxies, analyze the velocity dispersion profiles, and evaluate in a homogeneous way cluster velocity dispersions and virial masses.  We apply the same procedures already recently applied on a sample of nearby clusters ($z<0.15$, Girardi et al. 1998b) in order to properly analyze the possible dynamical evolution of galaxy clusters. We remark problems induced by the poor sampling and the small spatial extension of the sampled cluster region in the computation of velocity dispersion.  We do not find any significant difference between nearby and distant clusters.  In particular, we consider the galaxy spatial distribution, the shape of the velocity dispersion profile, and the relations between velocity dispersion and X--ray luminosity and temperature.  Our results imply little dynamical evolution in the range of redshift spanned by our cluster sample, and suggest that the typical redshift of cluster formation is higher than that of the sample we analyze.   \\vspace*{6pt} \\noindent {\\em Subject headings: } galaxies: distances and redshifts - X-rays: galaxies - cosmology: observations. ",
        "introduction": " ",
        "conclusions": ""
    },
    "0010/astro-ph0010134_arXiv.txt": {
        "abstract": "We report on a detailed kinematic study of the galactic-scale outflow in the Seyfert galaxy NGC 2992.  The TAURUS-2 Imaging Fabry-Perot Interferometer was used on the Anglo-Australian 3.9-m telescope to derive the two-dimensional velocity field of the H$\\alpha$-emitting gas over the central arcminute of NGC 2992. The complete two-dimensional coverage of the data combined with simple kinematic models of rotating axisymmetric disks allows us to differentiate the outflowing material from the line-emitting material associated with the galactic disk.  The kinematics of the disk component out to R = 3.0 kpc are well modeled by pure circular rotation in a plane inclined at $i = 68\\arcdeg \\pm 3\\arcdeg$ from the plane of the sky and with kinematic major axis along P.A. = 32$\\arcdeg \\pm 3\\arcdeg$. The outflow component is distributed into two wide cones with opening angle $\\approx$ 125 -- 135$\\arcdeg$ and extending $\\sim$ 2.8 kpc (18$\\arcsec$) on both sides of the nucleus at nearly right angles ($\\phi \\approx 116\\arcdeg \\pm 5\\arcdeg$) to the disk kinematic major axis.  The outflow on the SE side of the nucleus is made of two distinct kinematic components interpreted as the front and back walls of a cone. The azimuthal velocity gradient in the back-wall component reflects residual rotational motion which indicates either that the outflowing material was lifted from the disk or that the underlying galactic disk is contributing slightly to this component.  A single outflow component is detected in the NW cone.  A biconical outflow model with velocities ranging from 50 to 200 km~s$^{-1}$ and oriented nearly perpendicular to the galactic disk can explain the data.  The broad line profiles and asymmetries in the velocity fields suggest that some of the entrained line-emitting material may lie inside the biconical structure rather than only on the surface of the bicone. The mass involved in this outflow is of order $\\sim$ 1 $\\times$ 10$^7$ n$_{e,2}^{-1}$ M$_\\odot$, and the bulk and ``turbulent'' kinematic energies are $\\sim$ 6 $\\times$ 10$^{53}$ n$_{e,2}^{-1}$ ergs and $\\sim$ 3 $\\times$ 10$^{54}$ n$_{e,2}^{-1}$ ergs, respectively. The most likely energy source is a hot bipolar thermal wind powered on sub-kpc scale by the AGN and diverted along the galaxy minor axis by the pressure gradient of the ISM in the host galaxy.  The data are not consistent with a starburst-driven wind or a collimated outflow powered by radio jets. ",
        "introduction": "Active galactic nuclei (AGN) and nuclear starbursts may severely disrupt the gas phase of galaxies through deposition of a large amount of mechanical energy in the centers of galaxies.  As a result, a large-scale galactic wind (``superwind'') that encompasses much of the central regions of these galaxies may be created (e.g., Chevalier \\& Clegg 1985; Schiano 1985; Tenorio-Tagle \\& Bodenheimer 1988; Tomisaka \\& Ikeuchi 1988; MacLow \\& McCray 1988; MacLow, McCray, \\& Norman 1989).  Depending upon the extent of the halo and its density and upon the wind's mechanical luminosity and duration, the wind may ultimately blow out through the halo and into the intergalactic medium.  The effects of these superwinds may be far-reaching. Bregman (1978) has suggested that the Hubble sequence can be understood in terms of a galaxy's greater ability to sustain winds with increasing bulge-to-disk ratio. Superwinds may affect the thermal and chemical evolution of galaxies by depositing large quantities of hot, metal-enriched material on the outskirts of galaxies. This widespread circulation of matter and energy between the disks and halos of galaxies may be responsible for the metallicity-radius relation within galaxies and the mass-metallicity relation between galaxies (e.g., Larson \\& Dinerstein 1975; Vader 1986; Franx \\& Illingworth 1990; Carollo \\& Danziger 1994; Zaritsky, Kennicutt, \\& Huchra 1994; Jablonka, Martin, \\& Arimoto 1996; J{\\o}rgensen, Franx, \\& Kjaergaard 1996; Pahre, Djorgovski, \\& de Carvalho 1998; Trager et al. 1998). Superwinds also offer a natural way to create a cosmically evolving population of large, metal-enriched, kinematically-complex gaseous halos, in many ways resembling the sharp metal lines and Lyman-limit systems observed in quasar spectra.  Strong evidence for spatially-resolved starburst-driven winds now exits in several galaxies (e.g., Bland \\& Tully 1988; Heckman et al. 1990; Lehnert \\& Heckman 1996; Marlowe et al. 1995; Meurer et al. 1992; Veilleux et al. 1994).  The case for spatially resolved outflows in AGN is equally strong (e.g., Cecil, Bland, \\& Tully 1990; Cecil, Wilson, \\& Tully 1992; Baum et al. 1993; Wilson et al. 1992; Veilleux, Tully, \\& Bland-Hawthorn 1993; Colbert et al. 1996a, 1996b, 1998) although here the exact mechanism responsible for the outflow is sometimes uncertain. Possible mechanisms include : (1) thermal wind from a circumnuclear starburst (e.g., Baum et al. 1993), (2) collimated radio jets (e.g., Wilson 1981; Taylor, Dyson, \\& Axon 1992), and (3) thermal wind produced by the AGN torus (e.g., Krolik \\& Begelman 1986; Balsara \\& Krolik 1993) or entrained along the radio ejecta (e.g., Bicknell et al. 1998).  Over the past ten years, our group has been conducting an optical survey of nearby active and starburst galaxies combining Fabry-Perot imaging spectrophotometry with radio and X-ray data to track the energy flow of galactic winds through the various gas phases (see reviews of recent results in Cecil 2000 and Veilleux 2000). The complete spatial and kinematic sampling of the Fabry-Perot data is ideally suited to study the complex and extended morphology of the warm line-emitting material which is associated with the wind flow. The radio and X-ray data complement the Fabry-Perot data by probing the relativistic and hot gas components, respectively.  The high level of sophistication of recent hydrodynamical simulations (e.g., Tomisaka 1990; Slavin \\& Cox 1992; Mineshige et al. 1993; Suchkov et al. 1994; Strickland \\& Stevens 2000) has provided the theoretical basis to interpret our data and to predict the evolution and eventual resting place (disk, halo, or intergalactic medium) of the outflowing material. In this paper, we present our results on the Seyfert galaxy NGC 2992.   The proximity of NGC~2992 ($z$ = 0.0078 or 31 Mpc assuming H$_0$ = 75 km s$^{-1}$ Mpc$^{-1}$; see \\S 4.2 below) provides a spatial scale (150 pc arcsec$^{-1}$) which permits a detailed study of the galactic outflow in this galaxy. The AGN host is a Sa galaxy whose high inclination ($i$ $\\approx$ 70$\\arcdeg$) is well suited for studies of the extraplanar emission. The galactic disk is rich in HI (Hutchmeier 1982) and CO (Sanders \\& Mirabel 1985) and is crossed by a disturbed dust lane oriented along P.A. $\\approx$ 30\\arcdeg.  Less than 3$\\arcmin$ to the south-east of NGC~2992 is the companion galaxy NGC~2993. The NGC~2992 -- NGC~2993 system (= Arp 245; Arp 1966) is connected by a HI tidal bridge and displays two spectacular tidal tails to the north of NGC~2992 and to the south-east of NGC~2993 (see, e.g., Plate 2 in Ward et al. 1980). The tidal forces associated with this galactic encounter have long been suspected to be the triggering mechanism of the nuclear activity in NGC~2992 (e.g., Burbidge et al. 1972; Osmer, Smith, \\& Weedman 1974; Ward et al. 1980).  The evidence for AGN activity in NGC~2992 is seen at nearly all wavelengths. The pioneering optical work of Burbidge et al. (1972) and Osmer et al. (1974) revealed a nuclear spectrum dominated by strong high-excitation emission lines. The detection of a broad (5,500 km s$^{-1}$) component in the profile of nuclear H$\\alpha$ and the identification of NGC~2992 with a strong HEAO-I X-ray source led Ward et al. (1978, 1980) to suggest the presence of a partially obscured AGN in this galaxy. The existence of a broad-line region (BLR) in this object has since been confirmed at optical wavelengths (e.g., Shuder 1980; V\\'eron et al. 1980; Durret \\& Bergeron 1988) and in the near-infrared (e.g., Rix et al. 1990; Goodrich, Veilleux, \\& Hill 1994; Veilleux, Goodrich, \\& Hill 1997). X-ray and HI absorption studies have confirmed the presence of a significant amount of absorbing material in front of the nucleus of NGC 2992 (N$_{\\rm H}$ $\\approx$ 10$^{22}$ cm$^{-2}$; Gallimore et al. 1999; Weaver et al. 1996 and references therein).  The presence of a galactic-scale outflow in NGC~2992 has been suspected for several years. The 20-cm radio emission from NGC~2992 extends approximately 25$\\arcsec$ (3.75 kpc) along the major axis of the galaxy, but also presents a one-sided 90$\\arcsec$ (13.5 kpc) extension along P.A. $\\sim$ 100$\\arcdeg$ -- 130$\\arcdeg$, i.e. close to the galaxy minor axis (Ward et al. 1980; Hummel et al. 1983). On smaller scales, a striking ``figure-8'' structure suggestive of limb-brightened bubbles or magnetic arches is seen at 6 cm extending along P.A. = 160$\\arcdeg$, about 3$\\arcsec$ (5$\\arcsec$) north (south) from a weak nuclear source (Ulvestad \\& Wilson 1984; Wehrle \\& Morris 1988). The nuclear source has a size $\\la$ 0$\\farcs$1 -- 0$\\farcs$3 (15 -- 45 pc) and appears to have a flat radio spectrum (Condon et al. 1982; Ulvestad \\& Wilson 1984; Unger et al. 1986; Sadler et al. 1995). A recent re-analysis of the ROSAT HRI data on NGC~2992 (Colbert et al. 1998) reveals that the soft (0.2 -- 2.4 keV) X-ray emission from this object is extended on a scale of 35 -- 45$\\arcsec$ (5.3 -- 6.8 kpc) along the galaxy minor axis. The X-ray nebula is therefore roughly cospatial with the large-scale radio emission in this galaxy, suggesting that both are produced by a large-scale galactic outflow. Supporting evidence for a galactic outflow also exists at optical wavelengths. The [O~III]-emitting gas is distributed into two conical regions roughly aligned along the galaxy minor axis (e.g., Durret \\& Bergeron 1987; Werhle \\& Morris 1988; Allen et al. 1999).  Long-slit optical spectroscopy (e.g., Heckman et al. 1981; Colina et al. 1987; M\\'arquez et al. 1998; Allen 1998) indicates that some of the line-emitting gas in NGC~2992 does not follow simple galactic rotation. Near-infrared, extended emission embedded within the northern radio loop was recently detected by Chapman et al. (2000) and deduced to be of non-stellar origin, adding support to the galactic outflow scenario.  However, several fundamental questions remain: What is the geometry of the optical outflow?  How is it oriented with respect to the plane of the host galaxy?  What are the energetics involved in the outflow? What is powering the outflow?  The limited spatial coverage of the long-slit data published so far cannot satisfactorily answer these questions. In this paper, we present the first complete two-dimensional velocity field of the line-emitting gas in NGC~2992. We choose to study the H$\\alpha$ emission rather than [O~III] to reduce the effects of internal dust extinction (important in this edge-on disk galaxy) and to constrain the kinematics of {\\em both} the galactic disk (bright in H$\\alpha$ but faint in [O~III]) and the outflowing material. Our paper is organized as follows. In \\S 2, we briefly describe the methods used to obtain and reduce our Fabry-Perot observations.  The technique used to decompose the emission-line profiles of NGC~2992 into disk and outflow components is discussed in \\S 3. The results of this analysis are discussed in \\S 4. In that section, we compare our Fabry-Perot data with published radio and X-ray maps, and constrain the three-dimensional kinematics of the outflow and galactic disk.  In \\S 5, we discuss the implications of our results on the nature and origin of the galactic wind in this active galaxy. We summarize our results in \\S 6 and discuss future avenues of research. ",
        "conclusions": "The complete two-dimensional coverage of our Fabry-Perot data has allowed us to decompose the circumnuclear emission-line region of NGC~2992 into two distinct kinematic components: a rotational component which lies in the plane of the galactic disk and an outflowing component which traces a biconical wind along the minor axis of the galactic disk. Outflow velocities ranging from 50 to 200 km~s$^{-1}$ are detected in the biconical region. The presence of two distinct kinematic components in one of the cones suggest that some (but probably not all) of the warm ionized material lies on the surface of the bicone.  The mass ($\\sim$ 1 $\\times$ 10$^7$ n$_{e,2}^{-1}$ M$_\\odot$) and kinematic energy ($\\sim$ 4 $\\times$ 10$^{54}$ n$_{e,2}^{-1}$ ergs) involved in this outflow are typical of other starburst- and AGN-driven outflows.  Evidence for a slight rotational component in some of the entrained material located in front of the galactic disk suggests that this material has been lifted from the disk or that emission from the underlying disk is contributing to this component.  The diffuse radio morphology and absence of direct evidence for a powerful nuclear starburst in this galaxy seem to rule out collimated radio jets or starburst-driven wind as the source of the outflow. By process of elimination, the most likely scenario is that of a thermally-dominated wind diverted along the galaxy minor axis by the pressure gradient of the ISM and powered on sub-kpc scale by (yet undetected) low-energy radio jets. This picture is consistent with the jet-driven outflow model of Colbert et al. (1996a), Bicknell et al. (1998) and Allen et al. (1999). A similar process was suggested by Veilleux et al. (1999) to explain the outflow in NGC~4388.  In NGC~2992, shock velocities of $\\sim$ 400 km~s$^{-1}$ would be sufficient to power the optical line emission from the entrained warm ionized material.  The currently available data on NGC~2992 do not allow us to distinguish between the jet-driven wind of Bicknell et al. (1998) and the torus-driven wind of Krolik \\& Begelman (1986). Deep radio maps on VLBI scale will be useful to look for the presence of a collimated pc-scale jet. Chandra data of this object would also be helpful to constrain the geometry, temperature and pressure of the X-ray emitting gas and hence the dynamical role of the hot thermal gas in the galactic outflow of NGC~2992. The shock velocities derived from the X-ray data could be directly compared with those necessary to ionize the ENLR clouds.  \\clearpage"
    },
    "0010/astro-ph0010244_arXiv.txt": {
        "abstract": "Gamma-ray bursts  with long durations are widely thought to arise from the  collapse of massive stars, where the wind environment is unavoidable. It is also believed that $\\gamma$-ray bursts come from jets. Considering these two points in this paper, we calculate the evolution of a highly collimated jet that expands in a stellar wind environment and the expected afterglow from such a jet. We use a set of refined dynamical equations and a realistic lateral speed of the jet, and find: (1) There is no observable break at the time when the  Lorentz factor of the jet is equal to the inverse of its initial half-opening angle. (2) No obvious break appears at the time when the blast wave transits from the relativistic to the non-relativistic phase. (3) For the wind case, there is no flattening tendency even up to $10^9$ s. (4) Compared with the homogeneous medium case, our calculated  flux is weaker in the stellar wind case. Finally, we find that two kinds of  GRB models (neutron star mergers and massive star collapses) may be discriminated  in our numerical results. ",
        "introduction": "Since the discovery of the GRB 970228 afterglow by BeppoSAX, research on gamma-ray bursts (GRBs) has evolved. Now we know that GRBs are one of the most energetic phenomena at the cosmological distance. Optical afterglows have been observed from about a dozen GRBs (Klose 2000), most of which concentrate at the distance scale of $z\\sim 1$, corresponding to the luminosity distance of about 3.0 Gpc. In addition, some GRBs' host galaxies have been discovered. All these discoveries  leave no doubt that GRBs, the origins of which had puzzled people since their discovery more than 30 years ago, are of cosmological origin.   Whether a jet exists or not in GRBs is a fundamental problem. We believe that jets should exist based on the following facts: (1) The isotropic energy release per GRB is generally in the range of $10^{51} - 10^{52}$ ergs. It can be explained well by a stellar-mass progenitor. However for two GRBs, GRB 990123 (Kulkarni et al. 1999a) and  GRB 990510 (Harrison et al. 1999), the isotropic energy is so enormous that it is difficult  to explain it by any stellar progenitor model, which forces some theorists to deduce that the radiation must be highly collimated in these cases. (2) The steepening of some afterglow light curves observed at the optical band  is argued as evidence that jets exist in GRB radiation (Kulkarni et al. 1999a; Harrison et al. 1999; Stanek et al. 1999; Huang et al. 2000a). Rhoads (1997a, 1999), Sari, Piran  \\& Halpern (1999), M\\'{e}sz\\'{a}ros \\& Rees (1999b) have shown that the lateral expansion of a relativistic jet will lead to a more rapid deceleration, causing a sharp break in the afterglow light curve. For GRB 990123, the power law index of the afterglow light curve is $\\alpha = 1.0 \\pm$ 0.03 in the 2 days after the burst. After 2 days, there is a sudden steepening in the light curve (Kulkarni et al. 1999b). Similarly, for GRB 990510 the power law index changes from $\\alpha$ = 0.76 to $\\alpha = 2.4 \\pm$ 0.02 after t = 1.0 day (Stanek et al. 1999). Recently, a rapid decay with $\\alpha =$ 1.73 was found in GRB 970228 (Galama et al. 2000). All these breaks may be due to the lateral expansion of jets (Huang et al. 2000a). (3) The observed radio flare may provide an independent and excellent indication of a jet-like geometry in GRBs (Harrison et al. 1999;  Kulkarni et al. 1999).  As argued by Waxman, Kulkarni and Frail (1998), the radio afterglow from a spherical fireball must rise to a peak flux on a timescale of a few weeks, but  because of lateral expansion, the radio afterglow from the forward shock of a jet must fade a few days after the burst. Therefore, the relative faintness of the observed late-time radio emission implies the existence of a jet (Sari \\& Piran 1998a). GRB 990510 may be a good example: its radio radiation began to decline one day after the burst (Harrison et al. 1999). (4) The polarization observed in the afterglows may also be evidence for jets. Gruzinov (1999) has argued that optical afterglows from jets can be strongly polarized, in principle up to tens of percents, if co-moving magnetic fields parallel and perpendicular to the jet have different strengths and if we observe the afterglows at a right viewing angle.  However, Sari (1999) argued that even if the magnetic fields had a well-defined orientation relative to the direction of the shock, the polarization is unlikely to exceed 20\\%. Furthermore, taking into account the dynamics of jets, the polarization first rises to the peak around the jet break time and then decays. (5) Observed light curves of some GRBs steepen simultaneously at  different bands. This may be further evidence that there is collimated ejecta in GRB radiation (Harrison et al. 1999). (6) Observational characteristics of  some GRBs are similar to those of BL Lacs, in which jets are unavoidably involved (Paczy\\'{n}ski 1993; Dermer \\& Chiang 1999; Cheng, Fan \\& Dai 1999). This implies that GRBs may arise from jets.  Generally, the broken light curves of some afterglows are explained to be due to the expansion of jets  into a homogeneous interstellar medium (Harrison et al. 1999; Stanek et al. 1999; Huang et al. 2000a). However,  other explanations have also been proposed. For example, when a spherical fireball evolves in the Wolf-Rayet star wind, the light curve can also  steepen  (Chevalier \\& Li 1999, 2000; M\\'{e}sz\\'{a}ros, Rees \\&  Wijers 1998; Frail et al. 1999a, b; Dai \\& Lu 1998a). Dai \\& Lu (1999, 2000a, b) suggested that when a shock in a dense medium transits from the relativistic phase to the non-relativistic phase, a break  would occur in the light curve.   The most important problem in GRB research is the energy mechanism. It is widely believed that GRBs with long durations come from the  collapse of massive stars (Fryer et al. 1998, 1999a, b; In't Zand 1998; Ruffert \\& Janka 1998, 1999; Woosley, MacFadyen  \\& Heger 1999; Rees  1999; M\\'{e}sz\\'{a}ros \\& Rees 1999a). For a massive star at the end of its evolution, it throws away its envelope and the core collapses into a compact object, producing a jet. Then due to interactions between different shells inside the jet and between the jet and its surrounding medium, a GRB and its afterglow are produced, respectively. The recently discovered connection between supernovae and GRBs provides a strong support to such a collapsar model. At present, there are three GRBs which are most likely connected with supernovae: GRB 980425 (SN 1998bw) (Galama et al. 1998; Iwamoto et al. 1998), GRB 970228 (Galama et al. 2000; Reichart  1999), GRB 980326 (Bloom et al. 1999). Recently, two other GRBs were added, i.e. GRB 970514 (SN 1997cy) (Germany et al. 2000; Turatto, Suzuki \\& Mazzali 2000) and GRB 980910a (SN 1999e) (Kulkarni \\& Frail 1999c; Thorsett \\& Hogg 1999).  Livio \\& Waxman (1999) recently discussed the evolution of a jet in the wind environment  and gave an analytical result. They argued that at late stages (particularly after the break corresponding to $\\gamma=1/\\theta_{\\rm o}$), the light curve has a  flattening tendency. In this paper, we use some refined equations to describe the evolution of jets. First, in the adiabatic case, when blast wave is  extremely  relativistic, its dynamical evolution satisfies the Blandford-Mckee (1976) solution. But when it reaches the non-relativistic phase, it satisfies the Sedov-Taylor self-similar solution. However, the conventional dynamical model can not transit correctly from the ultra-relativistic phase to the non-relativistic phase. This has been stressed by Huang et al. (1998a, b, 1999a, b). Here we use the refined dynamical equations proposed by Huang et al. (1999a, b, 2000b, c, d), which can describe the overall evolution of jets from the ultra-relativistic phase to the non-relativistic phase. Second, for the lateral expansion speed of jets, it is reasonable to assume that it is just the co-moving local sound speed $c_{\\rm s}$. Usually, one has taken $c_{\\rm s}=c$ or $c/\\sqrt{3}$ (Rhoads 1997a, b, 1999; Sari, Piran \\& Halpern 1999),  where $c$ is the speed of light. In fact we expect $c_{\\rm s}$ to vary with time, and especially it will by no means be $c$ or $c/\\sqrt{3}$  when the blast wave decelerates into the non-relativistic stage. Huang et al. (2000b, c, d) have given the proper lateral expansion speed which depends on the bulk speed of the blast wave.  Based on these considerations, we calculate the evolution of jets in the wind environment from the relativistic stage to the non-relativistic stage and compare numerical results with those of jet evolution in the homogeneous medium. We describe our model in section 2. Our detailed numerical results are presented in section 3. Section 4 is a brief discussion of our final results. ",
        "conclusions": "Following Huang et al. (2000a, b, c, d), who considered the evolution of a jet in a homogeneous ISM environment, we investigated the detailed dynamical evolution of jets and their afterglows for the wind case from the ultra-relativistic stage to the non-relativistic stage. Recently, Kumar \\& Panaitescu  (2000) considered the evolution of a jet in stratified media. Compared with their studies, our model is refined in the following aspects: (1)\tKumar \\& Panaitescu (2000) considered  the evolution of an adiabatic jet, while we study the expansion of a partially radiative realistic jet. Furthermore, the dynamics presented are applicable to both ultra-relativistic and Newtonian jets, so we could follow the overall evolution of a jet using a  set of differential equations. (2) Similarly, Kumar \\& Panaitescu (2000) did not consider the variation of sound velocity  with time. We describe the lateral expansion of jets with a refined and more reasonable sound speed expression, which varies with the bulk Lorentz factor (Huang et al. 2000b, d).  \\begin{figure} \\begin{picture}(100,170) \\put(0,0){\\special{psfile=fig5.ps angle=-90 hoffset=-10 voffset=175 hscale=45 vscale=55 }} \\end{picture} \\caption { The evolution of the time index of the afterglow light curve  at the R band. Parameters and line styles are the same as in  Fig.1.} \\end{figure}  In addition, we considered the evolution of electron distribution with time. Despite these differences, one of our results is similar to that of Kumar \\& Panaitescu (2000): there seems no observable break around the time of $\\gamma =1/\\theta_{\\rm 0}$, which  conflicts with Rhoads' (1999) and Livio \\& Waxman's (1999) expectations. Furthermore, we also find that:  1.   Livio \\& Waxman (1999) predicted that a light curve flattening would occur when the blast wave evolves into the non-relativistic stage. For the wind case, we calculated up to $10^9{\\rm s}$ (corresponding to a Lorentz factor of 1.05), when the  blast wave has completely evolved into the non-relativistic stage. We do not find any flattening tendency in the light curve.   2.  The transition from the ultra-relativistic phase to the non-relativistic phase is also very smooth; the expected obvious break does not appear. This is very similar to the behaviour of isotropic fireballs (Wijers, Rees, \\& M\\'{e}sz\\'{a}ros1997, Huang et al. 2000b).  3.  If we use the same parameters (except for the differences in number density for the wind case and for the homogeneous ISM), we find that  the flux density in the wind case is obviously weaker than that in the homogeneous ISM case. This property is consistent with  Chevalier \\& Li's (2000) result.  Two currently popular models for GRB progenitors are the mergers of compact objects (neutron stars or black holes) and  the explosions of massive stars. It is widely believed that GRBs produced by the former model occur in the ISM with density n $\\sim 1 {\\rm cm}^{-3}$ and  GRBs produced by the latter model occur in the preburst stellar wind enviroment with mass density $\\rho \\propto R^{-2}$. As argued by Chevalier \\& Li (2000) and Livio \\& Waxman (1999), both ISM and wind cases should show the same emission feature during the lateral spreading phase, and in particular on a timescale of days, the wind density is similar to typical ISM densities so that an interaction with the wind would give results that  are not different from the ISM case. If GRBs are beamed, thus, their optical afterglow emission could not be used to discriminate the massive prognitor model from the compact binary progenitor model. However, our numerical results show that their optical afterglow emissions are different, particularly several days after the burst. Thus, it may be used to discriminate the two models from each other, and further observations may verify our numerical results."
    },
    "0010/gr-qc0010035_arXiv.txt": {
        "abstract": "Inspired by the Randall-Sundrum brane-world scenario, we investigate the possibility of brane-world inflation driven not by an inflaton field on the brane, but by a bulk, dilaton-like gravitational field. As a toy model for the dilaton-like gravitational field, we consider a minimally coupled massive scalar field in the bulk 5-dimensional spacetime, and look for a perturbative solution in the anti-de Sitter (AdS) background. For an adequate range of the scalar field mass, we find a unique solution that has non-trivial dependence on the 5th dimensional coordinate and that induces slow-roll inflation on the brane. ",
        "introduction": "It is now widely accepted that our spacetime is not 4-dimensional but higher dimensional from the unified theoretical point of view. Horava and Witten showed the possibility that desirable gauge fields may consistently appear on the 10-dimensional boundary of $Z_2$-symmetric 11-dimensional spacetime \\cite{Horava}. A low-energy, 5-dimensional realization of the Horava-Witten theory was first discussed by Lukas et al. \\cite{lukas}. They also analyzed brane-world cosmology in the context of this 5-dimensional theory \\cite{lukascos1,lukascos2}.  Recently, Randall and Sundrum wrote two very interesting papers \\cite{rs1,rs2} on another possible low-energy realization of the Horava-Witten theory. In \\cite{rs1}, they found an interesting $Z_2$-symmetric solution of the 5-dimensional Einstein equations with a negative cosmological constant. In this solution, two boundary branes with positive and negative tensions are embedded in the 5-dimensional anti-de Sitter (AdS) space and the tensions of the branes are chosen so that the effective cosmological constant on the branes vanishes and the 4-dimensional Minkowski space is realized on the branes. They then showed that the mass-hierarchy problem in particle physics may be solved if we live on the negative tension brane. This work has received much attention from the particle physics community, and subsequently a large number of papers have been published on it. However, it was soon realized that there exists the so-called radion mode that describes fluctuations of the distance between the two branes, and this mode causes an unacceptable modification of the effective gravitational theory on the negative tension brane, unless the radion is stabilized rather artificially\\cite{tanaka,tX,rubakov}.  In the second paper\\cite{rs2}, Randall and Sundrum showed that the negative tension brane may actually be absent if we live on the positive tension brane. They found because of the curvature of AdS, even if the extra-dimension is infinite, gravity on the brane is nicely confined around the brane and the Einstein gravity is effectively recovered on the brane\\cite{rs2,tanaka,sms,gid}. As a result, although the hierarchy problem remains unsolved, this model has attracted much attention from the relativity/cosmology community, and the brane-world cosmology has boomed\\cite{flanagan,pierre,kraus,muko,ida,% stoica,kodama,david,bruck,gong,yu,koyama,kim,dorca,sasaki,luis,olsen,roy}.  To name a few, the Friedmann equation on the brane is discussed in \\cite{kraus,muko,ida,kim,dorca,sasaki,roy}. The quantum creation of a brane-world is discussed in \\cite{sasaki,luis}. The formulation and evolution of cosmological perturbations in the brane-world are discussed in \\cite{kodama,david,bruck,koyama,dorca}. And inflation on the brane is discussed by many people \\cite{sasaki,luis,olsen,kazuya,a,b,c,d,e,f,g,h}.  In almost all of these works, the 5-dimensional bulk spacetime is assumed to be vacuum except for the presence of the cosmological constant, and the matter fields on the brane are regarded as responsible for the dynamics of the brane. However, from the unified theoretic point of view, the gravitational action is not necessarily the Einstein-Hilbert action. In fact, string theory tells us that the dimensionally reduced effective action includes not only higher-order curvature terms but also dilatonic gravitational scalar fields. Thus at the level of the ``low-energy'' 5-dimensional theory, it is naturally expected that there appears a dilaton-like scalar field in addition to the Einstein-Hilbert action \\cite{lukas}. Thus it is of interest to investigate how such a scalar field in the 5-dimensional theory affects the brane-world. In this connection, very recently Maeda and Wands have investigated dilaton-gravity in the brane-world scenario \\cite{wands}.  In this paper, we investigate whether it is possible or not to have inflation on the brane solely by a bulk gravitational scalar field. Thus we consider a scenario of inflation on the brane without introducing an inflaton field on the brane.\\footnote{This type of inflation is called ``modular inflation'' in \\cite{lukascos2}.} A similar model of inflation in 5-dimensional dilatonic gravity was discussed recently by Nojiri, Obregon and Odintsov\\cite{NoObOd}. We model the dilaton-like bulk gravitational scalar by a minimally coupled scalar field. It is known that a scalar-tensor gravitational theory is conformally equivalent to the Einstein theory plus a minimally coupled scalar field\\cite{maeda}. So our model may be regarded as a conformally transformed scalar-tensor gravitational theory.  Very recently, the quantum fluctuations in a similar brane-world scenario has been discussed by Kobayashi, Koyama and Soda\\cite{shinpei}. They have shown that the extra-dimensional corrections due to the so-called Kaluza-Klein modes are small and the scenario is observationally acceptable just as the ordinary 4-dimensional scenario of inflation.  This paper is organized as follows. In Sec.~II, we first recapitulate the effective 4-dimensional gravitational equations on the brane and derive the Friedmann equation on the brane under the presence of a bulk scalar field. Then we discuss general conditions for inflation to occur on the brane by the bulk scalar field. In Sec.~III, we solve the Einstein-scalar field equations in the bulk perturbatively, and give explicit solutions that give rise to inflation on the brane. Finally, we summarize our results and discuss the implications in Sec.~IV. ",
        "conclusions": "We have investigated the possiblity of brane-world inflation driven solely a bulk gravitational scalar field.  First we have clarified general (sufficient) conditions for inflation to occur on the brane by analyzing the effective 4-dimensional Einstein equations. Namely, we have found that the standard slow-roll inflation can occur if the bulk scalar field is sufficiently slowly varying both in space and time near the brane.  Then, we have modeled the effective potential of the gravitational scalar field by an new inflation type potential, and looked for a spherically symmetric 5-dimensional solution perturbatively. In our model, the lowest order solution is given by the AdS bulk with the de Sitter brane as the boundary, and non-trivial behavior of the scalar field appears at the first order. We have found there exists a regular solution for the scalar field in the separable form with respect to $r$ and $t$, where $r$ is the 5th dimensional radial coordinate. The solution we have obtained gives slow-roll inflation on the brane and is found to be unique, provided that $|m^{2}|/H^{2}\\ll 1$, which is the same condition for slow-roll inflation as in the usual 4-dimensional theory.  Although our solution is valid only perturbatively, we expect a regular spherically symmetric solution to exist as long as the scalar field potential is sufficiently flat. Then, an immediate question is whether our model can give the standard inflationary quantum fluctuation spectrum. In this respect, very recently, Kobayashi, Koyama, and Soda have calculated quantum fluctuations of a 5-dimensional minimally coupled massless scalar field, and concluded that the Kaluza-Klein modes give negligible contributions to the quantum fluctuation spectrum induced on the brane\\cite{shinpei}. Hence we expect the same conclusion holds also in the present model.  However, the situation seems to change drastically at the end of inflation when the scalar field undergoes damped oscillations. As one can notice from the discussions given in Sec.~\\ref{solution}, the radial function can be regular at the origin only if $m^2<0$. Thus there exists no regular solution in the separable form if $m^2>0$. This implies that any regular solution must be in the form, $\\sum_\\gamma \\psi_\\gamma(t)u_\\gamma(r)$, where the sum extends over infinite numbers of $\\gamma$. Then there seems no apparent reason why the sum should extend only over functions of $t$ and $r$. In other words, it may be that an infinitesimal non-spherical perturbation (the so-called Kaluza-Klein mode) can grow indefinitely. This may cause a problem in our scenario. But at the same time, if the non-spherical perturbations saturate at certain level, this rather chaotic behavior may give rise to an efficient mechanism of reheating. Furthermore, the Kaluza-Klein excitations left in the subsequent universe may become a good candidate for the cold dark matter of the universe. Certainly, this issue deserves further studies.  Another remaining issue is the naturalness of the model parameters. If we assume $\\kappa_5^2V_0$ is of the same order of $\\Lambda_5$, we have $\\ell_{eff}\\gtrsim\\ell$. Since the location of the brane $r_0$ is expected to be greater than $\\ell$ by factor of a few at least, in order for our classical picture of the infinitely thin brane to be valid, $r_0$ cannot be smaller than $\\ell_{eff}$. Hence the condition $|m^2|/H^2<1$ implies $|m^2|\\ell_{eff}^2\\ll1$ (see Eq.~(\\ref{srcondition})). That is, the mass parameter in our theory must be fine-tuned to a very small value compared with the natural mass scale of the theory. This is the same problem one encounters in 4-dimensional models of inflation. However, if we recall that our model corresponds to a conformally transformed scalar-tensor theory, this problem may be solved in the original conformal frame. Investigations in this direction is also left for future work."
    },
    "0010/astro-ph0010558_arXiv.txt": {
        "abstract": "In an attempt to determine the nature of the enigmatic cocoon stars in the Quintuplet Cluster, we have obtained mid-infrared imaging and spectrophotometry of the cluster, using the CAM and SWS instruments on ISO, using SpectroCam-10 on the Palomar 5m telescope, and NICMOS on HST.  The spectra show smooth continua with various dust and ice absorption features. These features are all consistent with an interstellar origin, and there is no clear evidence for any circumstellar contribution to these features.  We find no spectral line or feature that could elucidate the nature of these sources.  Detailed modeling of the silicate absorption features shows that they are best reproduced by the $\\mu$ Cep profile, which is typical of the interstellar medium, with $\\tausil \\simeq 2.9$.  The high spatial resolution mid-IR images show that three of the five cocoon stars have spatially extended and asymmetric envelopes, with diameters of $\\sim 20,000\\,$AUs.  A reddening law similar to that of Lutz (1999) but with silicate features based on the $\\mu$ Cep profile and normalized to our value of $\\tausil$ is used to deredden the observed spectrophotometry.  The dereddened energy distributions are characterised by temperatures of 750--925 K, somewhat cooler than determined from near IR data alone.  Models of optically thin and geometrically thick dust shells, as used by Williams et al.~(1987) for very dusty, late-type WC stars, reproduce the observed SEDs from 4 to $17\\mic$, and imply shell luminosities of $\\log (L/\\Lsun) \\simeq 4.5$--4.9 for the brightest four components.  An analysis of the various suggestions proposed to explain the nature of the cocoon stars reveals serious problems with all the hypotheses, and the nature of these sources remains an enigma. ",
        "introduction": "The Quintuplet Cluster (AFGL 2004) is one of the three young clusters known in the vicinity of the Galactic Centre.  It was first identified in the near infrared survey of Glass, Catchpole, \\&\\ Whitelock (1987) as a bright source coincident with AFGL 2004, and found to be a bright IRAS source with a cool energy distribution (Glass 1988).  It is located about 30 pc from the Centre in projection, and its cluster nature became apparent with the advent of IR cameras through the works of Glass, Moneti, \\& Moorwood (1990; GMM90), Nagata et al.~(1990), Okuda et al.~(1990), Moneti, Glass, \\& Moorwood (1992), and Moneti, Glass, \\&\\ Moorwood (1994 [MGM94]), who identified over a dozen stars in the cluster, including five with extraordinarily large luminosities ($\\sim 10^5 \\Lsun$), cool (600---1200 K) energy distribution, and featureless $K$-band spectra.  These characteristics led those authors to tentatively identify them as massive, dust-enshrouded young stars, henceforth \\em cocoon \\rm stars.  The possibility that they be OH/IR stars was discarded since (i) they have no CO bandhead at 2.3 and $4.6\\mic$, (ii) they show little or no variability (Glass et al.~1999 find peak-to-peak of $\\sim 0.45 \\pm 0.05\\,$mag and K for two of them, no variability for the remainder), (iii) they lack OH maser emission, (iv) they are cooler and more luminous than typical OH/IR stars, and (v) they are much younger than OH/IR stars, assuming they are coeval with the cluster ($\\approx 4$ Myr, Figer McLean and Morris 1999 [FMM99], Figer et al.~1999).  More recently, FMM99 suggested that the cocoon stars might be extremely dusty, late-type WC stars (DWCL, Williams et al.~1987, WHT), even though they do not detect the expected emission features in the $J$-band.  And finally, Glass et al.~(1999) suggested they could be self-obscured, very massive O stars close to the ZAMS, which, according to Bernasconi and Maeder (1996), go through a prolonged accretion phase of 2--2.5 Myr, comparable to the age of the Quintuplet Cluster, before becoming visible.  In an attempt to elucidate the nature of the cocoon stars, we have obtained spectroscopy and spectrophotometry in the mid-IR, where these sources are brightest, in order to (i) search for spectral features intrinsic to the stars, (ii) obtain accurate measurements of their spectral energy distribution, (iii) to investigate their spatial extension, and (iv) to study the shape of the silicate feature in particular to determine whether it could be partly intrinsic to the sources. ",
        "conclusions": "New mid-infrared spectroscopic observations of the enigmatic cocoon stars in the Quintuplet Cluster were obtained in order to put further constraints on their nature.  The high spectral/low spatial resolution spectroscopy did not reveal any feature that could be assigned to the cocoon stars themselves.  The low spectral/high spatial resolution spectrophotometry was used to study the form of the silicate absorption in these stars, and we conclude that (i) the depth of the absorption is the same for all the stars, and (ii) there is no evidence for any silicate emission or absorption intrinsic to the cocoon stars.  In fact, all of the absorption features are consistent with an  origin in the molecular clouds and in the more tenuous ISM along the line of sight.  We have examined carefully the hypothesis that the cocoon stars are extreme DWCL stars.  The SEDs of the cocoon stars can be reproduced by the models of DWCL shells presented by WHT, but the high UV optical thickness needed to reprocess most of the luminosity of the central star suggests that these models might not be applicable to such extreme stars.  Furthermore, the luminosities derived are significantly lower than those of known DWCL stars, and also these stars are extremely rare in the Galaxy, though they are concentrated in the GC.  Yet, stars of the carbon sequence are necessary to produce featureless spectra.  The new observations do not provide any new constraints in favour of any of the proposed identifications of the cocoon stars but they do provide new arguments against all three of the main hypotheses that have generally been considered, namely dust-enshrouded youg stars, late-type (OH/IR or AGB) stars, or late-type WC stars.  The nature of these stars remains an enigma."
    },
    "0010/astro-ph0010072_arXiv.txt": {
        "abstract": "We report the discovery of a highly coherent oscillation in a type-I X-ray burst observed from \\src\\ by the {\\it Rossi X-ray Timing Explorer}\\/ (\\xte\\/).  The oscillation was most strongly detected $\\approx 1$~s after the burst onset at a frequency of 269.3~Hz, and it increased in frequency over the following 4 seconds of the burst decay to a maximum of $\\simeq$272~Hz. The total measured drift of $3.58\\pm0.41$~Hz ($1\\sigma$) represents the largest fractional change in frequency ($1.32\\pm0.15$~\\%) yet observed in any burst oscillation. If the asymptotic frequency of the oscillation is interpreted in terms of a decoupled surface burning layer, the implied neutron star spin period is around 3.7~ms.  However, the expansion of the burning layer required to explain the frequency drift during the burst is around 80~m, substantially larger than expected theoretically (assuming rigid rotation).  The oscillation was not present in the persistent emission before the burst, nor in the initial rise. When detected its amplitude was 6--12\\% (RMS) with a roughly sinusoidal profile. The burst containing the oscillation showed no evidence for photospheric radius expansion, while at least 5 of the other 9 bursts observed from the source by \\xte\\/ during 1996 and 1998 did.  No comparable oscillations were detected in the other bursts.  A pair of kilohertz quasi-periodic oscillations (QPOs) has been previously reported from this source with a mean separation of $348\\pm 12$~Hz.  \\src\\ is the first example of a source where the burst oscillation frequency is significantly smaller than the frequency separation of the kHz QPOs. ",
        "introduction": "Type I X-ray bursts are signatures of unstable nuclear burning on the surface of accreting neutron stars (see \\citealt*{lew95} and \\citealt{bil98a} for recent reviews).  Strong evidence for surface brightness anisotropies during the bursts comes from observations of highly coherent `burst oscillations' in 7 sources to date (\\citealt{stroh96}; see also \\citealt{vdk00} and references therein). These oscillations have been suggested to result from initially localized nuclear burning, which spreads over the surface of the neutron star during the early stages of the burst \\cite[]{stroh96}. As the burst evolves, the oscillations typically increase in frequency by a few Hz; in X1658$-$298 the increase is $\\approx5$~Hz, the greatest increase seen so far \\cite[]{wij00}.  This frequency drift may occur as a consequence of the burning layer becoming decoupled from the star \\cite[]{stroh97,cb00}.  In several sources the oscillation approaches an asymptotic frequency that may be the spin frequency $\\nu_{\\rm spin}$ of the neutron star.  Although many observational characteristics of burst oscillations seem to be consistent with this picture, a growing body of observations points to a substantially more complex underlying mechanism \\cite[e.g.][]{mill99,stroh00}.  In this {\\it Letter} we report the discovery of a new burst oscillation in \\src.  The low-mass X-ray binary (LMXB) \\src\\ ($l=31.4\\arcdeg$, $b=-8.5\\arcdeg$) was discovered by the {\\it Uhuru}\\/ satellite in 1977 \\cite[]{4ucat}. {\\it EXOSAT} observations revealed irregular X-ray dipping behavior with a period of $\\approx50$~min \\cite[]{walter82,white82}, which optical observations of the $m_V=21$ companion confirmed was approximately the orbital period \\cite[]{grin88}. Analysis of \\xte\\/ observations during 1996 shows that the  source is a member of the `atoll' class \\cite[]{bloser00,boirin00b}.  Both high ($\\ga 100$~Hz) and low frequency QPOs were detected in the power spectra of the source as measured by \\xte.  In particular, a pair of high-frequency QPOs (`kilohertz QPOs') with a mean separation of $348\\pm 12$~Hz was simultaneously detected on five occasions \\cite[]{boirin00b}.  On four of these occasions, the observed separation was consistent with the mean value, but on one occasion was only $290\\pm 5$~Hz.  X-ray bursts exhibiting photospheric radius expansion resulting from super-Eddington luminosities indicate a source distance of 8.4--10.8~kpc \\cite[]{smale88}. The X-ray bursts observed by \\xte\\ during 1996 were searched for high-frequency oscillations, but none were found to a $3\\sigma$ limit of 3\\% RMS amplitude \\cite[]{boirin00b}. Here we present the results of a more comprehensive timing study of X-ray bursts in \\src\\ observed by \\xte, including both the previously analyzed 1996 data and more recent observations made during 1998. ",
        "conclusions": "We have conclusively detected a highly coherent $\\simeq$270~Hz oscillation during a type I X-ray burst from \\src, bringing the total number of known burst oscillation sources to eight.  If we interpret the frequency evolution of the oscillation during the burst in terms of a decoupled burning layer, then the maximum observed frequency implies a neutron star spin period of $\\approx 3.7$~ms.  However, the large change in frequency during this burst ($3.58\\pm0.41$~Hz or $1.32\\pm0.15$~\\%) would require an $\\approx80$~m expansion of the burning layer (assuming rigid rotation). The maximum peak flux of all the bursts observed is a lower limit to the Eddington flux $F_{\\rm Edd}$ for the source, in which case the peak flux for the 1998 August 1 burst (in which the burst oscillation was observed) is at most $0.5 F_{\\rm Edd}$. The maximum expansion predicted for a mixed H/He burst at this flux level is only 20~m \\cite[]{cb00}, different from our measurement at the $7\\sigma$ level.  That the binary period of the source is shorter than 80~min indicates that the mass donor must be very hydrogen-poor \\cite[]{nrj86}, suggesting a predominantly He burning layer. Interestingly, \\cite{cb00} predict an even smaller expansion in that case. The limited sample available suggests that burst oscillations in \\src\\ are more likely to be present in atypically weak bursts which do not exhibit photospheric radius expansion (see Table \\ref{obs}).  The other two sources with $\\sim300$~Hz burst oscillations (4U~1728$-$34 and 4U~1702$-$429) also show this tendency, while the sources with higher frequency oscillations behave in a distinctly different manner (Muno et al. 2001, in preparation).    In the four burst sources from which both a kHz QPO pair and a burst oscillation were previously known, the burst oscillation frequency $\\nu_{\\rm \\,burst}$ is comparable to (or else nearly twice) the peak separation of the kHz QPOs $\\Delta\\nu$ (see Table~2).  Precise measurements of $\\Delta\\nu$ have been made for two of these sources, and these detailed measurements show clearly that \\begin{equation} \\nu_{\\rm \\,burst} \\gtrsim n\\,\\Delta\\nu \\mbox{\\rm\\ \\ with $n=1$ or 2.} \\end{equation} \\cite[]{mend98,mend99b}. For the other two sources, one can only conclude that $\\Delta\\nu$ and $\\nu_{\\rm \\,burst}$ are equal within their uncertainties.  The relationship between $\\nu_{\\rm \\,burst}$ and $\\Delta\\nu$ given in (1) has been interpreted in terms of a beat frequency model for the kHz QPOs (\\citealt{stroh96}; \\citealt*{mlp98}).  In the original sonic-point beat frequency model of \\cite{mlp98}, the upper kHz QPO appears at the Keplerian frequency of the accretion flow's sonic point, and the lower kHz QPO appears at the beat between this frequency and $\\nu_s$.  In this simple picture, we expect $\\Delta\\nu=\\nu_{\\rm spin}$.  More recent calculations by \\cite{lm00} have shown that the accreting material's inward drift decreases the upper kHz QPO frequency below the sonic point's Keplerian frequency and increases the lower kHz QPO frequency above the beat frequency, resulting in $\\Delta\\nu\\lesssim\\nu_{\\rm spin}$.  This is consistent with the observed relation (1) if we have $\\nu_{\\rm \\,burst}\\approx n\\,\\nu_{\\rm spin}$ with $n=1$ or 2, as in the decoupled burning layer model for the burst oscillations.  Our study of \\src\\ has revealed the first example of a source where $\\nu_{\\rm \\,burst} < \\Delta\\nu$ to high significance ($\\ga7\\sigma$ for the range of values of $\\Delta\\nu$ reported in \\citealt{boirin00b}).  It is unclear how to understand this result in terms of the sonic point model, which would require the frequency of the upper kHz QPO to be higher than the Keplerian orbital frequency in order to match our observation. The numerical simulations presented by \\cite{lm00} do not show this trend, but such a possibility cannot be excluded a priori.  A different explanation for the kHz QPOs involving relativistic precession of the accretion disk has also been proposed.  Here, the upper kHz QPO appears at the Keplerian frequency at some characteristic disk radius, and the lower kHz QPO occurs at the periastron precession frequency at the same radius \\cite[]{sv98,pn00}.  In this picture, $\\Delta\\nu$ is approximately equal to the radial epicyclic frequency and is not inherently related to $\\nu_{\\rm spin}$, so that the striking observed similarity between $\\nu_{\\rm \\,burst}$ and $\\Delta\\nu$ (or $2\\Delta\\nu$) is not explicitly addressed.  An additional model is therefore required to understand the burst oscillations in conjunction with the relativistic precession models for the kHz QPOs.  The requirements for such a model have been discussed, although none has yet been developed quantitatively.  If we assume that $\\nu_{\\rm \\,burst}\\approx \\nu_{\\rm spin}$, then it must be that $\\nu_{\\rm spin}$ is driven towards the maximum epicyclic frequency \\cite[]{stella99} or else that the kHz QPOs only reach detectable amplitudes when $\\Delta\\nu\\approx \\nu_{\\rm spin}$ or $\\Delta\\nu\\approx \\nu_{\\rm spin}/2$ due to a resonance effect \\cite[]{psa00} Alternatively, $\\nu_{\\rm \\,burst}$ might not be related to $\\nu_{\\rm spin}$ at all, but might instead be related to an accretion disk mode that occurs near the maximum epicyclic frequency at the innermost stable orbit \\cite[]{tlm98,pn00}, such as the $g$-modes predicted in the case of accreting black holes \\cite[see e.g.][]{kato90,nw91}. Unfortunately, the absence of a detailed link between the burst oscillations and the relativistic precession model makes it impossible to constrain this model using our results."
    },
    "0010/gr-qc0010021_arXiv.txt": {
        "abstract": "The evolution of an homogeneous and isotropic dissipative fluid is analyzed using dynamical systems techniques. The dissipation is driven by bulk viscous pressure and the truncated Israel-Stewart theory is used. Although almost all  solutions inflate, we show that only few of them can be considered as physical solutions since the dominant energy condition is not satisfied. ",
        "introduction": "There has been a renewed interest in the study of dissipative cosmologies related to the existence of inflationary solutions. As a matter of fact, dissipative processes are thought to be present in any realistic theory of the evolution of the universe. The simplest model of dissipative fluid is that which assumes the existence of bulk viscosity only. At a phenomenological level, bulk viscosity may be associated with particle production \\cite{Z-H},\\cite{barrow}. Also, qualitatively, bulk viscosity may be understood as the macroscopic expression of microscopic frictional effects that appear in mixtures \\cite{I-Z}. The easiest way to include bulk viscosity effects is through the Eckart's theory \\cite{eck}, which assumes that the bulk viscous pressure is proportional to the expansion. Several authors have used that theory to investigate the effects of viscosity on  cosmological models \\cite{viscos}.  As it is well known, however, Eckart's theory suffers important drawbacks \\cite{h-s} and a more complete theory  must be used. One of the proposed theories is the Israel-Stewart theory \\cite{is}   which complies with causality and stability. The non-linear terms of that theory are often neglected giving place to the so called truncated theory. Although the truncated theory may lead to a different behaviour as compared with the full theory, we shall for simplicity base our analysis on the truncated version of the Israel-Stewart approach. This truncated teory  was  firstly used by Belinskii et al. \\cite{b} in a cosmological context and after that by many authors ( see \\cite{Mr} and references therein). The  flat FRW models were studied in ref. \\cite{p} and has been extended recently to spatially curved solutions \\cite{p1}. In these last works the system of field equations was reduced to a second order differential equation and the stability of the stationary solutions was investigated by using a Lyapunov function. A different and more powerful approach was given by Coley and van den Hoogen \\cite{col} who used dynamical system techniques to analyze the asymptotic behaviour of the governing field equations. They used dimensionless equations of state that make the equation for the expansion to decouple from the system, obtaining a two dimensional system of differential equations.  Since these models were originally studied, it was soon realized that the type of equations of state used in them was  determinant in verifying whether the models underwent inflation. It is therefore important to study the effect that other reasonable equations of state will produce in the qualitative behavior of the solutions.  In this paper we analyze the asymptotic behaviour of  isotropic and spatially homogeneous models with a fluid with bulk viscosity. It has been pointed out \\cite{h-s} that although bulk viscosity is almost vanishing, in both the ultrarelativistic and the Newtonian regimes, its dynamical effects can not be neglected. On the other hand the evolution equations given by this type of matter can be used to model another kind of sources (e.g. the string dominated universe described by Turok \\cite{tur} or even a scalar field) The equations of state we assume are those introduced by Belinskii {\\it et al.} \\cite{b}. Unlike the case  studied in ref. \\cite{col}, now the expansion does not decouple and we should  deal with the complete three dimensional system, except for a particular value of the parameters. It is important to stress that the choice of the equations of state could  dramatically change the behaviour of the model and, therefore, it is worthwhile to study the implications of  different reasonable equations of state. ",
        "conclusions": "In this paper we have analyzed the qualitative behaviour of a particular type of dissipative fluids in FRW universes using the truncated Israel-Stewart theory. There have been a previous \\cite{p} analysis of the same system and with the same equations of state but only for the zero curvature FRW universes that was thereafter extended to the general case \\cite{p1}. In those analysis the system of equations was reduced to a second order differential equation whose stationary solutions were studied by using a Liapunov function. That method does not discriminate, however, between different equilibrium points. We have seen in this paper that when $m\\neq 1/2$ almost all the equilibrium points have $H=0$ and all of them were reduced to only one stationary solution in the mentioned papers.  An important feature that emerges from our analysis is that for most of the equilibrium points either the dominat energy condition or the positivity of the energy density are not satisfied. Solutions evolving towards these equilibrium points, therefore, should be ruled out as unphysical. That means that in order to obtain physical solutions we have to be very careful with both the initial conditions and the values of the parameters.  With respect to inflation, we see that, generically, the presence of bulk viscosity makes the solutions enter into an inflationary regime. Only few solutions do not inflate.  If we consider the fluid interpretation of our system of equations, we have seen that almost all the solutions evolve in such a way that they go outside  the regime in which the theory is valid (i.e. when $|\\sigma | \\ll p$). The only exceptions are those solutions which evolve towards the point $P_5$ when $m>1/2$ which describes an inflating universe when $\\gamma<2/3$ or solutions evolving towards $P^-$ with $m=1/2$ and $\\alpha$ very small.  As mentioned before we have used the truncated equation (5), instead of the more general transport equation derived from the Israel-Stewart theory. As it has been shown \\cite{m}, this latter equation may be written as \\begin{equation} \\tau_{*} \\dot\\sigma + \\sigma = - 3 \\xi_{*} H \\left[1+\\frac{1}{\\gamma c_b^2 }\\left(\\frac{\\sigma}{\\rho}\\right)^2\\right] \\label{A} \\end{equation} with \\begin{equation} c_b^2 = \\frac{\\xi}{(\\rho + p) \\tau} \\end{equation} and \\begin{equation} \\tau_* = \\frac{\\tau}{1 + 3 \\gamma \\tau H} \\; ; \\; \\xi_* = \\frac{\\xi}{1 + 3 \\gamma \\tau H}. \\end{equation} Therefore, close to equilibrium ($\\sigma << \\rho$), equation (\\ref{A}) leads to the truncated equation (5) with reduced relaxation time ($\\tau_*$) and bulk viscosity ($\\xi_*$) \\begin{equation} \\tau_{*} \\dot \\sigma + \\sigma = - 3 \\xi_{*} H. \\label{B} \\end{equation} The ammount of reduction depending on the size of $\\xi$ relative to $H$. Obviously if $\\tau H << 1$, then $\\tau_* \\approx \\tau$ and $\\xi_* \\approx \\xi$. However if $\\tau H$ is close to $1$, the reduction could be significant. On the other hand, qualitative changes in the results, outside the quasi-equilibrium regime, are difficult to forecast without considering specific models.  It is important to realize that the system of  equations studied in this paper, not only describes the evolution of a dissipative fluid with bulk viscosity, but also the evolution of a different kind of matter. It is well konwn, for instance, that when $m=2/3$ the above equations give the evolution of a string dominated universe \\cite{tur}. We shall see that this equivalence can be extended to a scalar field as well.  Indeed, a scalar field in both FRW and Bianchi type solutions is equivalent to a perfect fluid with $p$ and $\\rho$ defined as: \\begin{equation} \\rho=\\frac{1}{2}\\dot\\phi^2+V(\\phi),\\qquad p=\\frac{1}{2}\\dot\\phi^2-V(\\phi), \\end{equation} where $V$ is the scalar field potential. We can substitute in the above expression $p$ by $p+\\sigma$. If at the same time we consider a barotropic equation of state between the energy density and the equilibrium pressure of the form $p=(\\gamma-1)\\rho$ then we obtain that the viscous pressure is given by: \\begin{equation} \\sigma=\\frac{2-\\gamma}{2}\\dot\\phi^2-\\gamma V. \\end{equation} Taking derivatives with respect time and substituting into the Klein-Gordon equation of the scalar field \\begin{equation} \\ddot\\phi+3H\\dot\\phi+\\frac{dV}{d\\phi}=0 \\label{pul} \\end{equation} we obtain that $\\sigma$ satisfies an equation like (5) with  $\\tau$ and $\\xi$ given by \\begin{equation} \\tau=\\frac{1}{6H}, \\qquad \\xi=\\frac{1}{3H}\\left( \\gamma V+\\frac{1}{3H}\\frac{dV}{d\\phi}\\dot\\phi\\right). \\label{ul} \\end{equation}  Now, in dissipative systems there exists a peculiar state called ``critical point'', which corresponds to a specific value of a combination of thermodynamic variables. It has been shown \\cite{HeMa}, \\cite{HeDP} (and references therein) that a dissipative system in the critical point immediately (i.e. on a time scale of the order of the relaxation time) after its departure from equilibrium, behaves as if the effective inertial mass of any fluid element vanishes.  In the case of pure bulk viscosity the critical point is characterized by \\begin{equation} \\frac{\\xi}{\\tau}=2(\\rho+p). \\label{cp} \\end{equation} In this case however the critical point is forbidden by causality and stability requirements, demanding \\begin{equation} \\frac{\\xi}{\\tau}<(\\rho+p)\\left(1-c_s^2\\right) \\label{con} \\end{equation} where $c_s$ is the speed of sound (see \\cite{HeMa}).  Nevertheless in order to see what the critical point may imply in terms of a scalar field, let us assume (\\ref{cp}). Then, combining with (\\ref{ul}), we obtain after an elementary algebra \\begin{equation} \\frac{1}{3H}\\,\\frac{dV}{d\\phi}=\\frac{\\gamma \\dot{\\phi}}{2} \\label{ec} \\end{equation} and feeding back into (\\ref{pul}) \\begin{equation} \\ddot\\phi+3H\\dot\\phi \\left(1+\\frac{\\gamma}{2}\\right)=0. \\label{dp} \\end{equation} A simple integration of (\\ref{dp}) gives \\begin{equation} \\dot\\phi \\sim e^{-\\int{3H(1+\\gamma/2)dt}}. \\label{in} \\end{equation}  Let us consider two examples: \\begin{itemize} \\item de Sitter case, $H=$constant.  In this well known case, one gets from (\\ref{pul}) \\begin{equation} \\dot\\phi \\sim e^{-3H(1+\\gamma/2)t} \\end{equation} and therefore \\begin{equation} \\phi \\sim \\dot\\phi. \\end{equation} Then (\\ref{ec}) yields \\begin{equation} V \\sim \\phi^2. \\end{equation} This kind of potential is one of the most favoured by particle theorists, since it describes renormalizable particle theory \\cite{Ry}. \\item $H \\sim \\frac{2}{3\\gamma t}$  The asymptotic behaviour close to the point $P_5$ described above (see eq.(37)), yields, using (\\ref{in}) \\begin{equation} \\dot\\phi \\sim \\phi^{(1+\\gamma/2)} \\end{equation} and therefore the corresponding potential is \\begin{equation} V \\sim \\phi^{\\gamma+2}. \\end{equation} \\end{itemize} Thus, the critical point condition (\\ref{cp}), leads, for an scalar field interpretation of the source term, to polinomial potentials (for the two examples considered).  It is evident, therefore, that the constraints imposed by the fluid interpretation can be droped if  we consider the energy momentum tensor to correspond to another kind of matter (different from the fluid with bulk viscosity).  In such a case the only warnings to bear in mind are those coming from the dominat energy condition requirement."
    },
    "0010/astro-ph0010585_arXiv.txt": {
        "abstract": "We have obtained adaptive optics images and accurate radial velocities for 7 very low mass objects, In the course of a long term effort to determine accurate masses for very low mass stars (M$<$0.6~\\Msol). We use the new data, together with measurements from the litterature for some stars, to determine new or improved orbits for these 7 systems. They provide masses for 16 very low mass stars with accuracies that range between 0.2\\% and 5\\%, and in some cases a very accurate distance as well. This information is used in a companion paper to discuss the Mass-Luminosity relation for the V, J, H and K photometric bands. ",
        "introduction": "Accurate masses for binary stars provide a crucial test of our understanding of stellar physics (e.g. Andersen \\cite{andersen91}). Mass, the basic input of evolutionary models, is directly measured, and the models must reproduce the effective temperatures and luminosities (or radii) of both components, for a single age and a single chemical composition. Given the strong mass dependency of all stellar parameters however, this discriminating diagnostic only shows its power for relative mass errors $\\leq$1-3\\%.  Very accurate mass measurements have long been the exclusive province of double-lined detached eclipsing binaries (Andersen \\cite{andersen91}, \\cite{andersen98}), for which as a bonus the stellar radii are simultaneously determined. Such systems are however unfortunately rare, and only 44 pairs had yielded masses accurate enough to be included in Andersen \\cite{andersen91}'s critical compilation, mostly for intermediate mass stars. Relatively few eclipsing binaries have had their masses measured since then. In the mass range of interest here, the litterature still contains no more than three well detached eclipsing binaries with substantially subsolar component masses: YY~Gem (M0Ve, 0.6+0.6\\Msol; Bopp \\cite{bopp74}; Leung \\& Schneider \\cite{leung78}), the recently identified GJ~2069A (M3.5Ve, 0.4+0.4\\Msol; Delfosse et al. \\cite{delfosse99a}), and CM Dra (M4Ve, 0.2+0.2\\Msol; Lacy \\cite{lacy77}; Metcalfe et al. \\cite{metcalfe96}).  Angularly resolved spectroscopic binaries provide stellar masses in parts of the HR diagram where eclipsing systems are rare or missing, and in particular for very low mass stars. Until recently however, these measurements did not match the $\\sim$1\\% accuracy which can be obtained in detached eclipsing systems. As a consequence, the best representation to date of the empirical M-L relation for M dwarfs had to mostly rely on masses determined with 5-20\\% accuracy (Henry \\& McCarthy \\cite{henry93}; Henry et al. \\cite{henry99}). The last two years have seen a dramatic evolution in this respect, with two groups breaking through the former $\\sim$5\\% accuracy barrier. The first group to do so used the 1~mas per measurement astrometric  accuracy of the Fine Guidance Sensors (Benedict et al. \\cite{benedict99}) on {\\it HST} to determine a few masses of angularly resolved binaries with 2 to 10\\% accuracy (Franz et al. \\cite{franz98}; Torres et al. \\cite{torres99}; Henry et al. \\cite{henry99}; Benedict et al. \\cite{benedict00}). Slightly more recently, we have demonstrated that the combination of very accurate radial velocities with angular separations from adaptive optics imaging can yield masses for VLMS with even better accuracy, of only 1-3\\% (Forveille et al. \\cite{forveille99}; Delfosse et al. \\cite{delfosse99b}). Here we present 16 new or improved masses determined with the same method, with accuracies that now range between 0.2 and 5\\%. In a companion paper (Delfosse et al. \\cite{delfosse00a}) we rediscuss the VLMS mass-luminosity relation in the light of these new data. We first discuss the observing program and its sample in Sect.~2 and then present the observations and data processing in Sect.~3. Section~4 describes the orbit adjustment and the mass determination. ",
        "conclusions": "We have presented here mass measurements for 16 VLMS, with accuracies which range between 0.5 and 5\\%, and for masses down to below 0.1\\Msol. We will shortly publish a few additional masses whose derivation involve long baseline interferometric measurements % (S\\'egransan et al., in prep.). These results only represent a snapshot of  the progress of our mass measurement program: we continue to monitor the VLMS binaries with high accuracy radial-velocity observations and adaptive optics imaging, and have started using long baseline interferometry. Most of the masses which we present here will be improved in the future, and additional ones with similar accuracies will become available.  In a companion paper (Delfosse et al. \\cite{delfosse00a}), we show that these masses provide an impressive validation of the theoretical infrared M-L relations of Baraffe et al. (\\cite{baraffe98}), but point towards low level ($\\sim$0.5~mag) deficiencies of these models in the V band.  A logical next step for this type of work is the determination of accurate masses for even fainter objects, the very late M dwarfs and the L dwarfs. There is at present a large effort in determining the mass function across the stellar/substellar boundary in young open clusters (Bouvier et al. \\cite{bouvier98}, Zapatero Osorio et al., in prep.) and in the field (Reid et al. \\cite{reid99}, Delfosse \\& Forveille \\cite{delfosse00}). These programs need an accurate calibration of the mass as a function of both luminosity and age (due to the dominant effect of cooling for brown dwarfs). Up to now such relations are only available from models, which unfortunately meet with new difficulties for temperatures lower than were relevant in this paper, as dust condenses in the atmospheres of very cool dwarfs.    Until very recently, no binary of such mass was known with a period short enough for a mass determination over any realistic time scale. A few are now known (Mart\\'{\\i}n et al. \\cite{martin99}, \\cite{martin00}, Koerner et al. \\cite{koerner99}), even though their periods are either shorter or longer than would be ideal for a quick and accurate mass measurement. They will eventually provide mass determinations for brown dwarfs, but additional efforts to find more brown dwarfs binaries, and better suited ones, are certainly more than warranted. Several groups are doing this, following up the late-M and L dwarfs discovered by the SLOAN, 2MASS and DENIS surveys."
    },
    "0010/astro-ph0010066_arXiv.txt": {
        "abstract": "In a companion paper, Maiolino et al. (2000) presented various observational evidences for ``anomalous'' dust properties in the circumnuclear region of AGNs and, in particular, the reduced $\\rm E_{B-V}/N_H$ and $\\rm A_V/N_H$ ratios, the absence of the silicate absorption feature in mid-IR spectra of Sy2s and the absence of the carbon dip in UV spectra of reddened Sy1s. In this paper we discuss various explanations for these facts.  The observational constraints favor a scenario where coagulation, catalyzed by the high densities in the circumnuclear region, yields to the formation of large grains. The resulting extinction curve is featureless, flatter than Galactic and the $\\rm E_{B-V}/N_H$ and $\\rm A_V/N_H$ ratios are significantly reduced. These results should warn about an unappropriate use of the standard Galactic extinction curve and $\\rm A_V/N_H$ ratio when dealing with the extreme gas conditions typical of the circumnuclear clouds of AGNs.  We also investigated alternative scenarios for the observed anomalous properties of dust in AGNs. Some of these scenarios might explain some of the observed properties for a few objects, but they generally fail to account for all of the observational constraints obtained for the large sample of AGNs studied in these works. ",
        "introduction": "Some authors in the past have found evidences for anomalous properties of the dust in the circumnuclear region of AGNs. In particular, dust absorption and reddening appear to be systematically lower than  expected from the gaseous column density, for a Galactic gas-to-dust ratio and extinction curve (eg. Maccacaro et al. 1982, Reichert et al. 1985, Granato et al. 1997). In a companion paper (Maiolino et al. 2000, paper I) we have strengthened these previous results and provided additional evidences that the dust in the circumnuclear region of AGNs is characterized by anomalous properties. By comparing the reddening toward the Broad Line Region (BLR) and the gaseous N$_H$ inferred from the X-rays in a sample of 20 AGNs we found that, with the exception of three Low Luminosity AGNs, the $\\rm E_{B-V}/N_H$ ratio is lower than the Galactic standard value by a factor ranging from $\\sim$3 to $\\sim$100. By comparing the optical and X-ray properties of various AGN classes  (type 1.8--1.9 Sys, hard X-ray selected and radio selected QSOs, Broad Absorption Line QSOs and grism-selected QSOs) it was inferred that the $\\rm A_V/N_H$ ratio must be one or two orders of magnitude lower than Galactic. Despite the large mid-IR continuum absorption inferred from the PAHs equivalent width, the absence of a significant silicate absorption feature at 9.7$\\mu$m in the average ISO spectrum of Sy2s  indicates that dust grains in the circumnuclear region of AGNs have different properties with respect to the Galactic diffuse interstellar medium. Finally, the lack of the 2175\\AA \\ carbon dip in the UV spectra of some type 1 AGNs, which appear affected by dust reddening, suggests a depletion of small grains in the absorbing medium, as mentioned in paper I. These findings suggest that the dust grain distribution is biased in favor of large grains. The presence of large grains in the circumnuclear region of AGNs was also suggested by Laor \\& Draine (1993) to explain some of the infrared properties of AGNs.  In this paper we examine some possible explanations of the observational findings presented in paper I. In particular, we compare our models with the quantitative constraints given on the $\\rm E_{B-V}/N_H$ ratio, the equivalent width of the silicate feature and the depth of the 2175\\AA \\ carbon dip. Note that we often distinguish the $\\rm E_{B-V}/N_H$ and $\\rm A_V/N_H$ ratios since, when discussing extinction curves different from Galactic, reddening and absorption are no longer tied by the standard Galactic relation $\\rm A_V/E_{B-V}=3.1$. In other words, models which could account for a decreasing of $\\rm E_{B-V}/N_H$ do not necessarily lower the $\\rm A_V/N_H$ ratio, and vice versa.  In Section 2 we first analyze and, in practice, dismiss scenarios with a standard (Galactic) grain size distribution. The case for large grains is analyzed and discussed in Sect. 3 while in Sect. 4 we draw our conclusions. ",
        "conclusions": "We have discussed various interpretations of the observational evidences for ``anomalous'' properties of dust in the circumnuclear region of AGNs and, in particular, the reduced $\\rm E_{B-V}/N_H$, the reduced $\\rm A_V/N_H$, the absence of the silicate absorption feature in the mid-IR spectra of Sy2s and the absence of the carbon dip in the UV spectra of reddened Sy1s, which are reported in a companion paper (Maiolino et al. 2000, paper I).  A dust distribution dominated by large grains as a consequence of coagulation is the interpretation which best fit the various observational constraints. The formation of large grains is naturally expected in the high density environment characterizing the circumnuclear region of AGNs. A dust grain distribution biased in favor of large grains makes the extinction curve flatter than Galactic and featureless (especially for what concerns the silicate mid-IR feature and the UV carbon dip). Coagulation also yields a reduced $\\rm E_{B-V}/N_H$ and $\\rm A_V/N_H$.  It should be noted that nearly all of the papers, published so far, dealing with the effects of dust in AGNs have adopted a standard Galactic extinction curve (essentially with the only exception of Laor \\& Draine 1993, who first proposed the large grain scenario). The assumption that the properties of dust in the extreme conditions of the circumnuclear region of AGNs are similar to the Galactic diffuse interstellar medium is somewhat naive, given that even in the denser clouds of our Galaxy the extinction curve is already different with respect to the diffuse medium. Probably, many of the past studies considering the effects of dust obscuration and scattering in AGNs should be revised at the light of the findings obtained in this paper.  With regard to the scattered radiation (especially in obscured AGNs) it is worth noting that a dust distribution dominated by large grains would yield a nearly gray reflection, thus mimicking the scattering by free electrons. Large grains produced by dust coagulation would have a reduced scattering efficiency with respect to the standard Galactic dust mixture. However, while in some cases coagulation can result in a reduction of the scattering efficiency by more than an order of magnitude, for some of the dust grain distributions producing a flat extinction and scattering curve the reduction in the scattering efficincy is much lower (e.g. for the model discussed in 3.2.2 the scattering efficiency is reduced by a factor of $\\sim$2). Therefore, it is possible that some of the gray--reflected spectra found in some obscured AGNs, and ascribed to electron scattering, might actually be produced by large grains scattering.  We have also examined other possible interpretations for the ``anomalous'' dust properties discussed in paper I, and more specifically:\\\\  1) a metallicity higher than solar (as often found in the central region of AGNs) might affect the estimate of the equivalent N$_H$ from the photoelectric cutoff in the hard X-rays;\\\\ 2) the observed broad lines might be scattered by a reflecting medium through a path of much lower gaseous column than the X-rays;\\\\ 3) reduction of the scattering component of the extinction curve as a consequence scattering of the radiation into our beam;\\\\ 4) the BLR might have a size comparable to that of the absorbing medium (at variance with what assumed by the standard unified model) and, therefore, suffer less absorption with respect to the nuclear X-ray source;\\\\ 5) the hard X-rays might be partly absorbed by the same clouds which produce the broad lines;\\\\ 6) the sublimation of dust at the inner face of the obscuring torus would produce a gaseous dust-free region;\\\\ 7) the absorbing medium might simply be characterized by a low dust--to--gas ratio (in terms of mass).  Although some of these cases might apply to some of the objects examined in paper I, none of them can fit all of the observational constraints for most of the objects, at variance with the large grains scenario."
    },
    "0010/astro-ph0010299_arXiv.txt": {
        "abstract": "We present an analysis of 40 cool helium-rich white dwarfs found in the Hamburg/ESO survey. They were selected for follow-up spectroscopy because of their $U-B$ colour below -0.18, the absence of strong absorption lines, and a continuum shape similar to that of a quasar. Effective temperatures for individual stars were determined by fitting model atmospheres of nearly pure helium with a small admixture of hydrogen. As a consequence of the selection criteria all but one sample stars have \\Teff\\ below 20000\\,K. Four stars clearly show helium and hydrogen lines in their spectra. In the spectra of another three, helium, hydrogen, and metal lines can be detected. For these stars hydrogen and metal abundances were also determined by fitting appropriate model atmospheres containing these elements. Seven sample stars most likely have helium-rich atmospheres but do not show any helium lines. They either have featureless spectra or show calcium lines. ",
        "introduction": "\\begin{table} \\caption[]{Journal of observations} \\label{obslog} \\begin{tabular}{llllllllr} \\hline\\noalign{\\smallskip} Date     & Tel.    & Instr./CCD   & Grism & Resol. \\\\ &         &              & Grat. & [\\AA ] \\\\[0.1em] \\hline\\noalign{\\smallskip} Dec 1990 & 3.6\\,m  & EFOSC1/RCA   & B300   & 26 \\\\ Apr 1991 & 3.6\\,m  & EFOSC1/RCA   & B300   & 26 \\\\ Feb 1992 & 2.2\\,m  & EFOSC2/Tek   & \\#\\,1  & 40 \\\\ Feb 1992 & 3.6\\,m  & EFOSC1/Tek   & B300   & 16  \\\\ Apr 1992 & 3.6\\,m  & EFOSC1/Tek   & B300   & 19 \\\\ Sep 1992 & 3.6\\,m  & EFOSC1/Tek   & B300   & 21  \\\\ Mar 1993 & 3.6\\,m  & EFOSC1/Tek   & B300   & 21  \\\\ Mar 1993 & 1.52\\,m & B\\&C/LORAL & \\#\\,2  & 15 \\\\ Feb 1994 & 1.52\\,m & B\\&C/LORAL & \\#\\,16 & 8 \\\\ Sep 1994 & 3.6\\,m  & EFOSC1/Tek   & B300   & 21 \\\\ Nov 1994 & 1.52\\,m & B\\&C/LORAL & \\#\\,13 & 15  \\\\ Oct 1995 & 1.52\\,m & B\\&C/LORAL & \\#\\,13 & 15  \\\\ Mar 1996 & 1.52\\,m & B\\&C/\\#\\,39 & \\#\\,13 & 15 \\\\ Oct 1996 & 1.52\\,m & B\\&C/\\#\\,39 & \\#\\,13 & 15 \\\\ Oct 1997 & 1.52\\,m & B\\&C/\\#\\,39 & \\#\\,13 & 15 \\\\ Sep 1998 & 1.52\\,m & B\\&C/\\#\\,39 & \\#\\,13 & 15 \\\\ Nov 1998a & 1.54\\,m & DFOSC/LORAL & \\#\\,4  & 10 \\\\ Nov 1998b & 1.54\\,m & DFOSC/LORAL & \\#\\,11  & 15 \\\\ Nov 1998c & 1.54\\,m & DFOSC/LORAL & \\#\\,7  & 5 \\\\[0.1em] \\hline \\end{tabular} \\end{table}  \\begin{table*} \\caption[]{ Coordinates, magnitudes, cross-references and spectral types, if already existing, for the observed objects. Cross-references were taken from the Simbad database and McCook \\& Sion (\\cite{mccook}). Most of the EC objects are listed in Kilkenny et al. (\\cite{kilkenny97}). Details of individual observing campaigns can be found in Table~1. The survey fields, given in column ``field'', are ESO/SERC atlas fields. The $B_J$ magnitudes are accurate to better than 0\\fm 2. \\label{objects}} \\begin{tabular}{lllllrcccc}\\hline\\noalign{\\smallskip} Name & Other Names & Type & \\multicolumn{2}{c}{Campaign} & \\multicolumn{1}{c}{$t$ [s]} & Field & $\\alpha$(2000) & $\\delta$(2000) & $B_J$\\\\[0.1em] \\hline\\noalign{\\smallskip} HE 0004$-$1452  & PHL665             &         & 98a & Nov 1.54m & 1800 & 607 & 00 06 36.0 & $-$14 36 14 & 17.1\\\\ HE 0052$-$2511  & CT195               &         & 95  & Oct 1.52m &  900 & 474 & 00 54 46.7 & $-$24 55 29 & 17.9\\\\ HE 0104$-$3056  & $^c$                &         & 94  & Sep 3.6m  &  300 & 412 & 01 07 12.4 & $-$30 40 16 & 17.3\\\\ HE 0108$-$3036  &                     &         & 92  & Sep 3.6m  &  300 & 412 & 01 11 06.3 & $-$30 20 34 & 17.7\\\\ HE 0122$-$2244  & KUV01223-2245       &         & 97  & Oct 1.52m &  300 & 476 & 01 24 44.6 & $-$22 29 07 & 16.8\\\\ HE 0127$-$3110  & BPM47178, GD1363    & DAH$^a$ & 98a & Nov 1.54m &  900 & 413 & 01 29 56.2 & $-$30 55 08 & 16.0\\\\ &                     &         & 92  & Sep 3.6m  &  300 &     &            &             & \\\\ &                     &         & 98c & Nov 1.54m & 1800 &     &            &             & \\\\ HE 0149$-$2518  & $^b$                &         & 97  & Oct 1.52m &  600 & 477 & 01 51 59.5 & $-$25 03 15 & 17.1\\\\ HE 0200$-$5556  &                     &         & 96  & Oct 1.52m &  300 & 153 & 02 01 57.9 & $-$55 42 16 & 17.5\\\\ HE 0203$-$5013  & JL281$^d$           &         & 96  & Oct 1.52m &  300 & 197 & 02 05 03.1 & $-$49 59 03 & 17.0\\\\ HE 0308$-$1313  & $^c$                &         & 98b & Nov 1.54m &  300 & 616 & 03 11 07.2 & $-$13 01 53 & 17.1\\\\ HE 0309$-$2105  & EC03097$-$2105        &         & 98c & Nov 1.54m & 1000 & 547 & 03 11 58.1 & $-$20 54 09 & 15.6\\\\ HE 0319$-$4344  &                     &         & 95  & Oct 1.52m &  300 & 248 & 03 21 10.1 & $-$43 34 00 & 17.7\\\\ HE 0413$-$3306  &                     &         & 92  & Feb 2.2m  &  300 & 360 & 04 15 20.6 & $-$32 59 11 & 15.9\\\\ HE 0423$-$5502  &                     &         & 95  & Oct 1.52m &  900 & 157 & 04 24 32.5 & $-$54 55 38 & 16.9\\\\ HE 0442$-$3027  &                     &         & 96  & Oct 1.52m & 1660 & 421 & 04 44 29.4 & $-$30 21 36 & 16.2\\\\ HE 0446$-$2531  &                     &         & 97  & Oct 1.52m &  480 & 485 & 04 49 01.4 & $-$25 26 36 & 16.9\\\\ HE 0449$-$2554  &                     &         & 98a & Nov 1.54m & 1200 & 485 & 04 51 53.8 & $-$25 49 15 & 16.5\\\\ &                     &         & 94  & Nov 1.52m & 1800 &     &            &             & \\\\ &                     &         & 98a & Nov 1.54m & 1200 &     &            &             & \\\\ HE 0453$-$4423  &                     &         & 95  & Oct 1.52m &  900 & 251 & 04 55 22.8 & $-$44 18 19 & 16.8\\\\ HE 0956$-$1121  & EC09565$-$1121        & DB      & 93  & Mar 3.6m  &  300 & 709 & 09 59 01.3 & $-$11 35 24 & 16.9\\\\ HE 1002$-$1929  &                     &         & 93  & Mar 3.6m  &  300 & 567 & 10 05 02.8 & $-$19 44 29 & 17.1\\\\ HE 1102$-$1850  &                     &         & 93  & Mar 3.6m  &  300 & 570 & 11 05 09.2 & $-$19 07 05 & 17.1\\\\ HE 1109$-$1953  & EC11091$-$1953        & DB      & 93  & Mar 3.6m  &  300 & 570 & 11 11 40.2 & $-$20 09 37 & 16.0\\\\ HE 1128$-$1421  & EC11285$-$1421        & DB      & 96  & Mar 1.52m &  300 & 641 & 11 31 05.5 & $-$14 38 09 & 16.7\\\\ HE 1128$-$1524  &                     &         & 96  & Mar 1.52m &  300 & 641 & 11 30 51.7 & $-$15 41 26 & 17.4\\\\ HE 1130$-$0111  &                     &         & 96  & Mar 1.52m &  300 & 858 & 11 33 25.9 & $-$01 28 30 & 17.6\\\\ HE 1145+0145    &                     &         & 96  & Mar 1.52m &  300 & 858 & 11 48 33.6 &   +01 28 59 & 17.2\\\\ HE 1146$-$0109  &                     &         & 96  & Mar 1.52m &  300 & 858 & 11 48 51.7 & $-$01 26 12 & 17.3\\\\ HE 1149$-$1320  & PG1149-133, EC11492 & DB2     & 92  & Feb 2.2m  &  300 & 642 & 11 51 50.5 & $-$13 37 14 & 16.1\\\\ HE 1214$-$2017  &                     &         & 96  & Mar 1.52m &  300 & 573 & 12 16 49.9 & $-$20 33 56 & 17.6\\\\ HE 1243$-$2134  & EC12436$-$2134        &         & 96  & Mar 1.52m &  300 & 574 & 12 46 18.5 & $-$21 50 57 & 16.3\\\\ HE 1308$-$1458  &                     &         & 91  & Apr 3.6m  &  300 & 646 & 13 11 12.9 & $-$15 14 28 & 17.2\\\\ HE 1350$-$1612  &                     &         & 92  & Feb 3.6m  &  300 & 649 & 13 53 34.9 & $-$16 27 05 & 17.3\\\\ HE 1352+0026    & PG1352+004          & DB4     & 96  & Mar 1.52m &  300 & 865 & 13 55 32.4 &   +00 11 24 & 16.1\\\\ HE 1409$-$1821  &                     &         & 93  & Mar 3.6m  &  300 & 578 & 14 11 48.6 & $-$18 35 04 & 15.6\\\\ HE 1420$-$1914  &                     &         & 96  & Mar 1.52m &  300 & 579 & 14 23 24.1 & $-$19 27 54 & 17.3\\\\ HE 1421$-$0856  &                     &         & 93  & Mar 3.6m  &  300 & 722 & 14 24 37.9 & $-$09 09 57 & 17.3\\\\ HE 1428$-$1235  & EC14289-1235  &         & 92  & Apr 3.6m  &  300 & 650 & 14 31 39.6 & $-$12 48 56 & 15.9\\\\ HE 2147$-$0950  & $^c$                &         & 98  & Sep 1.52m &  300 & 744 & 21 49 55.7 & $-$09 36 31 & 16.6\\\\ HE 2237$-$0509  & $^c$                &         & 98  & Sep 1.52m &  300 & 819 & 22 39 40.7 & $-$04 54 17 & 14.0\\\\ HE 2237$-$3630  & LHS 3841, LP984-76  &         & 98  & Sep 1.52m &  480 & 406 & 22 40 05.2 & $-$36 15 20 & 17.6\\\\ [0.1em] \\hline\\noalign{\\smallskip} \\multicolumn{10}{l}{$^a$ classified as magnetic WD without detectable polarization in McCook \\& Sion (\\cite{mccook}), Magnetic white dwarf with a}\\\\ \\multicolumn{10}{l}{\\hspace{0.9ex} polar field strength of 176 MG (Reimers et al.\\cite{reimersb}).}\\\\ \\multicolumn{10}{l}{$^b$ mistakenly identified as PHL1201 in Lamontagne et al. \\cite{lamontagne00}}\\\\ \\multicolumn{10}{l}{$^c$ coincidences with PHL objects (PHL3345, PHL1477, PHL166, PHL363) have been rejected using} \\\\ \\multicolumn{10}{l}{\\hspace{0.9ex} finding charts provided by E.~Chavira}\\\\ \\multicolumn{10}{l}{$^d$ finding chart from Jaidee \\& Lyng\\aa\\ \\cite{kilkenny97}}\\\\ \\end{tabular} \\end{table*}  The Hamburg/ESO Survey for bright quasars (HES, Wi\\-sotzki et al. \\cite{wisotzki}) was initiated in 1990 to perform a spectroscopic survey, based on digitized objective-prism photo\\-graphs taken with the ESO Schmidt telescope, in the southern hemisphere. A nominal area of $\\sim 9000$ deg$^2$ in the southern sky is covered with an average limiting magnitude on the prism plates of $B\\sim 17.5$. It was conceived as a complement to the Hamburg Quasar Survey (HQS, Hagen et al.  \\cite{hagen}), which aims at a complete coverage of the northern sky except for the galactic plane. Similar surveys have been conducted in the past to search for faint blue objects. Still in progress in the southern hemisphere are the Edinburgh-Cape survey (Stobie et al. \\cite{stobie97}), the Homogenous Bright Quasar Survey (Cristiani et al. \\cite{cristiani95}), and the Montreal-Cambridge-Tololo survey (Lamontagne et al. \\cite{lamontagne00}).  The main goal of both the HES and the HQS survey is the compilation of a large sample of bright QSOs. In a first step QSO candidates are selected by various broadband color criteria and feature detection algorithms applied to spectra derived from low resolution scans of the prism plates (Wisotzki et al. \\cite{wisotzki00}). Since among UV excess objects (i.e. $U-B < -0.4$) less than 10\\%\\ can be expected to be QSOs, the Hamburg Quasar Survey and the Hamburg/ESO Survey are also a rich source of white dwarfs of various types, and early-type subdwarfs. In a recent paper, Homeier et\\clearpage al. (\\cite{homeier}) present an analysis of 80 DA white dwarfs detected through follow-up observations of the HQS. Reimers et al. (\\cite{reimersa}, \\cite{reimersb}, \\cite{reimersc}) reported on the detection of several new magnetic white dwarfs in the HES. A summary of helium-rich subluminous stars can be found in Heber et al. (\\cite{heber96}) and an analysis of 47 sdO stars from the HQS in Lemke et al. (\\cite{lemke98}).  In this paper we present the analysis of 40 helium-rich white dwarfs using LTE model atmospheres. The objects have been found in the HES and have entered the quasar candidate sample because of $U-B\\le -0.18$. Due to the high spectral resolution of the HES spectra ($\\sim 15$\\,{\\AA} at H$\\gamma$), it is possible to identify most of the hot stars in the quasar candidate sample (see Reimers \\& Wisotzki \\cite{reimers97} for example spectra) by their hydrogen and/or helium absorption lines. Featureless, {\\em very\\/} hot stars can be identified by their continuum shape. Therefore, the sample of stars presented in this paper is restricted to a temperature regime of 10000\\,K$<$ \\Teff\\ $\\le$ 20000\\,K -- as the analysis shows -- in which neither strong absorption lines of helium nor a continuum shape notably different from that of a quasar are present. Because it was not possible to make a definitive distinction between quasars and stars on the basis of the HES objective prism spectra alone, spectroscopic follow-up observations have been performed. ",
        "conclusions": "\\subsection{DB white dwarfs}  \\subsubsection{HE~0127$-$3110} HE~0127$-$3110 was observed several times during the follow-up observations performed for the HES: Spectra were obtained in September 1992 (used here), September 1994, November 1994, and November 1998. Spectra from 1994 show broad absorption troughs, varying in depth and profile, whereas the September 1992 spectrum resembles a DB spectrum. After a careful inspection of individual telescope pointings to exclude the accidental observation of a different star, we conclude that HE~0127$-$3110 is variable.  The only common feature of all spectra is an absorption line at 5876\\,{\\AA} which can be attributed to \\ion{He}{i}. The broad absorption features in the spectra of 1994 led to the classification as a magnetic DA white dwarf with a polar field strength of 176 MG (Reimers et al. \\cite{reimersb}). In this scenario the putative \\ion{He}{i} line at 5876\\,{\\AA} is due to a hydrogen H$\\alpha$ component ($2s_0\\rightarrow 3p_0$), which becomes stationary only at a field strength of 235 MG with a minimum wavelength of 5830\\,{\\AA}. It is interesting to mention that HE~2201$-$2250, whose spectrum looked almost identical to the 1994 spectra of HE~0127$-$3110 (Reimers et al. \\cite{reimersb}), meanwhile also changed its spectroscopic appearance to a DB white dwarf. A detailed discussion will be published in a forthcoming paper.   \\subsection{DBA white dwarfs}  \\begin{figure} \\resizebox{\\hsize}{!}{\\includegraphics{aa98_db.fig3}} \\caption[]{DBAZ white dwarfs} \\label{figur3} \\end{figure}  \\begin{table*} \\caption[]{Results from the temperature and $\\log g$ determination and derived spectral types for the analyzed stars. Equivalent widths are given in \\AA ngstr\\\"om } \\label{temp} \\begin{tabular}{llllll}\\hline\\noalign{\\smallskip} Name & Type & \\Teff\\ & $\\log g$ & \\Teff\\ for & Remarks \\\\ &      &        &          & $\\log g=8$ &\\\\[0.1em] \\hline\\noalign{\\smallskip} HE 0004$-$1452 &   DB & 16800$\\pm$100 & 8.17$\\pm$0.06 & 16700$\\pm$100 &\\\\ HE 0104$-$3056 &   DB & 12600$\\pm$320 & 8.11$\\pm$0.30 & 12500$\\pm$580 & \\\\ HE 0108$-$3036 &   DB & 13600$\\pm$260 & 8.36$\\pm$0.21 & 13200$\\pm$430 &\\\\ HE 0127$-$3110 &   DB & 13800$\\pm$120 & 8.48$\\pm$0.09 & 13500$\\pm$130 & variable (?), possibly magnetic\\\\ HE 0149$-$2518 &   DB & 16900$\\pm$280 & 8.21$\\pm$0.17 & 16500$\\pm$310 & \\\\ HE 0200$-$5556 &   DB & 18500$\\pm$250 & 8.26$\\pm$0.12 & 17900$\\pm$230 & \\\\ HE 0203$-$5013 &   DB & 16200$\\pm$310 & 8.37$\\pm$0.20 & 16000$\\pm$290 & \\\\ HE 0308$-$1313 &   DB & 18000$\\pm$170 & 8.90$\\pm$0.20  &18000$\\pm$350 &\\\\ HE 0319$-$4344 &   DB & 18400$\\pm$710 & 7.82$\\pm$0.39 & 18600$\\pm$800 & \\\\ HE 0413$-$3306 &   DB & 16600$\\pm$70 & 7.80$\\pm$0.05 & 16700$\\pm$70 & \\\\ HE 0423$-$5502 &   DB & 15600$\\pm$200  & 8.12$\\pm$0.14 & 15500$\\pm$200 & \\\\ HE 0446$-$2531 & DBAZ & 12600$\\pm$250 & 7.77$\\pm$0.23 & 12600$\\pm$280 & W($\\lambda$)=$50.7^{+5}_{-3}$/$54.3\\pm5$ (\\ion{Ca}{ii}, Oct97/Nov98),\\\\ &&&&&W($\\lambda$)=$8.6\\pm1$ (H$\\alpha$), W($\\lambda$)=$4.7\\pm 0.5$ (H$\\beta$),\\\\ &&&&&W($\\lambda$)=$2.9^{+0.5}_{-1.0}$ (H$\\gamma$)\\\\ HE 0449$-$2554 & DBAZ & 12500$\\pm$220 & 8.23$\\pm$0.15 & 12200$\\pm$290 & W($\\lambda$)=$25.0^{+2}_{-3}$/$22.8^{+2}_{-3}$ (\\ion{Ca}{ii}, Nov94/Nov98)\\\\ &&&&&W($\\lambda$)=$2.8\\pm 0.5$ (H$\\alpha$)\\\\ HE 0956$-$1121 &   DB & 18800$\\pm$90 & 8.02$\\pm$0.04 & 18400$\\pm$80 & \\\\ HE 1002$-$1929 &  DBA & 16500$\\pm$100 & 8.26$\\pm$0.06 & 16200$\\pm$100 & W($\\lambda$)=$3.6\\pm1$ (H$\\alpha$)\\\\ HE 1102$-$1850 &  DBA & 17000$\\pm$100 & 8.25$\\pm$0.06 & 16700$\\pm$120 & W($\\lambda$)=$2.4\\pm 0.6$ (H$\\alpha$), W($\\lambda$)=$1.1\\pm0.6$ (H$\\beta$)\\\\ HE 1109$-$1953 &   DB & 17100$\\pm$70 & 7.85$\\pm$0.04 & 17000$\\pm$60 & \\\\ HE 1128$-$1421 &   DB & 16900$\\pm$240 & 8.04$\\pm$0.14 & 16600$\\pm$240 & \\\\ HE 1128$-$1524 &   DB & 26800$\\pm$1420 & 8.33$\\pm$0.16 & 27800$\\pm$1120 & \\\\ HE 1130$-$0111 &   DB & 15000$\\pm$300 & 8.00$\\pm$0.19 & 15100$\\pm$330 &\\\\ HE 1145+0145 &   DB & 16900$\\pm$250 & 8.02$\\pm$0.15 & 16600$\\pm$250 & \\\\ HE 1149$-$1320 &   DB & 17300$\\pm$120 & 8.07$\\pm$0.07 & 17600$\\pm$130 & \\\\ HE 1214$-$2017 &   DB & 16700$\\pm$200 & 8.05$\\pm$0.12 & 15700$\\pm$210 & \\\\ HE 1243$-$2134 &   DB & 13800$\\pm$240 & 8.44$\\pm$0.15 & 13300$\\pm$280 & \\\\ HE 1308$-$1458 &   DB & 19500$\\pm$190 & 8.35$\\pm$0.07 & 18800$\\pm$170 & \\\\ HE 1350$-$1612 & DBAZ & 15000$\\pm$80 & 8.28$\\pm$0.06 & 14600$\\pm$90 & W($\\lambda$)=$8\\pm1$ (\\ion{Ca}{ii}),\\\\ &&&&& W($\\lambda$)=$2.2^{+0.3}_{-0.5}$ (H$\\alpha$), W($\\lambda$)=$1\\pm0.5$ (H$\\beta$)\\\\ HE 1352+0026 &   DB & 15700$\\pm$150 & 8.09$\\pm$0.10 & 15400$\\pm$170 & \\\\ HE 1409$-$1821 &   DB & 18400$\\pm$110 & 7.88$\\pm$0.06 & 18300$\\pm$130 & \\\\ HE 1420$-$1914 &   DB & 17900$\\pm$220 & 8.24$\\pm$0.12 & 17200$\\pm$370 & \\\\ HE 1421$-$0856 &   DB & 17100$\\pm$90 & 7.84$\\pm$0.05 & 17200$\\pm$90 & \\\\ HE 1428$-$1235 &   DB & 19500$\\pm$50 & 7.87$\\pm$0.02 & 19500$\\pm$60 & \\\\ HE 2147$-$0950 &  DBA & 14600$\\pm$220& 8.17$\\pm$0.13 & 14100$\\pm$410 & W($\\lambda$)=$3.5\\pm0.5$ (H$\\alpha$)\\\\ HE 2237$-$0509 &  DBA  & 13300$\\pm$360& 8.34$\\pm$0.27& 12900$\\pm$380 & W($\\lambda$)=$3.4\\pm1$ (H$\\alpha$)\\\\ [0.1em] \\hline &&&&&\\\\ \\multicolumn{6}{l}{Possible helium-rich stars without helium lines}\\\\[0.3em] \\hline\\noalign{\\smallskip} HE 0052$-$2511 &      & --- & --- & --- & possibly DC or magnetic\\\\ HE 0122$-$2244 &   DZ & --- & --- & $10000^{+2000}_{-1000}$ & W($\\lambda$)=$18.4^{+3}_{-5}$/$19.0\\pm 2$ (\\ion{Ca}{ii}, Oct97/Nov98)\\\\ HE 0309$-$2105 &   DC & --- & --- & --- & \\\\ HE 0442$-$3027 &      & --- & --- & --- & possibly DC or DQ\\\\ HE 0453$-$4423 &   DC & --- & --- & --- & \\\\ HE 1146$-$0109 &   DQ & --- & --- & --- & C$_2$ bands\\\\ HE 2237$-$3630 &   DC & --- & --- & --- & binary\\\\[0.1em] \\hline \\end{tabular} \\end{table*}   \\subsubsection{HE~1002$-$1929 and HE~1102$-$1850}  In the spectra of HE~1002$-$1929 and HE~1102$-$1850 the H$\\alpha$ line is clearly visible with equivalent widths of $(3.6\\pm 1)$\\,{\\AA} and $(2.4\\pm 0.6)$\\,{\\AA}, respectively. The H$\\beta$ line (W($\\lambda$)=$(1.1\\pm 0.6)$\\,{\\AA}) can be detected only in the spectrum of HE~1102$-$1850, despite the larger equivalent width of H$\\alpha$ in HE~1002$-$1929 and a similar temperature. From the equivalent widths we calculated hydrogen abundances of $(5^{+5}_{-2})\\cdot 10^{-5}$ and $(2^{+2}_{-1})\\cdot 10^{-5}$ for HE~1002$-$1929 and HE~1102$-$1850, respectively. Metal lines could not be detected.  \\subsubsection{HE~2147$-$0950 and HE~2237$-$0509} HE~2147$-$0950 and HE~2237$-$0509 show absorption features at the position of H$\\alpha$ with equivalent widths of $(3.5\\pm0.5)$\\,{\\AA} and $(3.4\\pm1.0)$\\,{\\AA}, respectively. Because both spectra are very noisy, this should be taken with caution. The measured equivalent widths would correspond to hydrogen abundances of $(4\\pm2)\\cdot 10^{-5}$ and $(6\\pm4)\\cdot 10^{-5}$, respectively.  \\subsection{DBAZ white dwarfs}  \\subsubsection{HE~0446$-$2531} The most prominent lines in the November 1998 spectrum of HE~0446$-$2531 are hydrogen lines of the Balmer series up to H$\\gamma$, the \\ion{He}{i} line at 4471\\,{\\AA} and a somewhat weaker \\ion{He}{i} line at 5876\\,{\\AA}.  The October 1997 spectrum is too noisy to let the Balmer lines be discerned. Equivalent widths for H$\\alpha$ and H$\\beta$ were determined between 6530\\,{\\AA} and 6590\\,{\\AA}, and 4820\\,{\\AA} and 4885\\,{\\AA}, respectively. Together with the equivalent width of H$\\gamma$ they result in a hydrogen abundance of $(7^{+3}_{-2})\\cdot 10^{-4}$.  The equivalent width of the calcium doublet was determined between 3850\\,{\\AA} and 4050\\,{\\AA} to comprise the whole doublet. This is justified because the \\ion{He}{i} line at 4026\\,{\\AA} is to weak to significantly contribute. The calcium abundance derived from an equivalent width of $(52.5^{+5}_{-6})$\\,{\\AA} (unweighted mean from the October 97 and November 98 spectrum) amounts to $(2\\pm 1)\\cdot 10^{-6}$. At 3704\\,{\\AA}, 3735\\,{\\AA}, 3834\\,{\\AA}, and 5170\\,{\\AA} lines can be detected, which are most probably due to neutral magnesium and neutral as well as ionized iron. With this assumption we derive abundances of $(2^{+3}_{-1})\\cdot 10^{-6}$ and $(2\\pm 1)\\cdot 10^{-6}$ for magnesium and iron, respectively.   \\subsubsection{HE~0449$-$2554} Since the November 94 spectrum is very noisy HE~0449-2554 was reobserved in November 1998. Besides the  doublet of calcium at 3934\\,{\\AA} and 3968\\,{\\AA} already seen in the November 94 spectrum, the new spectrum clearly shows an H$\\alpha$ line. At 4471\\,{\\AA} and 5876\\,{\\AA} also weak features can be spotted. They are consistent with line depths and line strengths of corresponding \\ion{He}{i} lines in a synthetic spectrum with \\Teff\\ = 12200\\,K. The equivalent width of the H$\\alpha$ line ($(2.8\\pm 0.5)$\\,{\\AA}) corresponds to a hydrogen abundance of $(3\\pm1)\\cdot 10^{-5}$. The unweighted mean equivalent width of the calcium doublet from both observing runs ($23.9^{+2}_{-3}$\\,{\\AA}) was taken to determine a calcium abundance of $(7\\pm 2)\\cdot 10^{-8}$.   \\subsubsection{HE~1350$-$1612} In addition to \\ion{He}{i} lines also H$\\alpha$, H$\\beta$, and the \\ion{Ca}{ii} doublet at 3934\\,{\\AA} and 3968\\,{\\AA} can be detected in the spectrum. With an effective temperature of 14600$\\pm$90\\,K, this star is one of the very few DBAZ stars in the temperature range of about 15000\\,K, where \\ion{He}{i} lines make temperature determinations reliable. The only other such star known is HS~2253+8023 (Friedrich et al.  \\cite{friedrich}; \\cite{friedrichb}), and perhaps GD40, with a published equivalent width of H$\\alpha$ of 1\\,{\\AA} (Greenstein \\& Liebert \\cite{greenstein}). Hydrogen and calcium abundances determined from equivalent widths amount to $(2\\pm 1)\\cdot10^{-5}$ and $(2^{+1.0}_{-1.5})\\cdot 10^{-7}$, respectively.   \\subsection{DZ white dwarfs}  \\subsubsection{HE~0122$-$2244}  The \\ion{Ca}{ii} doublet at 3934\\,{\\AA}/3968\\,{\\AA} and an otherwise featureless spectrum classify HE~0122$-$2244 as a DZ white dwarf. An approximate temperature of $10000^{+2000}_{-1000}$\\,K for $\\log g = 8$ was determined from the continuum slope of the photometric November 98 spectrum. For this temperature the equivalent width of the \\ion{Ca}{ii} doublet ($18.7^{+4}_{-5}$\\,{\\AA}, unweighted mean) corresponds to a calcium abundance of $(4^{+1}_{-2})\\cdot10^{-10}$.   \\subsection{DC and DQ white dwarfs}  \\begin{figure*} \\resizebox{\\hsize}{!}{\\includegraphics{aa98_db.fig4}} \\caption[]{DZ, DQ, and DC white dwarfs} \\label{figur4} \\end{figure*}  \\begin{table} \\setbox\\strutbox=\\hbox{\\vrule height9.5pt depth4.5pt width0pt} \\caption[]{Element abundances (relative to helium) for the DBA, DBAZ, and DZ white dwarfs} \\label{abunda} \\begin{tabular}{llcccc} \\hline object & Type & H & Mg & Ca & Fe \\\\ &      & $10^{-5}$ & $10^{-6}$ & $10^{-8}$ & $10^{-6}$\\\\[0.1em] \\hline\\noalign{\\smallskip} HE 0122$-$2244 & DZ & & & $0.04^{+0.01}_{-0.02}$ & \\\\ HE 0446$-$2531 & DBAZ & $70^{+30}_{-20}$ & $2^{+3}_{-1}$ & $200\\pm 100$ & $2\\pm 1$ \\\\ HE 0449$-$2554 & DBAZ & $3\\pm 1$ & & $7\\pm 2$ & \\\\ HE 1002$-$1929 & DBA & $5^{+5}_{-2}$ & & & \\\\ HE 1102$-$1850 & DBA & $2^{+2}_{-1}$ & & & \\\\ HE 1350$-$1612 & DBAZ & $2\\pm 1$ & & $20^{+10}_{-15}$ & \\\\ HE 2147$-$0950 & DBA & $4\\pm 2$ & & & \\\\ HE 2237$-$0509 & DBA & $6\\pm 4$ & & & \\\\[0.1em] \\hline \\end{tabular} \\end{table}  \\subsubsection{HE~0052$-$2511} Weak flux depressions at 5000\\,{\\AA} and 6300\\,{\\AA} are the only features in the observed spectral range from 3500\\,{\\AA} to 9200\\,{\\AA}. Therefore, this star might be of type DC, or possibly a magnetic white dwarf. If this star is magnetic, the main constituent of its atmosphere could be helium, because helium, contrary to hydrogen, has no stationary components and hence has no detectable features over a wide range of field strengths (Schmelcher, priv.comm.).  \\subsubsection{HE~0309$-$2105} HE~0309$-$2105 was already observed by Sefako et al. (\\cite{sefako99}). On the basis of its colours ($B-V=0.10\\pm 0.05$, $U-B=-0.08\\pm 0.05$) derived from $UBV$ photometry they were not able to classify the object. However, our synthetic colours derived from prism plates of the HES and confirmed by spectroscopy yield $B-V=-0.10\\pm 0.08$ and $U-B=-0.88\\pm 0.08$, respectively, which place HE~0309$-$2105 well in the regime of white dwarfs in the two-colour diagram. Together with its featureless spectrum it is therefore classified as DC.  \\subsubsection{HE~0442$-$3027} C$_2$ bands might be the reason for weak flux depressions at 4700\\,{\\AA} and 5100\\,{\\AA}. Therefore, HE~0442$-$302 might rather be a DQ than a DC.  \\subsubsection{HE~1146$-$0109} HE~1146$-$0109 displays strong C$_2$ bands at 4670\\,{\\AA} and 5140\\,{\\AA}. Since no other features can be detected in the spectrum, the star is classified as DQ.  \\subsubsection{HE~2237$-$3630} This star has a composite spectrum, which shows no clear features bluewards of 5800\\,{\\AA}, and strong TiO bands redwards of 5800\\,{\\AA}. There are hints for absorption features at 3730\\,{\\AA} and 4026\\,{\\AA}, which could be attributed to helium, and one at 5170\\,{\\AA}, which could possibly be due to magnesium and iron. We assume that HE~2237$-$3630 is a binary star with a helium-rich white dwarf component and a M-dwarf component."
    },
    "0010/astro-ph0010250_arXiv.txt": {
        "abstract": "We present H$\\alpha$ velocity fields for a sample of nearly face--on spiral galaxies observed with DensePak on the WIYN telescope. We combine kinematic inclinations and position angles measured from these data with photometric inclinations and position angles measured from $I$-band images to show that spiral disks are intrinsically non-circular. \\vskip -0.25in ",
        "introduction": " ",
        "conclusions": ""
    },
    "0010/astro-ph0010120_arXiv.txt": {
        "abstract": "A thermal pulse during the post-AGB phase of stellar evolution may lead to a unique mode of light element nucleosynthesis. The stage is set by the ingestion of the unprocessed envelope material into the hot He-flash convection zone below. If the temperature is sufficiently large and the \\czw\\ abundance high enough (e.g. $T_{\\mem{8}}>0.8$, $\\mem{X}(\\czw) \\simeq 0.4$ and $\\mem{X(H)} \\simeq \\nat{1}{-3}$) protons react faster with \\czw\\ and form \\cdr\\ than destroying \\besi. The latter forms by $\\alpha$-capture of \\hedr\\ after an initial reduction of the \\hedr\\ abundance to about $\\nat{3}{-5}\\mem{X}(\\hevi)$ by the ppI reaction $\\hedr(\\hedr,2\\p)\\hevi$ (for $T_{\\mem{8}} \\simeq 1$).  All \\hedr\\ is burned within minutes to weeks depending on the temperature. \\besi\\ is now present at about the previously mentioned level of \\hedr. Its further fate  is determined by the reactions $\\besi(\\emi,\\nu)\\lisi$ and the $\\alpha$-capture reactions of \\besi\\ and \\lisi. These captures lead to the production of \\bel\\ which in turn is finally destroyed by $\\bel(\\alpha,\\n)\\nvi$. The details of this mechanism of light element production in real stars is expected to be fairly dependent on the description of mixing. ",
        "introduction": "About 20~\\% of all post-AGB stars are hydrogen-deficient. Among them are the Wolf-Rayet type central stars of planetary nebulae (PNe) and the PG1159 stars, with their characteristic helium, carbon and oxygen abundance patterns \\cite{koesterke:97b,demarco:97,dreizler:98}.  In order to understand the evolutionary origin and status of hydrogen-deficient objects new stellar evolution calculations have been recently presented by Herwig \\cite{herwig:00a}. One channel of post-AGB evolution leading to a hydrogen-deficient surface involves a so-called very late thermal pulse (VLTP) as first proposed by Fujimoto \\cite{fujimoto:77}. During such an event the entire unprocessed envelope is mixed into the region of the ongoing He-flash on the convective time scale. Corresponding models by Herwig \\etal\\cite{herwig:99c} have been constructed using an advanced numerical algorithm for the treatment of convective nucleosynthesis. It ensures consistent abundance profiles and energy generation rates even when nuclear and convective time scales are of the same order of magnitude. This is achieved  by a simultaneous solution of the nuclear network equations and the equations of time-dependent convective mixing.  A group of stars which may be related to the hydrogen-deficient post-AGB stars are the R\\,CrBr stars, of which some are not only hydrogen-free but at the same time lithium-rich \\cite{asplund:00}. The famous PN central star \\emph{Sakurai's object} (V4334 Sgr) resembles the R\\,CrBr stars. However, it showed a rapid evolution of temperature, luminosity and surface abundances over the last decade \\cite{asplund:99a,duerbeck:00}. Its evolution in the Hertzsprung-Russell diagram is similar to that of stellar models during a VLTP.  Its lithium abundance exceeds the initial solar abundance by 0.5 to 1.0 dex. This large lithium abundance is surprising given the fragility of lithium ($T_{\\mem{min}}(\\lisi + \\p) \\sim \\nat{2}{6}\\kelv$). We have thus investigated the fate of the light elements, in particular of \\hedr, which is present in the envelope at the time of ingestion into the convectively unstable intershell region of the ongoing He-flash beneath. Here, we present a new reaction channel which is to be considered as\\emph{ hot H-deficient \\hedr\\ burning.} \\begin{figure}[t] \\hspace{2cm} \\begin{minipage}[h]{4cm} \\include{he-la-inc} \\end{minipage} \\caption{\\label{fig:alpha-cap} The ppII-chain and the additional reactions considered for ingestion of light elements into He-burning conditions. At relevant temperatures around $T_\\mem{8} \\sim 1.5$ the $\\alpha$-capture on \\lisi\\ and \\bel\\ are four to five and three orders of magnitude larger than the $\\cdr\\ + \\alpha$ reaction, respectively. \\besi\\ is more similar to the carbon isotopes with its $\\alpha$-capture being about tenfold that of \\cdr\\ and the \\p-capture exceeding that of \\czw\\ by a factor of five. The \\emi-capture of \\besi\\ has a time scale of $0.1 \\jahre$.} \\end{figure} ",
        "conclusions": "We have presented a new mode of light element nucleosynthesis which likely occurs when unprocessed material is ingested into an active and convectively unstable helium-shell, e.g.\\ during a very late post-AGB thermal pulse. We have, however, not yet modelled in detail the mixing processes which ultimately bring the lithium into the atmosphere. Given the fragile constitution of lithium, most other sources of lithium in stars like \\emph{Sakurai's} object can be ruled out. We thus conclude that the proposed mechanism might be responsible for some of the lithium abundance anomalies observed in stars."
    },
    "0010/astro-ph0010316_arXiv.txt": {
        "abstract": "The variability properties of accreting compact objects predicted by dynamical models are characterized by a number of distinct frequencies that are specific to each model.  Because the accretion disks around neutron stars and black holes are hydrodynamic flows, these modulation frequencies cannot be strictly equal to test particle frequencies. I discuss several implications of hydrodynamic corrections to the modulation frequencies predicted by dynamical models. Finally, I show that the recent detection of yet a third kHz QPO in three neutron-star systems favors a previously developed model that attributes the various QPO frequencies to fundamental general relativistic frequencies in the accretion flow. ",
        "introduction": "The accretion flows around compact objects are turbulent and are, therefore, expected to be highly variable. The frequencies that correspond to the dynamical timescales can be as high as $\\simeq 1$~kHz and depend strongly on the distance from the central object. For example, for a Kerr spacetime and in the limit of slow rotation, the Keplerian orbital frequency at a radius $r$ scales as $\\sim r^{-3/2}$, the periastron precession frequency as $\\sim r^{-5/2}$, whereas the nodal precession frequency (if it is caused by frame dragging) scales as $\\sim r^{-3}$ (see, e.g., Perez et al.\\ 1997; Stella, Vietri, \\& Morsink 1999). Because of such strong dependences, it is surprising that the observed power-density spectra of accreting systems are characterized by a distinct number of frequencies in the form of relatively coherent ($\\Delta \\nu/\\nu\\gtrsim 10^{-2}$) quasi-periodic oscillation (QPO) peaks.  Modeling theoretically the presence of multiple QPO peaks is challenging. Current models seek to identify characteristic radii in the accretion flow that pick only a small range of frequencies and hence produce narrow QPO peaks in the power spectra. Such radii include the radius of the innermost stable circular orbit (which is important in the diskoseismic models of, e.g., Nowak \\& Wagoner 1991; Perez et al.\\ 1997; Silbergleit \\& Wagoner 2000), the magnetospheric radius (Alpar \\& Shaham 1985, Strohmayer et al.\\ 1996), or the sonic-point radius (Miller, Lamb, \\& Psaltis 1998).  The simultaneous presence of low- and high-frequency QPOs is then attributed either to multiple characteristic radii in the accretion flow (as discussed, e.g., in Miller et al.\\ 1998; Psaltis et al.\\ 1999b) or to different characteristic frequencies occurring at the same region in the flow, as in the case of the diskoseismic models (see Wagoner 1999) or the relativistic models of Stella et al.\\ (1999) and Psaltis \\& Norman (2000).  In the first class of models, the observed tight correlations between the different QPO frequencies (see, e.g., Psaltis et al.\\ 1999b; Psaltis, Belloni, \\& van der Klis 1999a) may be accounted for by combining the dependence of each QPO frequency on the mass accretion rate (see, e.g., Psaltis et al.\\ 1999b). In the latter case, frequency correlations can be easily reproduced since all QPO frequencies are determined almost entirely by a single parameter (e.g., the characteristic transition radius; Psaltis \\& Norman 2000).  A characteristic signature of each QPO model is the spectrum of QPO frequencies that it predicts beyond the ones currently detected (see, e.g., Miller 2000). For this reason, the inadequacy of current data to distinguish between different models can be overcome with the detection of additional QPO peaks. Even though some attempts for detecting expected QPO peaks at different frequencies have given negative results (M\\'endez \\& van der Klis 2000), a third kHz QPO has been recently discovered in three atoll sources (Jonker, M\\\"endez, \\& van der Klis 2000). This new QPO is significantly weaker than the other kHz QPOs and has a frequency that is $\\simeq 60$~Hz higher than that of the previously known lower kHz QPO.  Motivated by this detection, I discuss in this {\\em Letter} the importance of the hydrodynamic corrections to the predicted QPO frequencies that are introduced by the previously developed dynamical model that identifies the various QPO frequencies with general relativistic frequencies in the accretion disk (Psaltis \\& Norman 2000). I then use this property to show that the model can account for the observed properties of the newly discovered QPO. ",
        "conclusions": "In the previous section, I discussed a number of implications of the hydrodynamic corrections introduced to the predicted QPO frequencies by dynamical models. Moreover, I showed that the previously developed model that attributes the observed QPOs to general relativistic frequencies in the accretion flow can account for the presence of multiple kHz QPOs, as recently observed by Jonker et al.\\ (2000).  In this section I will concentrate to the case of the well studied source 4U~1728$-$34 (Jonker et al.\\ 2000) and discuss the implications of the newly observed QPOs for various models. In this source, four QPOs are observed simultaneously at the following frequencies: $\\nu_1=41.5\\pm 0.2$~Hz (the low-frequency QPO), $\\nu_2=795$~Hz (the lower kHz QPO), $\\nu_{2b}=859\\pm 2$~Hz (the newly discovered QPO), and $\\nu_3=1138\\pm 2$~Hz (the upper kHz QPO).  According to beat-frequency models (see, e.g., Strohmayer et al.\\ 1996; Miller et al.\\ 1998), the highest QPO frequency $\\nu_3$ is that of the Keplerian frequency of a characteristic radius in the accretion disk and the difference $\\nu_3-\\nu_2$ is comparable to the spin frequency of the neutron star. In fact, oscillations during Type~I X-ray bursts have been detected from 4U~1728$-$34 at a frequency of $\\simeq 363$~Hz, which are also thought to correspond to the spin frequency of the neutron star (Strohmayer et al.\\ 1996). In this picture, the presence of the newly discovered QPO at a frequency that is $\\simeq 64$~Hz higher than $\\nu_2$ might be accounted for if the radiation that emerges from the neutron star is modulated not only at the stellar spin frequency, but also at a frequency equal to $\\simeq 64$~Hz, introducing sidebands to the stellar spin (see, however, Alpar 1986 for an alternative way of generating sidebands in a beat-frequency model, because of the radial drift of the accreting material). As pointed out by Jonker et al.\\ (2000), this latter frequency is not equal to the frequency of the observed QPO at $\\nu_1$ and therefore the accretion flow needs to produce modulations at {\\em two\\/} distinct, unrelated, low frequencies. One such low frequency is the nodal precession frequency $(\\Omega-\\Omega_\\perp)/2\\pi$.  Note, however, that, as also discussed by Psaltis et al.\\ (1999b) for other similar cases, the QPO at frequency $\\nu_1$ cannot be caused by frame-dragging effects if the spin-frequency of the neutron star is equal to $\\sim 363$~Hz. (In the relativistic model described in \\S2 the frequency $\\nu_1$ is related to the frame-dragging frequency, but the peak separation of the kHz QPOs is not equal to the spin frequency of the neutron star.)  This can also be extended to the case of the $\\simeq 64$~Hz frequency, which is even higher than $\\nu_1$. Indeed, such a high nodal precession frequency would require an unrealistically high neutron-star moment of inertia of (see, e.g., Psaltis et al.\\ 1999b) \\begin{equation} \\frac{I_{45}}{M/M_\\odot}\\gtrsim 3.1\\left(\\frac{\\nu}{64~\\mbox{Hz}}\\right) \\left(\\frac{\\nu_{\\rm spin}}{363~\\mbox{Hz}}\\right)^{-1} \\left(\\frac{\\nu_3}{1138~\\mbox{Hz}}\\right)^{-2}\\;, \\end{equation} where $I_{45}\\equiv I/10^{45} g\\,cm^3$ is the moment of inertia, $M/M_\\odot$ the mass in solar units, and $\\nu_{\\rm spin}$ the spin frequency of the compact object. Even if a frequency of $\\simeq 64$~Hz corresponded to the first overtone of the nodal precession frequency, the required moment of inertia [$I_{45}/(M/M_\\odot)\\gtrsim 1.55$] would be higher than allowed by general relativity for a slowly rotating star, if any of the current equations of state for neutron-star matter were valid up to $\\simeq 1.5$ times the nuclear saturation density. As a result, none of the two required low frequencies can be equal to the frame-dragging frequency, if the spin frequency of the neutron star in 4U~1728$-$34 is equal to the burst oscillation frequency.  In the case of the dynamical model described in Psaltis \\& Norman (2000), the presence of multiple QPO peaks in the power-density spectra is a direct consequence of the fact that in strong-field general relativity the three dynamical frequencies (eq.~[\\ref{eq:omegak}]--[\\ref{eq:omegap}]) are significantly different from each other. In particular, all the QPO peaks observed so far can be accounted for by only the two lowest-order modes ($m=1,n=0$) and ($m=1,n=1$).  In this interpretation, the difference $f_{2b}-f_2$ depends mostly on the radial epicyclic frequency and the inverse sound-crossing time (see, eq.~[\\ref{dif}]). For a standard $\\alpha$-disk, the latter is equal to $\\Omega_{\\rm c}\\simeq (h/r) \\Omega_\\perp$, where $(h/r)$ is the scale-height of the disk (see, e.g., Psaltis \\& Norman 2000). As a result, the predicted frequency separation between the two lower-kHz QPO peaks becomes \\begin{equation} f_{2b}-f_2 \\simeq 66\\left(\\frac{h/r}{0.12}\\right)^2 \\left(\\frac{\\Omega/2\\pi}{1138~\\mbox{Hz}}\\right)^2 \\left(\\frac{\\kappa/2\\pi}{280~\\mbox{Hz}}\\right)^{-1}~\\mbox{Hz} \\end{equation} and hence the observed frequencies can be easily accounted for, for typical values of the model parameters (see also Fig.~2).  Higher-order modes will also contribute to the variability power at frequencies comparable to $f_2$ and may produce additional structure to the lower kHz QPO peak. The overall spectrum of observed modes will depend mostly on the amplitude of the perturbations that drive the variability outside the characteristic transition radius.  A detailed analysis of the various predicted resonant frequencies and their amplitudes is beyond the scope of this paper and will be presented elsewhere."
    },
    "0010/astro-ph0010470_arXiv.txt": {
        "abstract": "On 26 May 1999, one of the Sloan Digital Sky Survey (SDSS) fiber--fed spectrographs saw astronomical first light. This was followed by the first spectroscopic commissioning run during the dark period of June 1999. We present here the first hour of extra--galactic spectroscopy taken during these early commissioning stages: an observation of the Coma cluster of galaxies.  Our data samples the Southern part of this cluster, out to a radius of 1.5 degrees ($1.8\\,h^{-1}$ Mpc, approximately to the virial radius) and thus fully covers the NGC 4839 group.  We outline in this paper the main characteristics of the SDSS spectroscopic systems and provide redshifts and spectral classifications for 196 Coma galaxies, of which 45 redshifts are new.  For the 151 galaxies in common with the literature, we find excellent agreement between our redshift determinations and the published values, {\\it e.g.}, for the largest homogeneous sample of galaxies in common (63 galaxies observed by Colless \\& Dunn 1996) we find a mean offset of 3\\, ${\\rm km\\,s^{-1}}$ and an RMS scatter of only 24 ${\\rm km\\,s^{-1}}$. As part of our analysis, we have investigated four different spectral classification algorithms: measurements of the spectral line strengths, a principal component decomposition, a wavelet analysis and the fitting of spectral synthesis models to the data. We find that these classification schemes are in broad agreement and can provide physical insight into the evolutionary histories of our cluster galaxies.  We find that a significant fraction (25\\%) of our observed Coma galaxies show signs of recent star--formation activity and that the velocity dispersion of these active galaxies (emission--line and post--starburst galaxies) is 30\\% larger than the absorption--line galaxies.  We also find no active galaxies within the central (projected) $200\\,h^{-1}$ Kpc of the cluster. The spatial distribution of our Coma active galaxies is consistent with that found at higher redshift for the CNOC1 cluster survey.  Beyond the core region, the fraction of bright active galaxies appears to rise slowly out to the virial radius and are randomly distributed within the cluster with no apparent correlation with the potential merger or post-merger of the NGC 4839 group.  We briefly discuss possible origins of this recent galaxy star-formation. ",
        "introduction": "The Coma cluster is the richest cluster of galaxies in our local universe and has thus attracted considerable attention over the last century (see reviews by Biviano 1998 \\& West 1998).  In the optical, for example, Goldwin, Metcalfe \\& Peach (1983; hereafter GMP83) have published an extensive photometric study of the cluster providing accurate positions, colors, magnitudes and ellipticities for 6724 bright galaxies over 2.63 square degrees centered on Coma. Several authors have explored the fainter dwarf galaxy population of Coma (see Bernstein et al. 1995; Kashikawa et al. 1998; Adami et al. 1998 \\& 2000). The dynamics of the cluster have also been well studied. \\markcite{KG82}Kent \\& Gunn (1982) assembled approximately 300 optical redshifts from the literature to determine the cluster mass distribution. This initial work was extended by Colless \\& Dunn (1996; CD96), who collected 556 redshifts (based on new and literature redshifts), and Geller, Diaferio \\& Kurtz (1999) who have extended the dynamical study of Coma to large radii (10 degrees) and larger numbers (1693 redshifts) thus measuring the density profile of Coma well beyond the virial radius of the cluster. Hughes (1989) measured the total mass of Coma using early X-ray observations of the cluster. More recently, ROSAT observations of Coma have provided unprecedented detail of the intracluster gas morphology (e.g., \\cite{bri92}; \\cite{whi93}). ASCA observations of the cluster have provided important information on the temperature structure of the X--ray emitting gas (Honda et al. 1996) which have been recently complemented by XMM-Newton observations (Briel et al 2000; Arnaud et al 2000; Neumann et al 2000)  The Coma cluster has long been regarded as the archetypal relaxed massive cluster of galaxies. However, recent studies of the cluster have shown otherwise. The current view of Coma is that the cluster is the product of one recent, and one ongoing, cluster--group merger.  A group centered on NGC 4839, in the southwest region of the cluster, is falling into (CD96), or has just passed through (\\cite{bur94}), the main body of the cluster and may have triggered new star--formation in galaxies in Coma (Caldwell \\& Rose 1997). The velocity dispersion of this southwest group is approximately $\\frac{1}{3}$ of that of the main cluster.  Meanwhile, the core of Coma has two dominant galaxies, NGC 4874 and NGC 4889, which seem to be the relic central galaxies of previous groups that have merged into the current cluster. The X--ray and dynamical data reveal that both of these dominant galaxies do not appear to sit at the bottom of the cluster potential (CD96; \\cite{whi93}). The lack of a cooling flow and the existence of an extended radio halo support this merging history of the cluster.  The Coma cluster was thus an ideal first target for the Sloan Digital Sky Survey (SDSS; \\cite{yor00}) spectroscopic commissioning program because of its location (the North Galactic Pole), the pre--existence of wide--field galaxy photometry (e.g. GMP83), and the high--density of known redshifts for comparison and testing. Moreover, there remains interesting scientific questions that can be addressed using the unique SDSS spectroscopic hardware {\\it e.g.} the influence of cluster merger events on the star--formation rates of galaxies. The quality and quantity of SDSS spectral data will allow us to study such problems in great detail and in this paper, we start by outlining and comparing the different analysis techniques for classifying SDSS galaxy spectra as well as quantifying their star--formation rates.  Robust, automated, spectral classification methods are necessary to help us understand the distribution and evolution of the galaxies physical properties that define and characterize their spectral energy distributions.  The problem of galaxy spectra classification has been copiously treated in the astronomical literature. In general, methods are based on the measurement of spectral continuum and line features (e.g., the Lick/IDS system: Faber et al 1985; Burstein, Faber \\& Gonz\\'alez 1986) which are then used to classify and derive the galaxies physical properties. Stellar population synthesis models can, for example, be compared to these measurements, or the entire spectrum using template fitting, to provide a physical understanding of the galaxy properties. Recently, other techniques have been investigated. Amongst them, the principal component analysis has attracted a lot of interest (e.g, Connolly et al 1995; Bromley et al 1998; Folkes et al 1999; Ronen et al 1999). The technique is based on decomposing the spectra into a basis that highlights the galaxy differences. Such decomposition can therefore be used to classify the spectra. Different implementations change the way the basis is constructed or how the resulting coefficients of the decomposition are used to generate a classification method. Some authors, for instance, use artificial neural networks to build a classification scheme from the principal component coefficients (e.g., Folkes, Lahav \\& Maddox 1996). Wavelets also provide an orthogonal basis in which the spectra can be decomposed and therefore can in principle be used as a classification method. Along these lines, Pando \\& Fang (1996) and Theuns \\& Zaroubi (2000) have used wavelets to study quasar spectra.  In this paper, we present the first hour of extra--galactic data taken by the SDSS spectroscopic system. Only one of the ten plates originally designed in the Coma region has been observed producing nearly 200 Coma redshifts and thus illustrating the capabilities of this new instrumentation.  These spectra will allow us to investigate several classification schemes and test their applicability to the future SDSS dataset. The paper is structured as follows.  In \\S 2, we briefly highlight the main characteristics of the SDSS spectroscopic system. In \\S 3, we describe the selection of galaxy targets in the Coma cluster region. The observations of the Coma cluster plates are presented in \\S 4. In \\S 5, we describe the data analysis. We discuss the redshift measurements in \\S 6 and classify the galaxy members in \\S 7. Finally, we discuss our data and present our conclusions in \\S 8. ",
        "conclusions": "\\subsection{SDSS Spectroscopic System}  We present in this paper the first extra--galactic spectroscopic data taken by the SDSS during the initial spectroscopic commissioning run in June 1999. Although the data was primarily obtained to commission all the SDSS spectroscopic subsystems (guider, plug--plates, spectrographs, control software), and are therefore far from the nominal quality expected for the SDSS when it is fully operational, these Coma spectra are of reasonable quality and provide a new insight into the galactic content of the Coma cluster, especially at large distances from the cluster center. This is a combination of the high throughput of the SDSS spectrographs, the large field of view of the spectroscopic observations ($3^{\\circ}$ diameter), the extensive wavelength coverage and resolution, and the ability to obtain many hundreds of spectra simultaneously. This single plug--plate observation of Coma illustrates the overall efficiency of the SDSS spectroscopic system and demonstrates the potential for further cluster and field galaxy observations.  As of October 2000, the entire SDSS spectroscopic system has been tested and commissioned. Both spectrographs are operational and are routinely collecting data. The target selection procedure, from determining target objects through to drilling plug-plates and shipping them to APO, has been exercised several times and the emphasis has now shifted to improving the overall efficiency of the observations. At the timing of writing, the SDSS has obtained $\\simeq50,000$ spectra (approximately two thirds galaxies and one third quasars) and these data are being used, amongst other things, to define our integration time, completeness limits as well as debug the SDSS spectroscopic software.  These data also represent one of the largest samples of galaxies and quasars presently in existence and are being used for several scientific programs.   We have obtained 320 spectra (half the number of a nominal SDSS plug-plate) in the central region of the Coma cluster. The availability of one spectrograph only allowed us to observed the South half of plate 133 (see Figure~\\ref{fig3}). Other engineering tasks and bad weather conditions prevented us from obtaining further Coma spectra and future data taken in this region will be acquired as part of the regular SDSS survey (both imaging and spectroscopy). Nevertheless, in one hour of observations with only half of the spectroscopic system commissioned, and without optimizing the target selection, we obtained 196 spectra of Coma galaxies. This number represents approximately $\\frac{1}{3}$ of the number of redshifts presently available in the literature within the same 1.5 deg radius from the cluster center (as estimated from the NED database in August 2000). Amongst the 320 fibers available, 11 were placed on blank sky (to allow sky-subtraction from the rest of the fibers), while the rest of the science fibers yielded 196 Coma galaxies spectra, 49 field galaxies, 5 quasars (which were previously known and targeted on purpose), 47 stars and 12 unidentified spectra.  Four of the unidentified spectra were due to ``spill-over'' from bright adjacent fibers$\\footnote{ Fibers are only separated on the CCD camera by approximately 6 pixels. The FWHM of a fiber on the CCD is approximately 2 pixels. Therefore, the light from bright target fibers can {\\it contaminate} adjacent fibers.}$.  We note here that our target selection attempted to maximize the number of available galaxy targets by including all candidate galaxies {\\it i.e.}, galaxies that were only classified as galaxies on one of the two photographic plates scanned by SuperCOSMOS.  In hindsight, this produced considerable stellar contamination even at bright magnitude, but does ensure a higher level of completeness.  Of the 196 Coma redshifts obtained, 45 represent new identifications. Because of the SDSS large, circular field--of--view, these new redshifts are located at large cluster radii in areas scarcely sampled by other spectroscopic surveys of Coma, {\\it i.e.}  the difference between a survey instrument and a more targeted observation focused on a particular science goal. Given this uniformity, and the quality of our data, we are thus able to obtain a better understanding of the spatial distribution of galaxy classifications within Coma.  For the 151 redshifts found in common with the literature, we found a negligible mean offset between our redshift determinations and those already published.  In particular, when we compared our redshifts to the largest homogeneous sample of redshift available in the literature, the 63 galaxies measured by CD96, we find only a 3 ${\\rm km\\,s^{-1}}$ mean offset with an RMS scatter of only 24 ${\\rm km\\,s^{-1}}$ (this is less than 2 SDSS spectrograph pixels).  This is remarkable given the non--optimal observing conditions, the lack of proper calibration frames, the difference in the spectral resolutions of the two instruments used, as well as the non--standard wavelength calibration used. Moreover, we have looked for correlations between the observed difference between CD96 and our redshift as a function of CCD position, plate position and matching distance and find no significant dependences with these observational parameters.  In summary, it is likely that we have over--estimated the error due to our wavelength calibration given the low RMS difference with CD96; the SDSS spectroscopic systems appear to be remarkably stable and linear over a one hour observation.  \\subsection{Dynamics and Spectral Classifications}  In Figure~\\ref{fig5}, we show the redshift distribution of our 196 Coma members which is qualitatively similar to that seen for the 465 redshifts in CD96 (their Figure~5).  This is not surprising since we have 148 galaxies in common with their analysis.  However, we do sample out to larger radius than CD96 and the fraction of galaxies measured in the NGC 4839 group (in the SW portion of the cluster), with respect to the total, is larger in our sample than in CD96. A single Gaussian fit to our data gives a mean velocity of $6950\\pm66\\: {\\rm km\\,s}^{-1}$ and a velocity dispersion of $\\sigma_v=916\\pm50\\: {\\rm km\\,s}^{-1}$, where we have corrected for the fact that we measure the velocity dispersion with a 3-$\\sigma$ clip criterion. These values are in agreement with previous measurements and, as expected, are an average of the mean velocities, and mean velocity dispersions, of the main cluster core and the NGC 4839 group (see CD96). As already noted by CD96, a single Gaussian fit to the data is an over-simplification with a clear excess of galaxies, compared to the single Gaussian, between $\\simeq 6950 \\rightarrow 7600\\: {\\rm km\\,s}^{-1}$ due to the NGC 4839 group. The central velocity of these excess galaxies is approximately coincident with the mean redshift quoted by CD96 for the NGC 4839 group, $7339\\: {\\rm km\\,s}^{-1}$. We have also investigated the dynamical differences between the absorption line galaxies and the active (emission--line and post--starburst galaxies) galaxies. We find that the {\\bf AB} galaxies given by the line strength classification have a mean velocity of $6923\\pm70\\: {\\rm km\\,s}^{-1}$ and a velocity dispersion of $\\sigma_{\\rm v}=846\\pm54\\: {\\rm km\\,s}^{-1}$, while the rest of the galaxies have $<cz>=7027\\pm160\\: {\\rm km\\,s}^{-1}$ and $\\sigma_{\\rm v}=1160\\pm134\\: {\\rm km\\,s}^{-1}$. The velocity dispersion of the active galaxies is then 30\\% higher than that of the absorption line galaxies. This is similar to the difference found by CD96 for Coma and Carlberg et al (1997) when comparing the velocity dispersions measured from the red and blue galaxies in the CNOC1 cluster survey.  We defer a more detailed examination of the dynamics of the whole Coma region until the future when we have both high quality photometric and spectroscopic data from the main SDSS survey in hand.   In this paper we present several different spectral classifications for our 196 Coma galaxies. The homogeneity and quality of the data allowed us to experiment with these different classification schemes to understand their applicability to the main SDSS survey. The first classification scheme was based on the presence or absence of emission lines. In this case, the H$\\alpha$ line served as the main classifier. For typical IMF and metallicities, H$\\alpha$ cannot be detected in emission $\\sim 25$ Myr after the end of active star formation. Therefore, this criterion efficiently separates galaxies with current star formation from those without.  The strength of the H$\\beta$ line in emission, and the strength of the H$\\delta$, H$\\gamma$ and H$\\beta$ in absorption, are used as additional diagnostics to improve the classification (see Section \\ref{lines}). Using these specific lines, we were able to sub--divide the galaxies into 5 spectral types which we discuss below.   Soon after the termination of a star formation episode, even if it only includes a few percent of the stellar population, the blue part of the optical spectrum of a galaxy is dominated by A stars which are characterized by the Balmer absorption features. If there is no further star formation activity the strength of these lines diminishes gradually with time. The Balmer absorption lines strength thus provides a rough estimate of the age of the stellar population, although the lines are also somewhat sensitive to metallicity. We use this criterion to separate absorption line systems into two sub--classes.  Galaxies without recent star formation, based on the absence of H$\\alpha$ in emission and the absence of strong Balmer absorption, were classed as {\\bf AB}. Galaxies in which their last episode of star formation ended recently, or post--starburst galaxies ({\\bf PS}), were characterized by their strong Balmer absorption lines as well as the absence of H$\\alpha$ in emission. The threshold between {\\bf AB} and {\\bf PS} was set at an equivalent width larger than 15~{\\AA} for the sum of H$\\delta$, H$\\gamma$ and H$\\beta$ equivalent widths. This corresponds to roughly 400-700 million years after the truncation of the active star--formation for solar metallicity.  We note here that our criteria are based on equivalent widths; thus this threshold also depends upon the underlying stellar population (or the previous star formation history) {\\it i.e.}, how many stars are contributing to the continuum.  Emission line galaxies were divided into three sub--classes ({\\bf AB+EM}, {\\bf EM+AB} and {\\bf EM}) depending upon the strength of the H$\\beta$ line. The $W_o$(H$\\beta$) cuts are difficult to translate into star formation rates given the different continuum strengths and the fact that we have not attempted to correct the intrinsic extinction. Nevertheless, the equivalent width of H$\\beta$ represents a measure of the strength of the star formation process with respect to the underlying stellar population.  This classification scheme has the advantage of being simple and easy to measure in optical spectra with sufficient wavelength coverage, and is meaningful as it classifies spectra according to their star formation activity relative to the overall stellar population. Undoubtedly, the use of H$\\alpha$, instead of H$\\beta$, to subdivide the emission line spectra would have been preferred, as it is a better indicator of star formation and is less affected by extinction. However, the nature of our data made us settle for H$\\beta$. We can use this classification as the test base to which compare our other schemes.  Other classification schemes based on spectral features have also been used in the literature. In particular, the MORPHS group (Dressler et al 1999; Poggianti et al 1999) and the CNOC group (Balogh et al 1999) have recently put forward similar classification schemes based on the [OII] and H$\\delta$ lines, in which they provide a more physical interpretation of their types based on evolutionary spectral synthesis models. They use the [OII] and H$\\delta$ lines as indicators of current star formation and the time elapsed after the last episode of star formation, respectively. The large wavelength coverage of our spectra allow us to use more spectral features, H$\\alpha$, H$\\beta$, H$\\gamma$ and H$\\delta$, for the same purpose. Here, we have taken the phenomenological approach of providing a simple classification scheme based on observational quantities which are directly related to the star formation processes taking place within galaxies. We defer to a future paper a more detailed interpretation of these classes in terms of the evolutionary phases of galaxy stellar populations.  There are 44 galaxies (22\\% of our sample) that show evidence of current star formation to some degree. There are five extra galaxies, classified as post starbursts, which ceased their star formation process in the last $\\sim500$ Myr. If we include these five galaxies into the active galaxy population, then 25\\% of our galaxy sample show some sign of recent star formation activity.  In Figure~\\ref{fig3}, we show the spatial distribution of all 196 Coma galaxies; the plotting symbol type reflects our line strength classification scheme discussed above and in Section \\ref{lines}. Clearly, there is no obvious correlation in the position of the different types of galaxies except that the active star--forming galaxies avoid the cluster center.  In addition to the line strength classification scheme, we have explored two objective schemes based on PCA and wavelets. For our PCA analysis, Figure~\\ref{fig7} shows that the first three eigenspectra for our dataset contains 87\\% of the variance while most of the remaining variance is likely due to the noise in our spectra. Each spectrum can therefore be expressed as a linear combination of these three eigenspectra which in turn can be presented in a spherical coordinate system with only two important parameters (see Figure~\\ref{fig10}). We have defined regions in this two--dimensional space according to our line strength classification scheme.  In Table~\\ref{tbl3}, we compare the line strength and PCA classifications to check whether the two--dimensional space defined by the eigencoefficient projection angles can be understand in term of the absorption and emission lines of the spectra. The agreement between both classifications is obviously good but nevertheless remarkable as there was no a priori guarantee that such a good separation should be achieved given that the PCA classification is based on the overall continuum subtracted SED while the line strength classification only on the Balmer lines. Figure~\\ref{fig10} demonstrates that simple arbitrary cuts can efficiently separate galaxies according to their spectral line strengths.  The only cut which does not clearly reproduce the line strength classification above is the division between {\\bf AB} and {\\bf AE} galaxies. In fact this is mostly due to fact that we have used H$\\alpha$ to split these two classes in the line strength classification but have only used the blue part of the spectrum in the PCA analysis. Some of these galaxies show H$\\alpha$ in emission while the rest of the Balmer series is seen in absorption. PCA thus classifies the galaxy as an {\\bf AB} galaxy while in the line strength classification it is an {\\bf AB+EM} galaxy based on the presence of faint H$\\alpha$ emission.  In addition, two galaxies classified as {\\bf AB+EM} by their line strengths are classified as {\\bf PS} by the PCA analysis (fibers 91 and 140). Both galaxies show strong broad Balmer absorption features which are filled with narrower emission. In these cases, it is difficult to separately measure the absorption and emission line components equivalent widths. We estimate that galaxy 91 has $W_o(H{\\delta}+H{\\gamma}+H{\\beta}) \\sim 10$~{\\AA} in absorption and galaxy 140 has $W_o(H{\\delta}+H{\\gamma}+H{\\beta}) \\sim 15$ {\\AA}. The latter would have been classified as a {\\bf PS} galaxy by the line strength criteria if it did not have the emission. These two galaxies have certainly experienced star formation activity that has been substantially decreased in the last Gigayear thus giving rise to the strong Balmer absorption lines. However, there is still some residual star formation activity going on and thus the weak emission on top of the absorption.  We have used a version of the PCA based on the implementation presented in Connolly et al (1995) and Connolly and Szalay (1999).  Our method uses the data themselves to generate the eigenbasis in which the observed spectra are projected. We then use another classification scheme, the line strength classification, to subdivide the eigencoefficent parameter space according to it.  The line strength classification can be translated strikingly well into separated regions on the two-parameter space given by the angles $\\theta$ and $\\phi$. Other classification schemes employing PCA have also been used in the literature. For example, Bromley et al (1998) use a PCA to classify the Las Campanas Redshift Survey. Their implementation is very similar to ours. They use the data themselves to generate the eigenbasis as we do. They weigh their spectra by a smoothed version of the mean spectrum, while we continuum subtract each spectrum with a low order polynomial. Their sample is composed of field galaxies instead of galaxy cluster members and therefore their first eigenspectrum (their Figure~1) is more emission line dominated. Unlike us, they only use their first two eigencoefficients to classify their spectra, neglecting the third eigencoefficient that appears to be able to differentiate post-starburst galaxies. They subdivide their two eigenspectra parameter space into six ``clans'' or spectral types, without using any other external classification scheme for that subdivision. Their mean spectra for their clans show (their Figure 3), nevertheless, a good correlation with star formation. Post-starburst galaxies are, however, not represented. Folkes et al (1999) classify a preliminary sample of the 2dF galaxy survey using a PCA as well. Their sample is made up of field galaxies. Their PCA implementation is similar to us except that they subtract a mean spectrum from their spectra before applying the PCA. Their eigenspectra (their Figure 5) reflect both facts with their mean spectrum being similar to our first eigenspectrum (modulo our continuum subtraction), their first eigenspectrum being similar to our second and their second similar to our third. They use another external classification to subdivide their first two eigencoefficient parameter space to generate their classification. They argue that their third eigencoefficients do not add much additional information for their classification scheme. The ESO-Sculptor Survey has also been classified with a PCA method by Galaz \\& de Lapparent (1998). They employ the Connolly et al (1995) implementation, that is, the same method we use. Again, their eigenspectra are more emission line dominated than ours as they study a field galaxy population. They decompose their spectra into the same projection angles we do. However, they classify their spectra comparing them to the Kennicutt (1992) spectra projected onto the same eigenbasis generated by the data themselves. They find that their spectra can be interpreted as a one parameter family in their $\\theta$-$\\phi$ space. Their sample is of comparable size and signal-to-noise to ours but of coarser resolution. On the other hand, both Bromley et al (1998) and Folkes et al (1999) samples are larger and contain noisier spectra than ours. Their classification (as well as Galaz \\& de Lapparent's) rely on the spectra being a one parameter family going along stellar activity strength. Our spectra, however, being a cluster sample, has more absorption line galaxies and more post-starburst galaxies that require another parameter to be differentiated from the star forming sequence. We argue that our classification requires two parameters: one, describing the stellar formation activity; the other, the strength of the Balmer absorption features.  There have been other approaches to construct classification schemes using PCA. Connolly et al (1995), Connolly \\& Szalay (1999), Sodr\\'e \\& Cuevas (1997) and Ronen et al (1999) use spectral synthesis models (or galaxy spectra already classified) to construct their PCA eigenbases. They, then, interpret their classifications in terms of their input, either their already-classified spectra or the synthesis model parameters.  Given that spectral synthesis models currently available do not have the spectral resolution of our observed spectra, we have chosen to use our own spectra to build our eigenbasis and apply the line strength classification to subdivide the eigencoefficient parameter space. Therefore, not losing the spectral resolution of our data.   We have also used wavelets to define a simple classification for these spectra which uses both the absorption and emission lines in conjunction with the continuum shape. As with the PCA analysis, we only utilize the blue part of the spectrum and thus have no information on H$\\alpha$. We also do not try and differentiate between the {\\bf EM}, {\\bf AB+EM} and {\\bf EM+AB} line strength types since, at present, we cannot measure the equivalent width of lines directly from the wavelet analysis (although in principle this could be achieved). In Table~\\ref{tbl4}, we compare the wavelet classification to the line strength classification. As above, there is good agreement between the two where the line strength sub--classes of {\\bf AB+EM} and {\\bf EM+AB} classifications are mostly shared between the {\\bf EM} and {\\bf AB} wavelet classes (19 {\\bf AB+EM} galaxies were placed in the wavelet {\\bf AB} class as expected since there is no information about H$\\alpha$). The only surprises in Table~\\ref{tbl4} are the four galaxies (fibers 84, 138, 201 \\& 244) classified as {\\bf AB} in wavelet space but {\\bf EM+AB} by the line strength criteria. A visual inspection of these spectra shows that the signal--to--noise of these data is, on average, low. However, all four have moderately weak emission lines that probably missed the top 8 wavelet coefficients because of their size. All four do have strong Calcium H \\& K indicative of an old stellar population. We also note that the wavelet algorithm has more {\\bf PS} galaxies than the line strength classification. This is primarily due to the hard threshold of $W_o(H{\\delta}+H{\\gamma}+H{\\beta})>15$ {\\AA} imposed in the line strength classification. These extra wavelet {\\bf PS} galaxies (three classified as {\\bf AB} and 7 as {\\bf AB+EM} galaxies by the line strength method) do have Balmer absorption lines, but either they have H$\\alpha$ in emission and/or the Balmer absorption is weak (although still stronger than galaxies dominated by old stellar populations). This apparent discrepancy in the classification schemes can be easily understood in terms of the thresholds used in assigning the {\\bf PS} classification, {\\it i.e.} if the 15 {\\AA} threshold used in the line classification scheme had been lowered, then many of these galaxies would have also been classified as {\\bf PS} by the line strength scheme. The wavelet scheme has simply picked extra {\\it older} post--starburst galaxies with weaker Balmer absorption lines.   In Table~\\ref{tbl5}, we show the cross-comparison between the PCA and wavelet classifications.  Again, the agreement is very good. The only exceptions are the same cases discussed above given that the PCA classification cuts were applied to reproduce the line strength classification. It is worth mentioning here that the extra wavelet {\\bf PS} classified galaxies discussed above reside in the ``bridge of galaxies'' that join the {\\bf AB} and {\\bf PS} regions shown in Figure \\ref{fig10}. This again explains why we find more wavelet classified {\\bf PS} galaxies compared to the other schemes since we are simply sampling more of the galaxy population between these two classes.   We have also fitted evolutionary spectral synthesis models to our Coma spectra (only the blue side of the spectrum) to understand the composition and history of the stellar populations in these galaxies. We note that these fits are performed for a particular model, with a chosen star formation prescription, and should be interpreted as such. The star formation history of a galaxy is more complex that the simple parameterized models used (e.g., Abraham et al 1999) and the best fit gives only an indication of the properties of the combined stellar population and helps to compress the spectral information into a handful of meaningful parameters.  We have chosen to fit the age of the stellar population and a parameter describing the star formation prescription using two families of models (see section~7.4 for further details). Both parameterizations give similar results (see table~\\ref{tbl2}). It is nevertheless clear from table~\\ref{tbl2} that in some cases the star formation prescription utilized is unable to fit properly the observed spectrum and a different prescription with combinations of different bursts is necessary. We have chosen, at this stage of simple comparison with other classification schemes, not to fit for the intrinsic absorption or the galaxy metallicity, assuming fixed solar metallicity and no intrinsic absorption. Indeed, there are known degeneracies between age--metallicity and age--intrinsic absorption when fitting synthesized SED to observed spectra (e.g., Worthey 1994; Thompson, Weymann \\& Storrie-Lombardi 2000). In general, a higher metallicity and the effect of neglecting internal extinction will be compensated by an older computed age.  Figure~\\ref{fig17} presents the age and time scale, $\\tau$, of the best fit synthesized SED using an exponential SFR law. The symbols represent the different line strength classifications while the size of the symbol represented the observed magnitudes of the galaxies.  Most of the Coma galaxies cluster along a strip in the age--$\\tau$ space. The large dynamic range shown for the $\\tau$ parameter gives the impression that galaxies of different line strength classes populate similar regions in this parameter space. However, the general trends of age-$\\tau$ values with respect to the spectral line strength classification are as expected: {\\bf AB} galaxies populate regions of old ages ($\\ge3-4$ Gyr) and short time--scales of star formation which produce old stellar population while the {\\bf AB+EM}, {\\bf EM+AB} and {\\bf EM} galaxies occupy regions next to the {\\bf AB} galaxies but gradually moving to younger ages and longer $\\tau$ as the emission lines strength increases. Finally, the {\\bf PS} galaxies populate the region of short time scale star formation, which correspond to short and intensive star formation histories, and short ages after the strong formation episode. To help visualize these separate regions of the age-$\\tau$ space, we show in Figure~\\ref{fig18} the 1, 2 and 3$\\sigma$ contours of the age-$\\tau$ regions occupied by three representative galaxies drawn from our three main spectral types {\\it i.e.}, an {\\bf AB} type galaxy (fiber 143), a {\\bf PS} type galaxy (fiber 95) and a {\\bf EM+AB} type galaxy (fiber 117). It is clear from this plot that the {\\bf AB} and {\\bf EM+AB} galaxy inhabit different regions of the age-$\\tau$ space (at the 3$\\sigma$ level) while there is no intersection for the {\\bf AB} and {\\bf PS} galaxies either. Only the {\\bf EM+AB} and {\\bf PS} galaxies share a similar parameter space (at the 3$\\sigma$ level). Nevertheless, it is clear from Figures~\\ref{fig18} and \\ref{fig17} that the fits of the exponential decaying star formation models to the observed spectra are degenerate along a line in the log $tau$--log age parameter space of slope $\\sim$ 3/2. In summary, regardless of this degeneracy, our spectral classifications have a broad relationship to expected physical models of star--formation and therefore, they can be used to study star--formation and galaxy evolution with the Coma cluster.  \\subsection{Galaxy Star--Formation in Coma}  Our analysis of the Coma cluster has revealed that a substantial fraction (25\\%) of galaxies in Coma (out to the virial radius) exhibit recent star--forming activity. Such analyses are at the heart of studies of galaxy evolution in clusters and, in more general terms, the ``Butcher--Oemler'' effect (Butcher \\& Oemler 1984).  Our large fraction of recent star--forming galaxies is more than initially expected for such a low redshift cluster but is in reasonable agreement with other spectroscopic studies of galaxy star--formation in clusters. For example, Caldwell \\& Rose (1997) found that $\\sim15\\%$ of galaxies in five nearby clusters showed signs of recent star--formation while the work of Ellingson et al. (2000) shows that $\\sim20\\%$ of galaxies within the virial radius ($\\sim r_{200}$) of CNOC1 clusters ($0.18<z<0.5$) show signs of recent star--formation. However, we do not address here the redshift dependence of the BO effect since that will require a large, homogeneous sample of clusters spanning a significant range in redshift (See Andreon \\& Ettori 1999; Margoninier \\& Carvalho 2000). We will address this issue using the main SDSS sample.  We explore here possible mechanisms for triggering star--formation in clusters of galaxies. Several authors (Burns et al. 1994; Caldwell \\& Rose 1997; Metevier, Romer \\& Ulmer 1999; Wang, Connolly \\& Brunner 1997) have proposed that shocks caused by cluster merger events can trigger star--formation in cluster galaxies. However, Figures~\\ref{fig3} and \\ref{fig19} show that our Coma galaxies with recent star--formation ({\\bf EM+AB},{\\bf AB+EM}, {\\bf EM} and {\\bf PS}) are random distributed throughout the cluster with no apparent correlation with the NGC 4839 group. This implies that this merger (or post--merger) has had little effect on the recent star--formation rates of Coma contrary to the scenario outlined in Burns et al. (1994) and Caldwell \\& Rose (1997). This difference may be the combination of several effects: {\\it 1)} We have a complete, homogeneous, spectroscopically confirmed, sample of Coma galaxies out to large cluster radii {\\it i.e.} we have targeted all bright galaxies out to the virial radius of Coma regardless of their location, color or morphological type. Caldwell \\& Rose (1997) were forced to selectively sample specific areas in Coma, because of the smaller field--of--view of their instrument, as well as preferentially targeting early--type galaxies; {\\it 2)} We have used different schemes, and thresholds, for defining post--starburst and star--forming galaxies. For example, if we lowered our threshold for the definition of a {\\bf PS} galaxy (to older and weaker lines) we may find a correlations with the NGC 4839 group.  We note however that the extra wavelet {\\bf PS} galaxies (which are older {\\bf PS} galaxies) show no sign of excess clustering with the NGC 4839 group; {\\it 3)} The NGC 4839 is approximately 5--10\\% (CD96) of the mass of the core of Coma and thus the merger of Coma and this group may not be severe enough to trigger the levels of star--formation seen in more extreme head--on cluster--cluster and group--group mergers (Metevier et al. 1999); {\\it 4)} If the NGC 4839 group is still ``spiralling'' into Coma then we could simply be seeing the combination of ram--pressure induced star--formation as the field galaxies first infall into the cluster (see Dressler \\& Gunn 1983) as well as the expected recent star--formation in the NGC 4839 group before it fell in (see Zabludoff et al. 1996). This would also explain the large velocity dispersion measured for the emission--line galaxies (Carlberg et al. 1997). We must await further data to fully discriminate between these possible models or others. It is important to note that we only sample galaxies brighter than $b=18$ and therefore our observations only apply to the brightest galaxy members and not to the dwarf population (see Rakos et al. 2000). Moreover, our observations do not distinguish between different morphological types (see Doi et al. 1995a,b); this will be addressed using the main SDSS survey data.  The characteristic radius of the Coma cluster is $r_{200}\\sim 1.5$ $h^{-1}$ Mpc (Geller et al 1999), or 1.25 degrees, and thus, we sample the galaxy population out to the virial radius and slightly beyond. This provides a unique opportunity to study the large--scale star--formation histories of galaxies in this nearby cluster.  In Figure~\\ref{fig20}, we present the fraction of passive (absorption) and active (emission plus post--starburst) galaxies as a function of projected distance from the center of Coma.  The azimuthally averaged fraction of active galaxies ({\\bf PS, AB+EM, EM+AB, EM} types) appears to increase slightly with radius from 0.15 to 1.2 degrees$\\footnote{At the distance of Coma, one degree on the sky corresponds to 1.2 $h^{-1}$ Mpc.}$, but the observed percentages are also consistent with a constant radial fraction of active galaxies in this region. This trend is different from that observed in the CNOC sample at higher redshift (Ellingson et al 2000), {\\i.e.}, they observe a steeper radial profile for the fraction of active galaxies in their high redshift clusters.  However, the results are consistent given the errors and the different analyses performed which render a detailed comparison difficult. Moreover, the target selection used different criteria, although similar, which need to be taken into account to provide a fair comparison.   We find no active galaxies in the central (projected) $\\sim 200$ $h^{-1}$ Kpc of the Coma cluster. This result is similar to that observed by Ellington et al. (2000) who found a deficit of star--forming galaxies within 0.5$r_{200}$ of the cores of CNOC1 clusters (see also Rakos et al. 2000 who found a similar trend in intermediate redshift clusters). Ellingson et al. (2000) propose that the dense intracluster medium in the cores of clusters inhibits star--formation producing a deficit of star--formation compared to the outskirts of the cluster and the general field population.  Our data certainly agrees with this scenario except we do see a reduction of active galaxies beyond $r_{200}$, although this reduction is not statistically significant. We defer a detailed discussion of Coma's star--forming galaxies, and their relation to the X--ray gas, to a future paper."
    },
    "0010/physics0010021_arXiv.txt": {
        "abstract": "New optical arrangements with two single-mode input fibers and a fiber-optic coupler are devised to measure the instantaneous velocity difference and local velocity.  The fibers and the coupler are polarization-preserving to guarantee a high signal-to-noise ratio. When the two input fibers are used to collect the scattered light with the same momentum transfer vector but from two spatially separated regions in a flow, the obtained signals interfere when combined via the fiber-optic coupler.  The resultant light received by a photomultiplier tube contains a cross-beat frequency proportional to the velocity difference between the two measuring points. If the two input fibers are used to collect the scattered light from a common scattering region but with two different momentum transfer vectors, the resultant light then contains a self-beat frequency proportional to the local velocity at the measuring point. The experiment shows that both the cross-beat and self-beat signals are large and the standard laser Doppler signal processor can be used to measure the velocity difference and local velocity in real time.  The new technique will have various applications in the general area of fluid dynamics.   \\vskip\\baselineskip \\hskip-\\parindent {OCIS codes: 120.7250, 060.2420, 060.1810, 030.7060.} ",
        "introduction": "Measurements of the velocity difference or the relative velocity, $\\delta{\\bf v}({\\bf \\ell})= {\\bf v}({\\bf x}+{\\bf \\ell})-{\\bf v}({\\bf x})$, between two spatial points separated by a distance $\\ell$ have important applications in fluid dynamics.  For example, in the study of turbulent flows one is interested in the scaling behavior of $\\delta{\\bf v}(\\ell)$ over varying distance $\\ell$, when $\\ell$ is in the inertial range, in which the kinetic energy cascades at a constant rate without dissipation.\\cite{Frisch95,Sreeni99}  If the separation $\\ell$ is smaller than the Kolmogorov dissipation length,\\cite{Frisch95,Sreeni99} the measured $\\delta{\\bf v}(\\ell)$ becomes proportional to the velocity gradient $\\partial v/\\partial r\\simeq \\delta{\\bf v}(\\ell)/\\ell$ (assuming $\\ell$ is known), a useful quantity which is needed to determine the energy dissipation and flow vorticity.  In many experimental studies of fluid turbulence, one measures the local velocity as a function of time and then uses Taylor's frozen turbulence assumption to convert the measured temporal variations into the spatial fluctuations of the velocity field.\\cite{Taylor38}  The frozen turbulence assumption is valid only when the mean velocity becomes much larger than the velocity fluctuations. For isotropic turbulent flows with a small mean velocity, direct measurement of $\\delta{\\bf v}(\\ell)$ is needed.  Over the past several years, the present authors and their collaborators have exploited the technique of homodyne photon correlation spectroscopy (HPCS) to measure $\\delta{\\bf v}(\\ell)$.\\cite{Narayan97,Yixue98} With the HPCS scheme, small particles seeded in a flowing fluid are used to scatter the incident laser light.  The scattered light intensity $I(t)$, which fluctuates because of the motion of the seed particles, contains Doppler beat frequencies of all particle pairs in the scattering volume.  For each particle pair separated by a distance $\\ell$ (along the beam propagation direction), their beat frequency is $\\Delta \\omega_2={\\bf q} \\cdot \\delta {\\bf v}(\\ell)$, where $\\bf q$ is the momentum transfer vector.  The magnitude of $\\bf q$ is given by $q=(4\\pi n/\\lambda)\\sin (\\theta/2)$, where $\\theta$ is the scattering angle, $n$ is the refractive index of the fluid, and $\\lambda$ is the wavelength of the incident light. Experimentally, the Doppler beat frequency $\\Delta \\omega_2$ is measured by the intensity auto-correlation function,\\cite{Berne76} \\begin{equation} g(\\tau)=\\frac{\\langle I(t+\\tau)I(t)\\rangle}{\\langle I(t)\\rangle ^{2}} =1+bG(\\tau), \\label{eq1} \\end{equation} where $b$ ($\\le 1$) is an instrumental constant and henceforth we set $b=1$.  The angle brackets represent a time average over $t$.  It has been shown that $G(\\tau)$ in Eq. (\\ref{eq1}) has the form\\cite{Tong88a} \\begin{equation} G(\\tau)=\\int_{0}^{L} dr~h(r) \\int_{-\\infty}^{+\\infty}d\\delta v~P(\\delta v,r) \\cos (q\\delta v \\tau), \\label{eq2} \\end{equation} where $\\delta v$ is the component of $\\delta {\\bf v}$ in the direction of $\\bf{q}$, $P(\\delta v,r)$ is the probability density function (PDF) of $\\delta v(r)$, and $h(r) dr$ is the number fraction of particle pairs with separation $r$ in the scattering volume. Equation (\\ref{eq2}) states that the light scattered by each pair of particles contributes a phase factor $\\cos(q\\tau \\delta {\\bf v})$ (because of the Doppler beat) to the correlation function $G(\\tau)$, and $G(\\tau)$ is an {\\it incoherent} sum of these ensemble averaged phase factors over all the particle pairs in the scattering volume. In many previous experiments,\\cite{Tong88a,Pak91,Tong92,Kellay95} the length $L$ of the scattering volume viewed by a photodetector was controlled by the width $S$ of a slit in the collecting optics.  While it is indeed a powerful tool for the study of turbulent flows, the HPCS technique has two limitations in its collecting optics and signal processing.  First, a weighted average over $r$ is required for $G(\\tau)$ because the photodetector receives light from particle pairs having a range of separations ($0<r<L$).  As a result, the measured $G(\\tau)$ contains information about $\\delta{\\bf v}(\\ell)$ over various length scales up to $L$.  With the single slit arrangement, the range of $L$ which can be varied in the experiment is limited. The lower cut-off for $L$ is controlled by the laser beam radius $\\sigma$.  The upper cut-off for $L$ is determined by the coherence distance (or coherence area) at the detecting surface of the photo-detector, over which the scattered electric fields are strongly correlated in space.\\cite{Berne76}  When the slit width $S$ becomes too large, the photodetector sees many temporally fluctuating speckles (or coherence areas), and consequently fluctuations in the scattered intensity $I(t)$ will be averaged out over the range of $q$-values spanned by the detecting area.  Recently, we made a new optical arrangement for HPCS, with which the weighted average over $r$ in Eq. (\\ref{eq2}) is no longer needed and the upper limit for $L$ can be extended to the coherence length of the laser.  In the experiment,\\cite{Yixue98} two single mode, polarization-maintaining (PM) fibers are used to collect light with the same polarization and momentum transfer vector $\\bf q$ but from two spatially separated regions in a flow.  These regions are illuminated by a single coherent laser beam, so that the collected signals interfere when combined using a fiber-optic coupler, before being directed to a photodetector.  With this arrangement, the measured $G(\\tau)$ becomes proportional to the Fourier cosine transform of the PDF $P(\\delta v,r)$.  The second limitation of HPCS is related to signal processing.  The correlation method is very effective in picking up small fluctuating signals, but the resulting correlation function $G(\\tau)$ is a time-averaged quantity.  Therefore, the correlation method is not applicable to unstable flows.  Furthermore, information about the odd moments of $P(\\delta v,r)$ is lost, because the measured $G(\\tau)$ is a Fourier cosine transform of $P(\\delta v,r)$.  In this paper, we present a further improvement for HPCS, which is free of the two limitations discussed above.  By combining the new fiber-optic method with the laser Doppler velocimetry (LDV) electronics, we are able to measure the instantaneous velocity difference $\\delta{\\bf v}(\\ell,t)$ and local velocity ${\\bf v}({\\bf x},t)$ at a high sampling rate.  With this technique, the statistics of $\\delta{\\bf v}(\\ell,t)$ and ${\\bf v}({\\bf x},t)$ are obtained directly from the time series data.  The new method of measuring ${\\bf v}({\\bf x},t)$ offers several advantages over the standard LDV. The remainder of the paper is organized as follows.  In Section 2 we describe the experimental methods and setup.  Experimental results are presented and analyzed in Section 3.  Finally, the work is summarized in Section 4. ",
        "conclusions": "\\subsection{Velocity difference measurements}  We first discuss the measurements of the velocity difference in rigid body rotation.  Figure 3(a) shows a sample oscilloscope trace of the analogue output from PMT2 when the separation L=1.0 mm.  The signal is amplified 1250 times and band-pass-filtered with a frequency range of 1-10 kHz.  This oscilloscope trace strongly resembles the burst signals in the standard LDV.  The only difference is that the signal shown in Fig. 3(a) results from the beating of two moving particles separated by a distance L.  Figure 3(a) thus demonstrates that the beat signal between the two moving particles is large enough that a standard LDV signal processor can be used to measure the instantaneous velocity difference $\\delta v(\\ell,t)$ in real time.  Figure 3(b) shows the measured beat frequency $\\Delta \\omega_2$ as a function of separation L.  The circles are obtained from the oscilloscope trace and the triangles are obtained from the intensity correlation function $g(\\tau)$.  The two measurements agree well with each other.  The solid line is the linear fit $\\Delta \\omega_2=41.78 L ~(10^3 ~rad/s)$, which is in good agreement with the theoretical calculation $\\Delta \\omega_2=k_s \\omega L= 42.28 L ~(10^3 ~rad/s)$.  This result also agrees with the previous measurements by Du et al.\\cite{Yixue98} Because $\\Delta \\omega_2$ increases with L, one needs to increase the laser intensity at large values of L in order to resolve $\\Delta \\omega_2$. The average photon count rate should be at least twice the measured beat frequency. \\begin{figure} \\psfull \\centerline{\\psfig{figure=fig3.eps,height=6truecm,clip=true}} \\caption{\\narrowtext(a) Oscilloscope trace of a typical beat burst between two moving particles separated by a distance L=1.0 mm. The signal is obtained in rigid body rotation.  (b) Measured beat frequency $\\Delta \\omega_2$ as a function of separation L.  The circles are obtained from the oscilloscope trace and the triangles are obtained from the intensity correlation function $g(\\tau)$.  The solid line shows a linear fit to the data points.} \\label{fig3} \\end{figure}  We now discuss the time series measurements of $\\delta v(L,t)$ in a jet flow using the LDV signal processor.  The jet flow has significant velocity fluctuations as compared with the laminar rigid body rotation. The measuring point is in the developing region of the jet flow, 3 diameters away from the nozzle exit and is slightly off the centerline of the jet flow.  Figure 4 shows the measured histogram $P(\\delta v)$ of the velocity difference $\\delta v(L,t)$ in the jet flow, when the separation L is fixed at L=0.5 mm (circles) and L=0.8 mm (squares), respectively.  It is seen that the measured $P(\\delta v)$ has a dominant peak and its position changes with L.  Because $\\delta v(L,t)$ increases with L, the peak position moves to the right for the larger value of L.  The solid curve in Fig. 4 is a Gaussian fit to the data points with L=0.5 mm. The obtained mean value of $\\delta v(L,t)$ is $\\langle\\delta v\\rangle =1.87~cm/s$ and the standard deviation $\\sigma= 0.171~cm/s$.  At L=0.8 mm, the measured $P(\\delta v)$ peaks at the value $\\langle\\delta v\\rangle=2.73 ~cm/s$. \\begin{figure} \\psfull \\centerline{\\psfig{figure=fig4.eps,height=6truecm,clip=true}} \\caption{\\narrowtext Measured histogram $P(\\delta v)$ of the velocity difference $\\delta v(L,t)$ in the jet flow.  The values of L are: L=0.5 mm (circles) and 0.8 mm (squares).  The solid curve is a Gaussian fit to the circles.} \\label{fig4} \\end{figure}  In the above discussion, we have assumed that each input fiber sees only one particle at a given time and the beat signal comes from two moving particles separated by a distance L.  In fact, when the seeding density is high, each input fiber may see more than one particle at a given time.  The scattered light from these particles can also beat and generate a self-beat frequency proportional to $\\delta v(\\ell_0)$, where $\\ell_0\\simeq 0.15~mm$ is the laser spot size viewed by each input fiber.\\cite{Yixue98}  The self-beating gives rise to a small peak on the left side of the measured $P(\\delta v)$.  Note that the peak position is independent of L, because $\\delta v(\\ell_0)$ is determined only by $\\ell_0$, which is the same for both measurements.  It is seen from Fig. 4 that the cross-beating is dominant over the self-beating under the current experimental condition.  The intensity correlation function $g(\\tau)$ is also used to analyze the beat signal.  In the experiment, we measure the histogram $P(\\delta v)$ and $g(\\tau)$ simultaneously, so that Eq. (\\ref{eq8}) can be examined in details.  Figure 5 shows the measured $g(\\tau)-1$ (circles) as a function of delay time $\\tau$ at L=0.5 mm.  The squares are the self-beat correlation function $G_s(\\tau)$ obtained when one of the input fibers is blocked.  As shown in Fig. 4, the measured $P(\\delta v)$ has a Gaussian form and thus the integration in Eq. (\\ref{eq8}) can be carried out. The final form of $g(\\tau)$ becomes \\begin{equation} g(\\tau) = 1+G_s(\\tau) \\left [b_s+b_c\\cos[q\\langle\\delta v\\rangle\\tau] e^{-(q\\sigma\\tau)^2/2}\\right ]. \\label{eq9} \\end{equation} The solid curve in Fig. 5 is a plot of Eq. (\\ref{eq9}) with $b_s=0.5$ and $b_c=0.13$.  The values of $\\langle\\delta v\\rangle$ and $\\sigma$ used in the plot are obtained from the Gaussian fit shown in Fig. 4. It is seen that the calculation is in good agreement with the measured $g(\\tau)$.  The fitted value $b_s=0.5$ agrees with the expected value at $I_1=I_2$.  The value of $b_c$ would be 0.5 if the collected signals from the two input fibers were fully coherent and the fiber-optic coupler mixed them perfectly.  The fact that the fitted value of $b_c$ is smaller than 0.5 indicates that the collected signals are not fully correlated.  This is caused partially by the fact that in the present experiment the scattered light suffers relatively large number fluctuations resulting from a changing number of particles in the scattering volume. These number fluctuations produce incoherent amplitude fluctuations to the scattered light and thus introduce an extra (additive) decay to $g(\\tau)$.\\cite{Tong93} Because the beam crossing time (proportional to the beam diameter) is much longer than the Doppler beat time $1/\\Delta \\omega_2$ (proportional to the wavelength of the scattered light), the slow decay due to the number fluctuations can be readily identified in the measured $g(\\tau)$.  This decay has an amplitude 0.4 and has been subtracted out from the measured $g(\\tau)$ shown in Fig. 5. \\begin{figure} \\psfull \\centerline{\\psfig{figure=fig5.eps,height=6truecm,clip=true}} \\caption{\\narrowtext Measured intensity correlation function $g(\\tau)-1$ as a function of delay time $\\tau$ at L=0.5 mm (open circles).  The squares are obtained when one of the input fibers is blocked.  The solid curve is a plot of Eq. (5).} \\label{fig5} \\end{figure}   It is shown in Eq. (\\ref{eq9}) that to accurately measure the mean velocity difference $\\langle\\delta v \\rangle$, the beat frequency $\\Delta \\omega_2=q\\langle\\delta v\\rangle$ must be larger than the decay rate $\\Gamma_s \\simeq q\\delta v(\\ell_0)$ for $G_s(\\tau)$ and also larger than the decay rate $\\Gamma_c \\simeq q\\sigma$ resulting from the fluctuations of the velocity difference.  From the measurements shown in Figs. 4 and 5, we find $\\Gamma_s \\simeq 1.33\\times 10^5 ~s^{-1}$ and $\\Gamma_c \\simeq 3.8\\times 10^4 ~s^{-1}$, which are indeed smaller than the beat frequency $\\Delta \\omega_2\\simeq 4.15\\times 10^5 ~s^{-1}$.  Because $g(\\tau)$ contains a product of $G_s(\\tau)$ and $\\exp[-(q\\sigma\\tau)^2/2]$ [see Eq. (\\ref{eq9})], its decay is determined by the faster decaying function. It is seen from Fig. 5 that the decay of $g(\\tau)$ is controlled by $G_s(\\tau)$, which decays faster than $\\exp[-(q\\sigma\\tau)^2/2]$. It should be noted that in the measurements shown in Fig. 4, the beat signals are analogue ones and we have used a band-pass filter together with a LDV signal analyzer to resolve the beat frequency.  Consequently, many low-frequency self-beat signals are filtered out.  This low-frequency cut-off is apparent in Fig. 4.  The measurements of $g(\\tau)$, on the other hand, are carried out in the photon counting mode, and therefore the measured $g(\\tau)$ is sensitive to all the self-beat signals as well as the cross beat signals.  With a simple counting of particle pairs, we find that the probability for cross beating is only twice larger than that for the self-beating.  \\subsection{Local velocity measurements}  We now discuss the local velocity measurements using the new optical arrangement shown in Fig. 2(a).  The velocity measurements are conducted on a freely suspended flowing soap film driven by gravity. Details about the apparatus has been described elsewhere,\\cite{Rutgers97,Goldburg98,Horvath00a} and here we mention only some key points.  $2\\%$ solution of detergent and water is introduced at a constant rate between two long vertical nylon wires, which are held apart by small hooks.  The width of the channel (i.e., the distance between the two nylon wires) is 6.2 cm over a distance of 120 cm.  The measuring point is midway between the vertical wires. The soap solution is fed, through a valve, onto an apex at the top of the channel.  The film speed $\\bar v$, ranging from 0.5 to 3 m/s, can be adjusted using the valve.  The soap film is approximately 2-6 $\\mu m$ in thickness and is seeded with micron-sized latex particles, which scatter light from a collimated laser beam.  The light source is an argon-ion laser having a total power of 1W.  The incident laser beam is oriented perpendicular to the soap film and the scattered light is collected in the forward direction.  \\begin{figure} \\psfull \\centerline{\\psfig{figure=fig6.eps,height=6truecm,clip=true}} \\caption{\\narrowtext  Measured intensity autocorrelation function $A(\\tau)$ as a function of delay time $\\tau$ with the measuring time T=30 ms.  The inset (a) shows an enlarged portion of $A(\\tau)$ for small values of $\\tau$ up to $\\tau=20 ~\\mu s$.  The inset (b) shows the frequency power spectrum $P(f)$ of the measured $A(\\tau)$.} \\label{fig6} \\end{figure} To measure the rapidly changing beat frequency $\\Delta \\omega_1$, we build a fast digital correlator board for PC.\\cite{Horvath00b} With a fast sampling rate $f_s$, the plug-in correlator board records the time-varying intensity I(t) (number of TTL pulses from the photomultiplier tube per sample time) over a short period of time T and then calculates the (unnormalized) intensity autocorrelation function, $A(\\tau)=\\langle I(t+\\tau)I(t)\\rangle$. Figure 6 shows an example of the measured $A(\\tau)$ as a function of delay time $\\tau$ with T = 30 ms and $f_s=14.32$ MHz.  Because the burst signal I(t) is a periodic function of t, the measured $A(\\tau)$ becomes an oscillatory function of $\\tau$.  The frequency of the oscillation apparent in the inset (a) is the beat frequency $\\Delta \\omega_1$. The amplitude of the oscillation decays at large $\\tau$. The inset (b) shows the power spectrum $P(f)$ of the measured $A(\\tau)$; it reveals a dominant peak at 755.1 kHz.  The power spectrum is obtained using a fast Fourier transform (FFT) program.\\cite{Press92}  \\begin{figure} \\psfull \\centerline{\\psfig{figure=fig7.eps,height=6truecm,clip=true}} \\caption{\\narrowtext Measured intensity autocorrelation function $A(\\tau)$ as a function of delay time $\\tau$ with the measuring time T=50 $\\mu s$.  The inset (a) shows the frequency power spectrum $P(f)$ obtained by FFT.  The inset (b) shows the frequency spectrum $Q(f)$ obtained by the Scargle-Lomb method.} \\label{fig7} \\end{figure} To increase the sampling rate of the velocity measurements, one needs to keep the measuring time T for each $A(\\tau)$ as short as possible. The signal-to-noise ratio for $A(\\tau)$ decreases with shorter measuring time T and with lower mean photon count rate, which was $\\sim 1$ MHz in the present experiment.  It is found that the shortest useful measuring time $T_c$ is roughly $50~\\mu s$.  For this value of T, $A(\\tau)$ is quite noisy and the corresponding peak in $P(f)$ becomes less pronounced [see Fig. 7 and inset (a)].  It is worth mentioning that if one only wants to know the periodicity of a function, rather than its actual power at different frequencies, the Scargle-Lomb method\\cite{SL} is a better alternative to FFT.  This method, which does not require evenly spaced sampling, compares the measured data with known periodic signals using the least-square fitting procedure and determines the relevant frequencies by the goodness of the fit $Q(f)$.  It can even utilize the uneven sampling to further increase the Nyquist frequency.  As shown in Fig. 7(b), the Scargle-Lomb method can still clearly identify the periodicity of the signal, even when the power spectrum $P(f)$ [Fig. 7(a)] becomes less reliable. The total time required for the measurement of the characteristic frequency is less than 1 ms.  Using the correlator board together with an average speed PC (300 MHz), we are able to conduct accurate measurements of the local velocity with a sampling rate up to 1 kHz."
    },
    "0010/astro-ph0010646_arXiv.txt": {
        "abstract": "{ We have analyzed the timing properties of the Narrow-line Seyfert 1 galaxy Mrk 766 observed with {\\it XMM-Newton} during the PV phase. The source intensity changes by a factor of 1.3 over the 29,000 second observation. If the soft excess is modeled by a black body component, as indicated by the EPIC pn data, the luminosity of the black body component scales with its temperature according to $L \\sim T^4$. This requires a lower limit 'black body size` of about $\\rm 1.3 \\cdot 10^{25}\\ cm^2$. In addition, we report the detection of a strong periodic signal with $\\rm 2.4 \\cdot 10^{-4}\\ Hz$. Simulations of light curves with the observed time sequence and phase randomized for a red noise spectrum clearly indicate that the periodicity peak  is intrinsic to the distant AGN. Furthermore, its existence is confirmed by the EPIC MOS and RGS data. The spectral fitting results show that the black body temperature and the absorption by neutral hydrogen remain constant during the periodic oscillations. This observational fact tends to rule out models in which the intensity changes are due to hot spots orbiting the central black hole. Precession according to the Bardeen-Petterson effect or instabilities in the inner accretion disk may provide explanations for the periodic signal. ",
        "introduction": "Mrk 766 is a nearby ($z$ =  0.013) and X-ray  bright ($\\rm \\sim 10^{-11}\\ erg\\ s^{-1}$) Narrow-Line Seyfert~1 galaxy (Walter \\& Fink 1993, Boller et al. 1996, Leighly et al. 1996, Leighly 1999). Due to the extreme spectral and X-ray properties found in Narrow-Line Seyfert 1 galaxies and due to its extreme X-ray brightness, Mrk 766 was proposed for the PV phase for observations with the X-ray satellite {\\it XMM-Newton}. The high throughput of {\\it XMM-Newton} compared to previous X-ray missions allows precise studies of the spectral properties, e.g. the shape of the soft excess and of the transition region between the soft excess and the power-law component. In addition, RGS observations allow to perform X-ray spectroscopy of accretion disk lines (e.g. Ross et al. 1999) as well as emission and absorption features originating from the broad- and narrow-line region. The {\\it XMM-Newton} spectral properties of Mrk 766 are discussed in detail by Page et al. (2001) and Pounds et al. (2001). The long-period orbit of XMM allows to measure the variability power spectra of ultrasoft Narrow-Line Seyfert 1 galaxies far better than has yet been possible before and allow to constrain nonlinear variability, to search for spectral variability and to search for any quasi-periodic oscillations (e.g. Papadakis \\& Lawrence 1995 and references therein). In this paper we  concentrate on the timing properties of Mrk 766 and present the discovery of an X-ray periodicity in Mrk 766. The object resembles IRAS 18325$-$5926 (Iwasawa et al. 1998), where a 58000 second periodicity in the 0.5$-$10 keV {\\it ASCA} energy band was found.  \\begin{figure} \\centering \\centerline{\\psfig{figure=xmm30_f1.eps,width=8.0cm,angle=-90,clip=}} \\centerline{\\psfig{figure=xmm30_f2.eps,width=8.0cm,angle=-90,clip=}} \\caption{ XMM Newton EPIC pn light curve of Mrk 766 in the 0.2--2 and 5--10 keV energy band, respectively. } \\end{figure}  \\begin{figure} \\centering \\psfig{figure=xmm30_f3.eps,width=8.0cm,height=5cm,clip=} \\vskip 0.2cm \\psfig{figure=xmm30_f4.eps,width=7.94cm,height=5cm,angle=-90,clip=} \\vskip 0.2cm \\psfig{figure=xmm30_f5.eps,width=8.45cm,height=5cm,angle=-90,clip=} \\caption{ Upper panel: Reduced $\\rm \\chi^2$ versus the folding period for the 0.2-10 keV energy range. Middle panel: Folded 0.2--2 keV light curve with the best fitting period of 4200 seconds. Lower panel: Power spectrum for Mrk 766 in the 0.2-10 keV energy band. } \\end{figure} ",
        "conclusions": "We first discuss the 'long-term` increase observed in the light curve (cf. Figure 1) which is characterized by a correlation between flux and temperature of the soft component. To our knowledge such a behavior has not been found in AGN spectra before. The correlation between the bolometric luminosity and the spectral temperature is close to the $L_{\\rm bb} \\sim T^4$ law, consistent with a radiating surface remaining constant during the variations. The projected size of the black body radiating disk surface can be estimated using the luminosities and temperatures given in Figure 8. We find that this projected disk size is $\\rm 1.3 \\cdot 10^{25}\\ cm^2$ which is rather a lower limit to the physical size of the emission region, because we have neglected a possible inclination of the disk, gravitational redshift effects and deviations from a black body due to scattering effects. We note that this area estimate requires a black hole mass of at least a few $\\rm 10^5$ solar masses.   The periodic signal may be a quasi-periodic oscillation (QPO). Investigations of QPOs in  galactic binary systems show a wide range of frequencies between about 0.1 and 1200 Hz (Psaltis et al. 1999). However, we note that for the galactic black hole microquasars GRS 1915+105 and GRO 1655-40 narrow frequencies of roughly 50 -- 300 Hz have been detected (Morgan et al. 1997, Remillard et al. 1999). The ratio between these QPO frequencies and the strongest peak frequency found in the data of Mrk 766 lies between $\\rm 10^{5-6}$ roughly consistent with the expected mass ratio of these objects. Specifically, QPOs could be produced by instabilities in the inner part of the accretion disk or by pulsating accretion if the rate is close to the Eddington limit. In these cases the period is expected to be of the order of the radial drift time scale (`instability time scale') $$ \\tau  \\sim \\epsilon \\cdot \\alpha^{-1} \\left(\\frac{r}{R_{\\rm S}}\\right)^{\\frac{3}{2}} \\cdot \\frac{R_{\\rm S}}{c}   \\eqno(1) $$ where $\\epsilon \\sim$ 5 to 10 (Sunyaev 2000). We note that $\\tau$ is of the order of a few thousand seconds for $\\alpha \\sim \\rm 0.1$, $r \\sim 6\\ R_{\\rm S}$ and $M \\sim \\rm  10^6$ solar masses.    If the periodic signal arises from X-ray hot spots orbiting the black hole (e.g. Sunyaev 1973; Guilbert, Fabian \\& Rees 1983) we expect a variation in temperature.  Using the radius dependent orbital velocity as given by Abramowicz et al. (1996), the boost model of Rybicki \\& Lightman (1979) and assuming an inclination angle of 45 degrees we find a temperature variation by a factor of about 1.6 and 1.4 at 3 and 4 Schwarzschild radii, respectively. As shown in Figure 7 the black body temperature remains constant within a few percent during the periodic variations requiring an inclination angle of the accretion disk of more than 85 degrees. For 85 degrees the temperature change according to the moving black body field is about a factor of 1.1 at 3 Schwarzschild radii, still inconsistent with the observations. Even for a nearly face-on geometry with an inclination angle of 89 degrees temperature changes by a factor of 1.06 are expected from the hot spot model.   Another possibility is that the observed variability is due to disk precession according to Bardeen and Petterson (1975). This causes the inner disk to settle into the plane of the black hole equator up to the so called transition radius, whereas the outer parts remain in their original orientation. The disk precession takes place in a ring which is in between. Nelson \\& Papaloizou (2000) found that the transition radius is of the order of about 10 to 30 gravitational radii. Since the precession of the disk is a geometrical effect it does not lead to spectral changes which is consistent with the observations. The absence of significant absorption by neutral gas above the galactic value shows that shadowing by the outer parts of the accretion disk is not responsible for the observed periodicity. The amplitude of the variations of about 7 per cent (see Figure 2) requires a precession angle dependent on the disk inclination according to $\\Delta i = \\frac{{\\rm 0.07}}{{\\rm cot}\\ i}$. For instance, for a disk inclination of 90 (45) degrees the precession angle is 32 (4) degrees. The angular shape of the light curve may reflect the expected disk warp. The second peak at 2100 seconds in the power spectrum and the possible third one at about 3000 seconds may be related to the warp as well. The precessing frequency produced by the Bardeen-Petterson effect depends on the disk radius $r$ as $$ \\omega =  \\frac{2G}{c^2} J r^{-3} \\eqno(2) $$ where $J$ is the angular momentum of the black hole. The maximum angular momentum of a Kerr black hole is $$ J_{\\rm max} = G\\ M^2\\ c^{-1} \\eqno(3) $$ For the ratio between $J$ and $J_{\\rm max}$ we get  $$ \\frac{J}{J_{\\rm max}} = \\frac{\\omega\\ c^3}{2 G^2} \\cdot \\frac{r^3}{M^2} \\eqno(4) $$ Equ. 4 can be used to estimate the lower limit of the black hole mass in Mrk 766: Taking $\\omega \\rm = 1.5 \\cdot 10^{-3}$ and  the black body disk radius of $\\rm 2 \\cdot 10^{12}\\ cm$ as an rough estimate for $r$, we find that a black hole mass of $M \\rm > 3 \\cdot 10^6$ solar masses is required not to violate the angular momentum limit.    Presently, we are unable to differentiate  between a strictly periodic behavior and quasi-periodic oscillations with high precision. We also need better data for the light curve and the power spectrum in order to constrain the models. In this context the forthcoming 150 ksec {\\it XMM-Newton} observation will be most useful."
    },
    "0010/gr-qc0010092_arXiv.txt": {
        "abstract": "In this work we have investigated the possibility of having a late time accelerated phase of the universe, suggested by recent supernova observation, in the context of Brans Dicke (BD) theory with potential having a time dependent mass squared term which has recently become negative and a matter field. We find that while a perfect fluid (pressureless and with pressure) cannot support this acceleration, a fluid with dissipative pressure can drive this late time acceleration for a simple power law expansion of the universe. We have also calculated some cosmological parameters in our model to match with observations. ",
        "introduction": "A lot of activity has been triggered by two recent observations \\cite{Rperl,Rgar} on the explosion of type Ia Supernovae. These datas favour the existence of a new kind of matter with positive energy density dominant at present universe and is also responsible for the present acceleration of the universe accounted by its negative pressure. This along with the observed location of the first acoustic peak of CMB temperature fluctuation corroborated by the latest BOOMERANG and MAXIMA data \\cite{Rber,Rhan}, favours a spatially flat universe whose energy density is dominated by a cosmological constant like term. Obviously the first natural choice to represent such special matter was the cosmological constant $\\Lambda$\\cite{Rbah,Rsah}. For a flat matter dominated universe with $\\Lambda$ having $\\Omega_\\Lambda\\sim 0.72$ in Einstein gravity best fits the data sets. But the candidature of $\\Lambda$ as the constituent of the major energy density is troubled by the fact that it has an energy scale which is $\\sim 10^{-123}$ lower than normal energy scale predicted by the most particle physics models. So to find some alternative candidate for this acceleration a dynamical $\\Lambda$\\cite{Rcal} in the form of a scalar field with some self interacting potential\\cite{Rpeebetc} is considered whose slowly varying energy density mimics an effective cosmological constant. The idea of this  candidate, called {\\it quintessence}\\cite{Rcal}, is borrowed from the inflationary phase of the early universe, with the difference that it evolves at a much lower energy scale. The energy density of this field, though dominant at present epoch, must remain subdominant at very early stage and has to evolve in such a way that it becomes comparable with the matter density $\\Omega_m$ now. This type of specific evolution, better known as ``{\\it cosmic coincidence}''\\cite{Rstei} problem, needs several constraints and fine tuning of parameters for the potential used to model quintessence with minimally coupled scalar field. A new form of quintessence field called ``{\\it tracker field}''\\cite{Rzla} has been proposed to solve the cosmic coincidence problem. It has an equation of motion with an attractor like solution in a sense that for a wide range of initial conditions the equation of motion converges to the same solution. \\\\  There are a number of quintessence models which have been put forward and most of which involve minimally coupled scalar field with different potentials dominating over the kinetic energy of the field. A purely exponential potential is one of the widely studied cases \\cite{Rfer}. Inspite of the other advantages the energy density is not enough to make up for the missing part. Inverse power law is the other potential (\\cite{Rpeebetc}-\\cite{Rzla}) that has been studied extensively for quintessence models, particularly for solving the cosmic coincidence problem. Though the problems are resolved successfully with this potential, the predicted value for the equation of state for the quintessence field, $\\gamma_Q$, is not in good agreement with the observed results. In search of proper potentials that would eliminate the problems, new types of potentials, like $V_0[\\cos h \\lambda\\phi-1]^p$\\cite{Rsah2} and $V_0\\sin h (\\alpha\\sqrt k_0\\Delta\\phi)^\\beta$\\cite{Rsah,Rlop} have been considered, which have asymptotic forms like the inverse power law or exponential ones. Different physical considerations have lead to the study of other types of the potentials also\\cite{Ruzan}. Recently Saini {\\it et al}  \\cite{Rsai} have reconstructed the potential in context of general relativity and minimally coupled quintessence field from the expression of the luminosity distance $d_L(z)$ as function of redshift obtained from the observational data. However, none of these potentials are entirely free of problems. Hence, there is still a need to identify appropriate potentials to explain current observations \\cite{Rfer}.\\\\  Most of the studies regarding accelerated expansion have been done with a minimally coupled scalar field representing the quintessence. It has been recently shown by Pietro and Demaret\\cite{Rpietro} that for constant scalar field equation of state, which is a good approximation for a tracker field solutions, the field equations and the conservation equations strongly constrain the scalar field potential, and most of the widely used potential for quintessence, such as inverse power law one, exponential or the cosine form, are incompatible with these constraints. The minimally coupled self interacting models will also be ruled out if the observations predict that the missing component of the energy density obeys an equation of state $p=\\gamma\\rho$ with $\\gamma<-1 (\\rho\\geq 0)$ , and these sort of equation of state is in reasonable agreement with different observations \\cite{Rcald}. Also the inequality $dH^{2}(z)/dz\\geq 3\\Omega_{m0}H_{0}(1+z)^{2}$ should satisfy for minimally coupled scalar field and its violation will certainly point towards a theory of non Einstein gravity such as scalar tensor theories where the scalar field is non minimally coupled to gravity.\\\\  There have been quite a few attempts of treating this problem with the non-minimally coupled scalar fields. Scaling attractor solutions are available in the literature with the exponential\\cite{Ruzan} and power law\\cite{Ruzan,Rlid} potentials in non-minimally coupled theories. Faraoni\\cite{Rfar} have studied different potentials with a non-minimal coupling term $\\psi R{\\phi^2\\over{2}}$ for the present acceleration. There have been different approach also for solving the problem in general scalar tensor theory, sometimes called {\\it extended} or {\\it generalised} quintessence, not only because this theory is considered to be the most natural alternative to general relativity, there are other strong motivations\\cite{Respo} also. People like Bertolo {\\it et al}\\cite{Rbert}, Bertolami {\\it et al} \\cite{Rberto}, Ritis {\\it et al}\\cite{Rriti} have found tracking solutions in scalar tensor theories with different types of power law potential. In another work Sen {\\it et al}\\cite{Rsens} have found the potential relevant to power law expansion in Brans Dicke cosmology. Like Saini {\\it et al}\\cite{Rsai}, Boisseau {\\it et al} \\cite{Rboi} have reconstructed the potential from the luminosity-redshift relation available from the observations in context of scalar tensor theory.\\\\  Very recently McDonald \\cite{Rmcdo} has investigated the possibility of modelling a dynamical cosmological constant with  a scalar field which has undergone a very recent phase transition. For this he has considered a standard $\\phi^{4}$ potential for the scalar field with an additional time dependent mass squared term in the potential which has become negative very recently. For this kind of model, phase transition occurs very recently at redshift $z\\leq 1.2$.\\\\  In this paper we have investigated whether non minimally coupled self interacting scalar field such as a Brans-Dicke (BD) type scalar field with this kind of potential can successfully drive the late time acceleration for the flat universe. In the context of Brans Dicke(BD) theory\\cite{Rbd} with a self interacting potential and a matter field, the action is given by \\be {\\cal{S}}=\\int d^4x \\sqrt{-g}[\\phi R-{\\omega\\over{\\phi}}\\phi^\\alpha\\phi_\\alpha -V(\\phi)+{\\cal{L}}_m]\\label{action} \\ee (We have chosen the unit $8\\pi G_0=c~=~1$.)\\\\ We have chosen the $\\phi^{4}$ potential with a time dependent mass squared term which has already become negative after a phase transition in recent time \\cite{Rmcdo}: \\be V(\\phi)=\\lambda \\phi^4-\\mu^2(t)\\phi^2 \\label{pot} \\ee where \\be \\mu^2(t)=\\bar{\\mu_0}^2\\left(\\frac{R_c}{R}\\right)^n=\\frac{\\mu_0^2}{R^n} \\ee ${\\mu_0^2}=\\bar{\\mu_0}^2{R_c}^n$. $\\lambda$ is a constant and $n$ is an integer. The time dependent mass squared term with integer $n$ can arise naturally in plausible models and one can have a detailed discussion in \\cite{Rmcdo}. \\\\  As a matter field we would consider first perfect fluid and then a fluid having negative pressure. An effective negative pressure and hence an acceleration can be achieved by dissipative mechanism modelled commonly by fluid viscosities. It has been proposed recently that the CDM must self interact in order to explain the detailed structure of the galactic halos\\cite{Rsperg}. This self interaction will naturally create a viscous pressure whose magnitude will depend on the mean free path of the CDM particles. An effective negative pressure in CDM can also be created from cosmic anti friction which is closely related to particle production out of gravitational field\\cite{Rzim}. Since the negative pressure can be modelled in different ways, we are not apriori assuming any specific model for this negative pressure.\\\\  In this work we find that it is not possible to have a late time power law accelerated expansion when the CDM is a perfect fluid, but a dissipative CDM fluid in BD cosmology with such a potential like (\\ref{pot}) can successfully drive a late time accelerated expansion. In the next section we treat the field equations and find the solutions in both the cases. We also calculate some cosmological parameters to match the accelerated model with observation. The third section is the concluding section where we have discussed different features of this model. ",
        "conclusions": "This work investigates the possibility of getting an accelerated universe in context of BD theory with a $\\phi^{4}$ potential having a time dependent mass squared term  and a matter field. In this work we have not used quintessence field to trace the missing energy. The BD scalar field , which is a geometric scalar field, plays the role of dynamical $\\Lambda$ and provides that missing energy. It is found that for a simplistic approach of power law expansion ($\\sim t^{\\alpha}$) a perfect fluid kind of matter (both pressureless and with pressure) cannot support a late time acceleration of the universe, if the scalar field have potential given by equation (\\ref{pot}). But a matter with a dissipative effect can provide the acceleration that agrees with the observational data sets. The dissipative effect accounted by the negative pressure can be modelled in two ways according to the recent investigations\\cite{Rzim,Rchim}. Particle production out of gravitational field can give rise to negative pressure while energy can also be dissipated by bulk viscous effect between the CDM particles. In this work though no particular model is considered for the origin of the negative pressure, it is found that $\\nu>1$. This is important for hydrodynamical description if the CDM is assumed to be a conventional dissipative fluid. We have also calculated different parameters like the time variation of gravitational coupling, age of the universe  and the luminosity-distance redshift relation. All of these cosmological parameters agrees quite well with the recent observations.\\\\  The accelerated solution depends crucially upon two parameters: $\\alpha$ and BD parameter $\\omega$, both of which are constrained by different physical conditions. Different combinations of $\\alpha$ and $\\omega$ can produce the required values for $\\gamma_\\phi$ ($-0.6>\\gamma_\\phi>-1$) and $\\Omega_m$  ($\\sim0.3$) that tallies with present observation suggesting acceleration. Small $\\alpha$ values restricts $\\omega$ to small negative values and support the late time acceleration scenario quite successfully.The cosmological parameters calculated with the small value of $\\alpha$ agrees with observations quite well. Many references \\cite{Rkapling,Rdiaz,Rsethi} are available in the literature where it has been shown that this value should be very close to $1$ to be consistent with observation. In one of our recent work\\cite{Rsen} it is shown that for simple power law expansion $(\\sim t^\\alpha)$ of the universe, the best fit value of $\\alpha$ with SNIa data is approximately 1.25, and due to (\\ref{17a}) and (\\ref{27a}), this will restrict $\\omega$ to small value. But this clearly contradicts the solar system limit $\\omega>600$. One should note that in our model, large value of $\\omega$ consistent with solar system limit,  is not restricted either by physical conditions such as positivity of energy density or by the requirements of specific ranges for $\\gamma_{\\phi}$ or $\\Omega_{m}$ consistent with the observations. Only the fact, that $\\alpha$ is not large, is the prediction of the datas obtained so far and this constrains $\\omega$ to small value. If future observations predicts large value for $\\alpha$ then that can also be accommodated in our model with a large value for $\\omega$.\\\\  One should also note that, for $\\alpha<1.33$ which is consistent with the present data, the dissipative pressure is not sufficient to drive the acceleration alone, and the BD scalar field along with the dissipative pressure in the CDM drives the acceleration whereas the BD scalar field plays the role of the missing components of the universe.\\\\  It is also expected that after the phase transition, one expects $\\phi$ to roll down from $\\phi=0$ taking different values in different directions causing large scale inhomogeneities. But it was argued by McDonald \\cite{Rmcdo} that after this recent phase transition the universe will be filled by non topological objects like ``{\\it axions}'' whose radius $r_{\\phi}$ is much smaller than $10 Mpc$ causing these axions to behave like a smooth dark energy components.\\\\  The range of $\\omega$, we obtained in our calculations in order to fulfil different physical conditions and also to have correct range of values for $\\gamma_{\\phi}$ , $\\alpha$, and $\\Omega_{m}$, is consistent with the range obtained by other authors \\cite{Rsens,Rsen,Rban}. Also, it can be seen in the figure 3 that for different values of $\\omega$ and $\\alpha$, it is difficult to distinguish between models  upto $z\\sim 1$. Hence our model is not very much fine tuned as far as the parameter $\\alpha$ and $\\omega$ are concerned. \\\\  It is also important to note that as $\\alpha$ remains constant throughout the age of the universe. This essentially means that the universe always accelerates for $\\alpha>1$ , which seriously contradicts the primeval nucleosynthesis scenario. As we have mentioned earlier that one way to overcome this problem is to consider $\\omega$ to be function of the scalar field $\\phi$ \\cite{Rban}. A choice of $\\omega$ (polynomial function of $\\phi$) can give a decelerating radiation era as well as accelerating matter dominated era. But then also $\\omega$ asymptotically acquires a small negative value for an accelerating universe at late time. In most of the  investigations done in scalar tensor theory\\cite{Rsens,Rsen,Rban} such conclusions have been arrived. Only exception, so far in our knowledge, is that done by Bertolami and Martins\\cite{Rberto}, where in Brans Dicke cosmology with a $\\phi^2$ potential, the solution of an accelerated universe is obtained with large $|\\omega|$. But there the positive energy condition on both matter and the the scalar field have not been considered. There are also other evidences in literature where small $|\\omega|$ has been supported. In the extended inflationary model by La and Steinhardt ~\\cite{LS}, the required value for $\\omega$ is 20. The structure formation in scalar tensor theory also contradicts the solar system bound on $\\omega$\\cite{Rgaz}.Thus the problem seems to appear in different scales (astronomical and cosmological). The theory has been tested by experiments so far only in the astronomical scales and no experiment had been done in cosmological scale as yet. And so the problem occurs in finding the compatibility between astronomical observation and cosmological requirements. Considering $\\omega$ to be a variable to have both decelerating and accelerating phases at different epochs, while large $\\omega$ values occur due to local inhomogeneities in astronomical scale to satisfy the solar system bound, can be a complete investigation and may give a satisfactory answer to this question."
    },
    "0010/astro-ph0010536_arXiv.txt": {
        "abstract": "The direct detection of  dark matter is central to particle physics and cosmology. Current fashionable supersymmetric models provide a natural dark matter candidate which is the lightest supersymmetric particle (LSP). Such models combined with fairly well understood physics like the quark substructure of the nucleon and the structure of the nucleus (form factor and/or spin response function), permit the evaluation of the event rate for LSP-nucleus elastic scattering. The thus obtained event rates are, however, very low or even undetectable. So it is imperative to exploit characteristic signatures, like the modulation effect, i.e. the dependence of the event rate on  the earth's annual motion, and the directional rate, i.e its dependence on the direction of the recoiling nucleus. In this paper we do this using various velocity distributions, isothermal (symmetric as well as only axially asymmetric) and non isothermal (e.g. due to caustic rings). ",
        "introduction": "The consideration of  exotic dark matter has become necessary in order to close the Universe \\cite {KTP}$^,$ \\cite{Jungm}. Furthermore in in order to understand the large scale structure of the universe one has to consider matter made up of particles which were non-relativistic at the time of freeze out, i.e  cold dark matter (CDM). The COBE data ~\\cite{COBE} suggest that CDM is at least $60\\%$ ~\\cite {GAW}. Recent data from the High-z Supernova Search Team \\cite{HSST} and Supernova Cosmology Project  ~\\cite {SPF} $^,$~\\cite {SCP} suggest the necessity of the cosmological constant $\\Lambda$. In fact the situation can adequately be described by  a barionic component $\\Omega_B=0.1$ along with the exotic components $\\Omega _{CDM}= 0.3$ and $\\Omega _{\\Lambda}= 0.6$. In another analysis Turner \\cite {Turner} gives $\\Omega_{m}= \\Omega _{CDM}+ \\Omega _B=0.4$. Since the non exotic component cannot exceed $40\\%$ of the CDM ~\\cite{Jungm}$^,$~\\cite {Benne}, there is room for the exotic WIMP's (Weakly  Interacting Massive Particles). In fact the DAMA experiment ~\\cite {BERNA2} has claimed the observation of one signal in direct detection of a WIMP, which with better statistics has subsequently been interpreted as a modulation signal ~\\cite{BERNA1}.  The above developments are in line with particle physics considerations. Thus, in the currently favored supersymmetric (SUSY) extensions of the standard model, the most natural WIMP candidate is the LSP, i.e. the lightest supersymmetric particle. In the most favored scenarios the LSP can be simply described as a Majorana fermion, a linear combination of the neutral components of the gauginos and Higgsinos \\cite{Jungm}$^,$~\\cite{JDV}$^-$\\cite{Hab-Ka}.  Since this particle is expected to be very massive, $m_{\\chi} \\geq 30 GeV$, and extremely non relativistic with average kinetic energy $T \\leq 100 KeV$, it can be directly detected ~\\cite{JDV}$^-$\\cite{KVprd} mainly via the recoiling of a nucleus (A,Z) in the elastic scattering process: \\begin{equation} \\chi \\, +\\, (A,Z) \\, \\to \\, \\chi \\,  + \\, (A,Z)^* \\end{equation} ($\\chi$ denotes the LSP). In order to compute the event rate one needs the following ingredients:  1) An effective Lagrangian at the elementary particle (quark) level obtained in the framework of supersymmetry as described in Refs.~\\cite{Jungm}, Bottino {\\it et al.} \\cite{ref2}, \\cite{Hab-Ka}, Kane {\\it et al.}, Castano {\\it et al.} and Arnowitt {\\it et al}. Our own SUSY input parameters will appear elsewhere \\cite {Gomez} : 2) A procedure in going from the quark to the nucleon level, i.e. a quark model for the nucleon. The results, for the scalar and the isoscalar axial couplings, depend crucially on the content of the nucleon in quarks other than u and d ~\\cite{Dree,Chen}.  3) Compute the relevant nuclear matrix elements~ \\cite{Ress}$^-$\\cite{DIVA00}.  Since the obtained rates are very low, one would like to be able to exploit the modulation of the event rates due to the earth's revolution around the sun. One also would like to know the directional rates, by observing the nucleus in a certain direction, which correlate with the motion of the sun around the center of the galaxy.  The purpose of our present review is to focus on this last point along the lines suggested by our recent work \\cite{JDV99,JDV99b}. ",
        "conclusions": "We have studied the parameters describing the event rates for direct detection of SUSY dark matter and, in particular, two characteristic signatures, which will aid the experimentalists in reducing background: The non directional modulated event rates, which are correlated with the motion of the Earth  and the directional event rates, which are correlated with both the velocity of the sun and the velocity of the Earth.   In the case of non isothermal models \\cite{SIKIVIE} from Table 1 we see that the maximun no longer occurs around June 2nd, but about six months later. The difference between the maximum and the minimum is about $4\\%$, smaller than that predicted by the symmetric isothermal models \\cite{JDV99,JDV99b}. In the case of the directional rate we found that the rates depend on the direction of observation. The biggest rates are obtained, If the observation is made close to the direction of the sun's motion. The directional rates are suppressed compared to the usual non-directional rates by the factor $f_{red}=\\kappa/(2 \\pi)$. We find that $\\kappa=r^u_z \\simeq 0.7$, if the observation is made in the sun's direction of motion, while $\\kappa\\simeq 0.3$ in the opposite direction. The modulation is a bit larger than in the non-directional case, but the largest value, $8\\%$, is not obtainrd along the sun's direction of motion, but in the x-direction (galactocentric direction).  In the case of the isothermal models we restricted our discussion to the directional event rates. The reduction factor is now given by the parameter $f_{red}=t_0/(4 \\pi~t)=\\kappa/(2 \\pi)$. We find that $\\kappa$ is around 0.6 for no asymmetry and around 0.7 for maximum asymmetry ($\\lambda=1.0$). In other words it is not very different from the naively expected $f_{red}=1/( 2 \\pi)$. The modulation of the directional rate increases with the asymmetry parameter $\\lambda$ and it depends, of course, on the direction of observation. For $Q_{min}= 0$ it can reach values up to $23\\%$. Larger values, up to $35\\%$, are possible for large values of $Q_{min}$, but they occur at the expense of the total number of counts.  In all cases our results indicate that $t$ for large reduced mass deviates from unity. Thus, if one  attempts to extract the LSP-nucleon cross section from the data to compare it with the predictions of SUSY models, one must take $t$ into account.  {\\it Acknowledgments:} One of the authors (JDV) would like to acknowledge partial support of this work by $\\Pi $ENE$\\Delta $ 1895/95 of the Greek Secretariat for Research and  TMR No ERB FMAX-CT96-0090 of the European Union."
    },
    "0010/astro-ph0010426_arXiv.txt": {
        "abstract": "We present a model for the steady state winds of super-Eddington systems. These radiatively driven winds are expected to be optically thick and clumpy as they arise from a porous atmosphere.  The model is then used to derive the mass loss of bright classical novae. The long duration super-Eddington outflows that are clearly observed in at least two cases (Nova LMC 1988 \\#1 \\& Nova FH Ser) are naturally explained. Moreover, the predicted mass loss and temperature evolution agree nicely with the observations, as do additional features. $\\eta$ Car is then used to double check the theory which predicts the observed mass shed in the great eruption. ",
        "introduction": "$\\eta$ Car was super-Eddington (hereafter SE) during its 20 year long eruption (e.g., Davidson \\& Humphreys 1997). Yet, its observed mass loss and velocity are inconsistent with a homogeneous solution for the wind (Shaviv 2000b). Basically, the sonic point obtained from the observed conditions necessarily has to reside too high in the atmosphere, at an optical depth of only $\\sim 1$ to $\\sim 300$, while the critical point in a homogeneous atmosphere necessarily has to reside at significantly deeper optical depths. The inconsistency arises because the sonic and critical points have to coincide in a steady state solution.  A solution was proposed in which the atmosphere of $\\eta$ Carinae is inhomogeneous, or porous (Shaviv 2000b). The inhomogeneity is a natural result of the instabilities of atmospheres that are close to the Eddington luminosity (Shaviv 2000a). The inhomogeneities, or ``porosity'', reduce the effective opacity and increase the effective Eddington luminosity (Shaviv 1998).  Here, we are interested in understanding the wind generated in cases in which the luminosity is SE.  To do so, it is advantageous to find a class of objects for which better data than for $\\eta$ Car exists. One such class of objects is novae. Their advantage over other types is that their mass and luminosity is known better than all other SE objects, allowing a detailed analysis. ",
        "conclusions": "We have tried to present the following coherent picture which explains the observed existence of SE atmospheres: Homogeneous atmospheres becomes inhomogeneous as the the radiative flux approaches the Eddington limit. This is due to a plethora of instabilities. The particular governing instability depends on the details of the atmosphere. Due to the inhomogeneity, the effective opacity is then reduced as it is easier for the radiation to escape; consequently, the effective Eddington limit increases.  SE configurations are now possible because the bulk of the atmosphere is effectively sub-Eddington. Very deep layers advect the excess total luminosity above Eddington by convection. Higher in the atmosphere, where convection is inefficient, the Eddington limit is effectively increased due to the reduced effective opacity. The top part of the atmosphere, where perturbations of order of the scale height become optically thin, has however to remain SE. Thus, these layers are pushed off by a continuum driven wind.  By identifying the location of the critical point of the outflow, one can obtain a mass-loss luminosity relation. The relation, given by eq.~1 is the main result of the paper. The relation has a universal dimensionless parameter which should be of order or somewhat larger than unity.  To check the result, we analyzed 2 novae and the star $\\eta$ Car. Although the two types of systems are markably different, as they have masses, luminosities and massloss rates which differ by orders of magnitude, the wind mass loss and the wind parameter are found to be in good agreement with the theoretical expectation. Moreover, the results are completely consistent with clumping --- a natural prediction of SEWs since the atmospheric layers beneath the sonic point are predicted to be inhomogeneous.  We also identify the occurrence of the ``transition phase'' observed in a majority of the novae with the advance of the atmosphere into the ``no steady state region''.  As the nova explosion progresses, its luminosity and radius decline. However, if the radius decreases too quickly, at some point the SEW predicted will be too heavy for the luminosity at the base to push to infinity. No steady state will then exist. The inconsistency might explain the strange behaviors observed in the transition phase of different novae.  The wind model presented is by no means a complete theory for novae since it cannot predict {\\em ab initio} the luminosity at the base of the wind. To obtain the latter, one needs to solve for the complete evolution of a nova taking into account the porosity and lowered opacity in the nova atmosphere. One expects that the lowered opacity increases the luminosity obtained in the core-mass luminosity relation, and super-Eddington values therefore arise naturally."
    },
    "0010/astro-ph0010610_arXiv.txt": {
        "abstract": "Various independent ways of constraining key cosmological parameters yielded a consensus range of values which indicates that at the present epoch the bulk of the universe's baryons is in the form of a warm ($\\sim$ 10$^6$ K) gas - a temperature regime which renders them difficult to detect.  The discovery of EUV and soft X-ray excess emission from clusters of galaxies was originally interpreted as the first direct evidence for the large scale presence of such a warm component. We present results from an EUVE Deep Survey (DS) observation of the rich cluster Abell 2199 in the Lex/B (69 - 190 eV) filter passband. The soft excess radial trend (SERT), shown by a plot against cluster radius $r$ of the percentage EUV emission $\\eta$ observed above the level expected from the hot intracluster medium (ICM), behaves as a simple function of $r$ which decreases monotonically towards $r = 0$; it smoothly  turns negative at $r \\sim$ 6 arcmin, inwards of this radius the EUV is absorbed by cold matter with a line-of-sight column density of $\\geq$ 2.7 $\\times$ 10$^{19}$ cm$^{-2}$. The centre of absorption is offset from that of the emission by $\\sim$ 1 arcmin, and the area involved is much larger than that of the cooling flow.  These facts together provide strong evidence for a centrally concentrated but cluster-wide distribution of clumps of cold gas which co-exist with warm gas of similar spatial properties. Further, the simultaneous modeling of EUV and X-ray data requires a warm component even within the region of absorption. The phenomenon demonstrates a three phase ICM, with the warm phase estimated to be $\\sim$ 5-10 times more massive than the hot. ",
        "introduction": " ",
        "conclusions": ""
    },
    "0010/astro-ph0010560_arXiv.txt": {
        "abstract": "{Observations have revealed strong evidence for powerful jets in the Low/Hard states of black hole candidate X-ray binaries. Correlations, both temporal and spectral, between the radio -- infrared and X-ray bands suggest that jet synchrotron as well as inverse Compton emission could also be significantly contributing at higher frequencies.  We show here that, for reasonable assumptions about the jet physical parameters, the broadband spectrum from radio through X-rays can be almost entirely fit by synchrotron emission. We explore a relatively simple model for a relativistic, adiabatically expanding jet combined with a truncated thermal disk conjoined by an ADAF, in the context of the recently discovered black hole binary XTE J1118+480.  In particular, the X-ray power-law emission can be explained as optically thin synchrotron emission from a shock acceleration region in the innermost part of the jet, with a cutoff determined by cooling losses. For synchrotron cooling-limited particle acceleration, the spectral cutoff is a function only of dimensionless plasma parameters and thus should be around a ``canonical'' value for sources with similar plasma properties.  It is therefore possible that non-thermal jet emission is important for XTE J1118+480 and possibly other X-ray binaries in the Low/Hard state. ",
        "introduction": "Observations are providing increasing evidence that black hole candidate (BHC) X-ray binaries (XRBs) produce powerful collimated outflows when in the Low/Hard X-ray state (LHS).  This state is characterized by a nonthermal power-law in the X-ray band and little, if any, thermal disk contribution (e.g., \\citeNP{Nowak1995,Poutanen1998}). At least three systems in the LHS (Cyg X-1, 1E 1740.7-2942 and GRS 1758-258) have directly resolved radio jets on scales from AU to parsecs.  However, XRB jets also reveal themselves in the broadband LHS spectra with a flat-to-inverted radio synchrotron spectrum, analogous to the signature emission of jets in compact radio cores of AGN \\cite{BlandfordKoenigl1979,HjellmingJohnston1988,FalckeBiermann1999}. Furthermore, new data suggest a continuation of this radio synchrotron emission to much higher frequencies.  The spatial, spectral and temporal evidence for powerful jets from XRB BHCs in the LHS are compiled and discussed in \\citeN{Fender2001}.  We know that jets play a significant role in the emission of Active Galactic Nuclei (AGN), even dominating the spectrum from radio through TeV $\\gamma$-rays in the case of BL Lacs, with emission from optically thin synchrotron up to even $100$ keV and higher (e.g. \\citeNP{Pianetal1998}). By analogy, if the flat,  optically thick synchrotron spectrum in XRBs, commonly attributed to jets, indeed extends into the NIR and optical regimes \\cite{Fender2001}, one would expect a corresponding optically thin power-law from shock acceleration at even higher frequencies. This is irrespective of whether the jet emission is boosted or not, since optically thick and optically thin emission would have similar Doppler factors.  Shock acceleration is likely to be present in XRBs, given that optically thin power-law spectra are observed during their radio outbursts (e.g., \\citeNP{FenderKuulkers2001}, and refs. therein).  Still, the majority of current models for the broadband (X-ray) spectra of BHC XRBs focus only on the contribution of thermal disk plus coronal inverse Compton (IC) emission (for a review, see \\citeNP{Poutanen1998}) and ignore any jet contribution, even though jets can be an integral part of X-ray binaries and their systems (e.g., \\citeNP{EikenberryMatthewsMorgan1998,MirabelRodriguez1999,Fender2001}).  As an example for the possible importance of jet emission, we use the recently discovered XRB XTE J1118+480 \\cite{Remillardetal2000}, which has been observed in the radio through X-rays (see \\citeNP{Hynesetal2000}, hereafter H00, \\citeNP{Fenderetal2001}, hereafter F01, and references therein), and is also at high enough Galactic latitude to allow the first ever EUV detections of an X-ray transient (H00).  The system is likely a BHC in the LHS (H00;\\citeNP{RevnivtsevSunyaevBorozdin2000,Woodetal2000}). Although jets were not directly resolved with MERLIN to a limit of $<65(d/{\\rm kpc})$ AU (at 5 GHz; F01), its radio emission shows the flat characteristic jet spectrum. ",
        "conclusions": "{}From our modeling we conclude that in the LHS of XTE J1118+480 a significant jet contribution from radio through IR is present, and with reasonable assumptions and a small efficiency could extend even up to the hard X-rays.  We cannot exclude that XTE J1118+480, with its rather low luminosity, is a special source. However, our model is scalable and thus likely applicable to other XRBs in the LHS, when a flat radio spectrum is present.  An important element of the model is that the inner ADAF-like flow is very hot and radiatively inefficient, injecting (mildly) relativistic electrons into the jet and allowing high electron energies because of low IC cooling.  With a more dominant disk contribution, e.g., in the High/Soft state, the jet spectrum would also change and possibly disappear due to increased cooling.  Interestingly, in the case of a low disk contribution, synchrotron-dominated cooling provides a natural ``canonical'' cutoff around 100 keV, which only depends on the dimensionless plasma parameters $\\xi$ and $u_{\\rm sh}/c$.  Small variations in the spectral index of the synchrotron power-law could explain the observed time lags in some sources \\cite{KotovChurazovGilfanov2001}.  The jet model for XTE J1118+480 may therefore provide an interesting new perspective for the modeling of XRBs in general."
    },
    "0010/astro-ph0010083_arXiv.txt": {
        "abstract": "Detailed VLA observations have been gathered for a number of sources classified as either BL Lacs or galaxies, derived from the REX survey. We focus in particular on the sources identified by us, for which we have in hand homogeneous optical data, to study in more detail than allowed by the NVSS the radio properties of these sources in the framework of AGN unified models. ",
        "introduction": "The REX survey (Radio Emitting X-ray sources; see Maccacaro et al. 1998, Caccianiga et al. 1999 for details of the survey) is based on pointed ROSAT/PSPC observations and NVSS (NRAO VLA Sky Survey, Condon et al. 1998) data. About 500 of the $\\sim 1600$ REX sources have an optical identification obtained working down from the highest X-ray fluxes, in order to have larger and larger completely identified subsamples  (see Caccianiga et al., these proceedings for the first results on the Xray--Bright-REX sample of BL Lacs).  \\begin{table} \\caption{Identification breakdown as of July 2000} \\begin{center} \\begin{tabular}[h]{|l r r r |} \\hline ID \t & New  & Literature & Total \\\\ AGN \t & 96  \t& 138  & 234\t \\\\ BL+cand. & 40 \t& 32   & 72\t \\\\ Galaxies & 55 \t& 121  & 176\t \\\\ \\hline \\end{tabular} \\end{center} \\label{id} \\end{table}   AGN and BL Lacs are the most common counterparts, followed by galaxies, either isolated or in clusters. The objects without emission lines are classified according to the amount of the CaII discontinuity in our optical spectral data. Values larger than 40\\% define galaxies. BL Lacs are defined as having a CaII discontinuity smaller than 25\\%, while objects with CaII discontinuity between 25\\% and 40\\% are defined as BL Lac candidates. Most of the galaxy classification from the literature are instead based on the optical extended morphology. In Table 1 % we list the updated identification breakdown. ",
        "conclusions": "From the REX survey we have extracted a number of BL Lacs and galaxies, to study their radio properties in the context of Unified Models. Our results seem to imply that objects classified spectroscopically as galaxies, i.e. that do not have emission lines and have a large predominance of stellar light, as measured form the CaII absorption, are not the same objects that are defined as classical radio galaxies, i.e. with a large radio structure of jets and lobes in the plane of the sky. The resolved morphology contrasts with the BL Lac classification from the optical spectrum in at least one striking case.  The two measures of orientation: core domninance and CaII depression, from the radio and optical band respectively, seem to be only weakly correlated, suggesting that a revision of the unifying models, in which not only the orientation contributes to change one class into the other, might be necessary."
    },
    "0010/astro-ph0010030_arXiv.txt": {
        "abstract": "Long-time behavior of spatial power-law correlation in one-dimensional self-gravitating system is numerically investigated. % The power-law structure persists even after the system is virialized. The structure gradually disappears with energy exchange among particles. Lifetime of the power-law structure is estimated to be proportional to the system size.  PACS 05.45 ",
        "introduction": "Spatial structure in many-body systems is an important subject in many branches of physics, such as molecular systems, gravitational systems, and so on. For various types of such structures, their origins are interesting subject of study. Some of them may be realized as equilibrium distribution, and others may be already set at initial conditions.  Recently we have discovered that spatial structure with power-law correlation spontaneously emerges from uniformly random initial conditions in one-dimensional self-gravitating system~\\cite{hktk-1}. What is important in this novel phenomenon is that the spatial structure is not given at the initial condition, but dynamically  created from a state without spatial  correlation. Succeeding research clarified that the structure is created first in small spatial scale then grows up to large scale through hierarchical clustering~\\cite{hktk-2}.  The question to be discussed now is the properties of the state which has power-law correlation. In particular we have to clarify whether the state persists for long time, or just a short transient and disappears quickly. If the state lasts for long time, we have good chance to observe it and it has much importance. On the other hand, if the state disappears quickly, it may not be observable easily and it is  less important to the whole dynamics of the system.   Another important question is  on the relaxation process~( See \\cite{Miller-2,sheet-tkg-1,sheet-tgk-2} and references therein.). Since the system is an isolated many-body system without dissipation, one would expect that the system will eventually relax to thermal equilibrium for sufficiently long time scale. Hence  relation between decay (if it does) of power-law structure and relaxation processes of thermalization is also important.  The purpose of this paper is to investigate the long-time behavior of the power-law structure; its lifetime and mechanism of relaxation. The model is introduced in the next section. In section \\ref{sec:results} we introduce several quantities which represent relaxation and investigate their dynamics to clarify the mechanism by which the power-law structure disappears. The final section is devoted to summary and discussions. ",
        "conclusions": "In this paper we numerically investigated the long-time behavior of power-law correlation structure~\\cite{hktk-1,hktk-2} in one-dimensional self-gravitating system. The power-law correlation exists even after the system is virialized. In this sense the structure persists for long time. The lifetime of the structure  is, however, not infinite and it gradually fades away. By comparing the timescale of the lifetime with other time scales of relaxation we see that the decay of the power-law structure is not caused by virialization process but by energy exchange among particles. The one-particle energy distribution $\\nu(\\varepsilon)$  is invariant through the relaxation of the structure, hence it is the microscopic relaxation~\\cite{sheet-tkg-1,sheet-tgk-2}. The system is not stationary if we see it by the exponent $\\alpha$ of correlation, but it is at the same time quasi-stationary if we see it by the energy distribution $\\nu(\\varepsilon)$ after the system is virialized.  The fact that power-law correlation structure lasts long implies that the structure is important in the whole  dynamical evolution of the system and we may be able to observe such structure in other systems even if the system is virialized.  Relation between the decay of power-law correlation and microscopic relaxation has an important consequence. It is known that the microscopic relaxation time depends linearly on the system size $\\displaystyle t_{micro}\\propto N$~\\cite{sheet-tgk-2}. Hence it is likely that the average lifetime of the power-law correlation has similar dependence on $N$, which is quite long for large $N$.    Let us now summarize schematically the entire life of the spatial power-law correlation structure in the model (\\ref{eq:sheet-hamiltonian}) clarified through a series of our works. \\begin{enumerate} \\item From  initial condition with zero velocity dispersion~\\cite{hktk-1}, \\ spatial structure with power-law correlation emerges in small spatial scale~\\cite{hktk-2}. Typical example of such initial condition is $\\forall i \\  v_i = 0$ , $x_i = \\mbox{random}$.  \\item The power-law correlation structure grows up to large scale through hierarchical clustering~\\cite{hktk-2}.  \\item The system is virialized. Power-law correlation structure persists.   \\item Microscopic relaxation takes place and power-law correlation structure disappears. Energy distribution remains invariant.   \\end{enumerate}  Related models with a uniform background or friction~\\cite{aurell-2001,miller-rouet-2001} are investigated from cosmological interest. In those models similar fractal structure has been observed and is more persistent. Comparison with our results may be helpful % to understand mechanism by which the structure is retained in their models.   It should also be noted that the structure formation we have studied have origin in dynamics.  It is interesting that, instead of monotonously relaxing to thermal equilibrium the system, initially does not have spatial correlation, creates spatial correlation of power-law type through dynamics. It seems difficult to explain it in terms of thermodynamics or equilibrium statistical physics, since the structure appears in states which are not thermally relaxed. Recently dynamical formation of spatial structure in long-range interacting systems is actively investigated. Relation between our results and other examples of structure formation with dynamical origin~\\cite{tk-cluster,onion,dauxois-2000,barre-2001} will be interesting.    In this paper we have  focused on  typical behavior of the  system. Several problems related to details of the behavior remain unsolved. Among them there are, for example, initial condition dependence, system size dependence, and so on. These are beyond the scope of this short paper and  of important subjects of future study.  In this paper we have studied the behavior of the system until power-law structure disappears through energy exchange and energy distribution. It will be interesting to investigate whether there are other relaxation processes effectively  working~\\cite{lynden-bell-1967,rouet-feix-1999}, and the behavior of the system in the course of approach to equilibrium~\\cite{sheet-tkg-1,sheet-tgk-2,yawn-miller-prl-1997,yawn-miller-pre-1997}.   The spontaneous emergence of power-law correlation we have studied so far is a novel phenomenon with rich implication. We hope our research serves as a foundation for future study, both fundamental and applied."
    },
    "0010/astro-ph0010206_arXiv.txt": {
        "abstract": "We investigate the evolution, final fate, and nucleosynthetic yields of rotating and non-rotating very massive stars (VMS) of zero metallicity.  First we address the issue of mass loss during hydrogen burning due to vibrational instabilities.  We find that these objects are much more stable than what was found in previous studies of VMS of solar composition, and expect only negligible mass loss driven by the pulsations.  As these stars thus reach the end of their evolution with massive helium cores, they encounter the pair-creation instability. We find that for helium core masses of $\\sim64\\ldots133\\,\\Msun$ these stars are completely disrupted with explosion energies of up to $\\sim\\Ep{53}\\,\\erg$ and eject up to $\\sim60\\,\\Msun$ of \\I{56}{Ni}. Stars with more massive helium cores collapse into black holes.  We present the first calculations that follow the collapse of such a massive rotating star and predict that X-ray burst and significant gravitational wave emission could result. ",
        "introduction": "Recently, three-dimensional cosmological simulations have reached sufficient resolution on small scales to begin to address the star formation problem of the first generation of so-called Population III (Pop III) stars \\cite{OG96,abe98}.  Though the formation and nature of the first generation of metal-free stars have been investigated since over thirty years \\cite[and many others]{SS53,Yon72,HC77,PSS83,Sil83}, the extent to which they differed from present day stars in ways other than composition is still debated.  While considerable uncertainty remains, these simulations suggest that the first generation of stars may have been quite massive, $\\sim 100-1000\\,\\Msun$ \\cite{Lar99,ABN00,BCL99}, giving us motivation to examine the properties, evolution, and fate of such stars. ",
        "conclusions": "Recent results on the formation of Pop~III stars indicate that very massive stars of zero metallicity could have formed.  Our analysis of the pulsation properties indicates that stars of $\\lesssim500\\,\\Msun$ are far less vibrationally unstable than previously though and thus may reach the end of their evolution as objects with massive helium cores that encounter the pair-creation instability.  This has interesting consequences for the nucleosynthetic yield from these objects, and may even allow for the observation of their explosions despite the high red shift.  Above an initial mass of $\\sim300\\,\\Msun$ the core collapses into a black hole and may cause an X-ray burst or a strong gravity wave signal that may be accessible to experiments in the near future.  The survival of these objects as very massive stars until the end of their evolution also has interesting consequences for the re-ionization of the early universe, as these VMS have an effective surface temperature of $\\sim\\Ep5\\,\\K$ \\cite{EC71,BHW00}, which means that they very efficiently turn nuclear energy into ionizing photons.  \\textbf{Acknowledgements.\\/} This research was supported, in part, by Prime Contract No. W-7405-ENG-48 between The Regents of the University of California and the United States Department of Energy, the National Science Foundation (AST 97-31569, INT-9726315), and the Alexander von Humboldt-Stiftung (FLF-1065004)."
    },
    "0010/astro-ph0010176_arXiv.txt": {
        "abstract": "We present a comprehensive study to measure the locations of $\\gamma$-ray bursts (GRBs) relative to their host galaxies.  In total, we find the offsets of 20 long-duration GRBs from their apparent host galaxy centers utilizing ground-based images from Palomar and Keck and space-based images from the Hubble Space Telescope (HST).  We discuss in detail how a host galaxy is assigned to an individual GRB and the robustness of the assignment process. The median projected angular (physical) offset is 0.17 arcsec (1.3 kpc).  The median offset normalized by the individual host half-light radii is 0.98 suggesting a strong connection of GRB locations with the UV light of their hosts. This provides strong observational evidence for the connection of GRBs to star-formation.  We further compare the observed offset distribution with the predicted burst locations of leading stellar-mass progenitor models. In particular, we compare the observed offset distribution with an exponential disk, a model for the location of collapsars and promptly bursting binaries (e.g.~helium star--black hole binaries). The statistical comparison shows good agreement given the simplicity of the model, with the Kolmogorov-Smirnov probability that the observed offsets derive from the model distribution of $P_{\\rm KS} = 0.45$.  We also compare the observed GRB offsets with the expected offset distribution of delayed merging remnant progenitors (black hole--neutron star and neutron star--neutron star binaries). We find that delayed merging remnant progenitors, insofar as the predicted offset distributions from population synthesis studies are representative, can be ruled out at the $2 \\times 10^{-3}$ level. This is arguably the strongest observational constraint yet against delayed merging remnants as the progenitors of long-duration GRBs.  In the course of this study, we have also discovered the putative host galaxies of GRB 990510 and GRB 990308 in archival HST data. ",
        "introduction": "For some thirty years since the discovery of gamma-ray bursts \\citep[GRBs;][]{kso73}, a basic understanding of the nature of the brief intense flashes of $\\gamma$-rays remained elusive.  Throughout much of the 1970s and 1980s the prevailing view was that GRBs arise from the surface of neutron stars in and around our Galaxy \\citep[see][for a review]{lam95}, though, by the mid-1990s, the isotropic distribution of GRBs on the sky \\citepcf{fm95} served as the cornerstone of mounting evidence suggesting an extra-galactic origin \\citep[see][for a review]{pac95}.  The main impedance to progress was the difficulty of localizing bursts to an accuracy high enough to unequivocally associate an individual GRB with some other astrophysical entity.  In large measure the localization problem was due to both the transient nature of the phenomena and the fact that the incident direction of $\\gamma$-rays are difficult to pinpoint with a single detector; for example, the typical 1-$\\sigma$ uncertainty in the location of a GRB using the {\\it Burst and Transient Source Experiment} (BATSE) was $4$--$8$ degree in radius \\citep{bpk+99}. The Interplanetary Network \\citep[IPN; see][]{cbb+99} localized GRBs using burst arrival times at several spacecrafts throughout the Solar System and provided accurate localizations (3 $\\sigma$ localizations of $\\sim$few to hundreds $\\times$ arcmin$^2$) to ground-based observers; however, the localizations were reported with large time delays (days to months after the GRB).  The crucial breakthrough came in early 1997, shortly following the launch of the BeppoSAX satellite \\citep{bbp+97}. On-board instruments \\citep{fcd+97,jmb+97} were used to rapidly localize the prompt and long-lived hard X-ray emission of the GRB of 28 February 1997 (GRB 970228) to a 3 $\\sigma$ accuracy of 3 arcmin (radius) and relay the location to ground-based observers in a matter of hours.  Fading X-ray \\citep{cfh+97} and optical \\citep{vgg+97} emission, the so-called ``afterglow,'' associated with GRB 970228 were discovered. Ground-based observers noted \\citep{mkd+97,vgg+97} a faint nebulosity in the vicinity of the optical transient (OT) afterglow. Subsequent {\\it Hubble Space Telescope} (HST) imaging resolved the nebulosity \\citep{slp+97} and showed that the morphology was indicative of a distant galaxy \\citep{slp+97}.  Three years later, we now know the redshift of this faint, blue galaxy is $z=0.695$ \\citep{bdk01}.  The next prompt localization of a GRB yielded the first confirmed distance to the GRB through optical absorption spectroscopy: GRB 970508 occurred from a redshift \\hbox{$z \\ge 0.835$} \\citep{mdk+97}. The first radio afterglow was detected from GRB 970508 which, through observations of scintillation, led to the robust inference of superluminal motion of the GRB ejecta \\citep{fkn+97}.  These measurements (along with the dozen other redshifts now associated with individual GRBs) have effectively ended the distance scale debate and solidified GRBs as one of the most energetic phenomena known \\citepcf{kbb+00,fks+01}.  The cosmological nature of GRBs now frames our basic understanding of the physics of GRB phenomena\\footnotemark\\footnotetext{The emergent picture described herein is reserved to the so-called long duration GRBs, those lasting for a duration $\\age 2$ sec, since no short-duration bursts have yet been well-localized on rapid timescales ($\\ale $ few days).}.  The general energetics are well-constrained: given the observed fluences and redshifts, approximately $10^{51}$--$10^{53}$ erg in $\\gamma$-ray radiation is released in a matter of a few seconds in every GRB. The GRB variability timescale suggests that this energy is quickly deposited by a ``central engine'' in a small volume of space (radius of $\\sim 30$ km) and is essentially optically thick to $\\gamma$-ray radiation at early times. This opaque fireball of energy then expands adiabatically and relativistically until the $\\gamma$-ray radiation can escape; the emitting surface of the GRB is likely to be 10$^{15}$--$10^{17}$ cm from the explosion site and probably arises from the interaction of internal shocks initiated by the central engine \\citepeg{frrw99}. Only a small amount of baryonic matter ($\\sim 10^{-5} M_\\odot$) can be entrained with the fireball since too much baryonic matter, a condition referred to as the ``baryonic loading'' problem, would essentially stall the relativistic expansion of the fireball. The transient afterglow phenomena is thought to be due to synchrotron radiation arising from the interaction of the relativistic ejecta and the ambient medium surrounding the burst site \\citep[see][for reviews]{cpkw00,kbb+00,dfk+01a}. The relativistic nature of the expanding shock (which also gives rise to the GRB) is required to avoid the so-called compactness problem \\citepcf{pir99} and, as mentioned above, was observationally confirmed with radio scintillation measurements of the afterglow of GRB 970508 \\citep{fkn+97}.  While the GRB emission and the afterglow phenomenon are now reasonably well-understood, one large outstanding question remains: what makes a $\\gamma$-ray burst?  Specifically what are the astrophysical objects, the ``progenitors'', which produce $\\gamma$-ray bursts?  To-date several theoretical considerations appear to implicate the progenitors as stellar-mass systems involving a compact source, probably a black hole (BH).  First, the implied (isotropic) energy release in $\\gamma$-rays are typically 10$^{-3}$--$10^{-1}$ times the rest-mass energy of the Sun.  The estimated efficiency of conversion of the initial input energy (either Poynting flux or baryonic matter) to $\\gamma$-rays ranges from $\\sim$1\\% \\citepeg{kum99} to as much as $\\sim$60\\% \\citepeg{ks01}; therefore, the best-guess estimate of the total energy release is roughly comparable to the rest-mass energy of one solar mass. Second, the variability timescale (few ms) observed implies the energy deposition takes place in a small region of space (radius of $c \\times 1$ ms $\\approx 30$ km). Third, the inferred rate of GRB occurrence, about 4 per day in the Universe above current detection thresholds, and the lack of burst repetition \\citepeg{hmp+98} suggest that GRB events are rare \\citep[$\\sim 10^{-7}$ yr$^{-1}$ Galaxy$^{-1}$;][]{feh93,wbbn98} and catastrophically destroy the individual progenitors.  The progenitor models which most naturally explain these observables in the GRB phenomena fall in to two broad classes---the coalescence of binary compact stellar remnants and the explosion of a massive star (``collapsar'').  An active galactic nucleus (AGN) origin is another possibility, however the variability timescale still requires the energy source to be stellar-mass objects \\citep{car92,cw99}.  We briefly summarize the popular progenitor models and refer the reader to \\citet{fwh99} for a more in-depth review. In both the collapsar and the merging remnant class of progenitors a spinning BH is formed.  The debris, either from the stellar core of the collapsar or a tidally disrupted neutron star, forms a temporary accretion disk (or ``torus'') which then falls into the BH releasing a fraction of gravitational potential of the matter. In this general picture \\citep[see][for a review]{ree99}, the lifetime of the accretion disk accounts for the duration of the GRB and the light-crossing time of the BH accounts for the variability timescale.  The GRB is powered by the energy extracted either from the spin energy of the hole or the from the gravitational energy of the in-falling matter.  The coalescing compact binary class \\citep{pac86,goo86,eic+89} was favored before the first redshift determination because the existence of coalescence events of a double neutron star binaries (NS--NS) was assured; at least a few NS--NS systems in our Galaxy (e.g., PSR 1913+16, PSR 1534+12) will merge in a Hubble time thanks to the gravitational radiation of the binary orbital angular momentum \\citepcf{tay94}.  Further, the best estimate of the rate of NS--NS coalescence in the Universe \\citepeg{phi91,npp92} was comparable to an estimate of the GRB rate \\citep{feh93,wbbn98}.  Recently, stellar evolution models have suggested that black hole--neutron star binaries (BH--NS) may be formed at rates comparable to or even higher than NS--NS binaries \\citepeg{bb98}, though no such systems have been observed to-date.  There are other merging remnant binaries which may form GRBs, notably merging black hole--white dwarf (BH--WD) binaries \\citep{fw98} and black hole--helium star binaries (BH--He) \\citepcf{fwh99}.  The collapsar class is comprised of a rotating massive star, either isolated or in a binary system, whose iron core subsequently collapses directly to form a black hole \\citep{woo93}.  To avoid baryon loading the progenitor star should have lost most, if not all, of its extended gas envelope of hydrogen by the time of collapse.  The progenitors of collapsars---likely Wolf-Rayet stars---are then closely related to the progenitors of hydrogen-deficient supernova, namely type Ib/Ic supernovae \\citepcf{mw99}. Perhaps one distinguishing difference is that high angular momentum is necessary in collapsars. High angular momentum centrifugally supports a transient torus around the BH, creating a natural timescale for mass-energy injection.  The efficiency of energy conversion is also helped around a spinning BH. Further, angular momentum creates a natural rotation axis along which large density gradients allow for the expanding blastwave to reach relativistic speeds.  How can this large variety of viable GRB progenitors be distinguished? Direct associations with other known astrophysical entities is possible.  For massive stars, the energy release from the collapse of the core of the star, just as in supernovae, is sufficient to explode the star itself.  This may result in a supernova-like explosion at essentially the same time as a GRB.  The first apparent evidence of such a supernova associated with a cosmological GRB came with the discovery of a delayed bright red bump in the afterglow light curve of GRB 980326 \\citep{bkd+99}.  The authors interpreted the phenomena as due to the light-curve peak of a supernova at redshift $z \\sim 1$. Later, \\citet{rei99} and \\citet{gtv+00} found similar such red bump in the afterglow of GRB 970228.  Merging remnant progenitors models (e.g., BH--NS, NS--NS systems) have difficulty producing these features in a light curve on such long timescales and so the supernova interpretation, if true, would be one of the strongest direct clues that GRBs come from massive star explosions. However, the supernova story is by no means complete.  For instance, in only one other GRB (000911) has marginal ($\\sim$2 $\\sigma$) evidence of a SN signature been found \\citep{laz+01}; further, many GRBs do not appear to show any evidence of SNe signatures \\citepeg{hhc+00}. Even the ``supernova'' observations themselves find plausible alternative explanations (such as dust echoes) that do not strictly require a massive star explosion \\citep{eb00,rei01,wd00}. We note, however, that all other plausible explanations of the observed late-time bumps require high-density environments found most readily in star forming regions.  \\citet{cl99b} emphasize that if a GRB comes from a massive star, then the explosion does not take place in a constant density medium, but in a medium enriched by constant mass loss from the stellar winds.  One would expect to see signatures of this wind-stratified medium in the afterglow \\citep[e.g., bright sub-millimeter emission at early times, increasing ``cooling frequency'' with time; see][]{pk00b,kbb+00}. However, afterglow observations have been inconclusive \\citep{kbb+00} with no unambiguous inference of GRB in such a medium.  Recent work has begun to focus on the immediate environments of GRBs as a means toward divining the nature of the progenitors.  This has come primarily from detections of line features in now five GRB afterglows \\citep[e.g., 970508 and 970828][]{pcf+99,yno+99}. The most recent and convincing detection so far comes from observations of the afterglow of GRB 991216 \\citep{pir01}.  Individually, the observational significance of the line detections are marginal but on the whole there appears to be a good case for line emission features in the afterglow of some GRBs. If so, the inescapable conclusion is that there must exist dense matter in the vicinity of the explosion \\citepeg{wmkr00,vpps99,lcg99a}, a seeming discordance with the expectations of NS--NS merger models.  We emphasize that even the connection of GRBs to stellar-mass progenitors has yet to be established.  The most compelling arguments we have outlined (e.g.~temporal variability) rely on theoretical interpretations of the GRB phenomena.  Further, direct observational results (SNe signatures and transient Fe-line emission) are not yet conclusive.   In this paper we examine the observed locations of GRBs with respect to galaxies.  We find an unambiguous correlation of GRB locations with the UV light of their hosts, providing strong indirect evidence for the connection of GRBs to stellar-mass progenitors. Beyond this finding, we aim to use the location of GRBs to distinguish between stellar-mass progenitor models. In \\S \\ref{sec:location} we review the expectations of GRB locations from each progenitor model.  Then in \\S\\S \\ref{sec:offdata}--\\ref{sec:astlevels} we discuss the instruments, techniques, and expected uncertainties involved in constructing a sample of GRB locations about their host galaxies.  In \\S \\ref{sec:indivoff} we comment on the data reductions specific to each GRB in our sample.  The observed distribution is shown and discussed in \\S \\ref{sec:offdist} and then statistically compared with the expected offset distribution of leading progenitor models (\\S \\ref{sec:compare}). Last, in \\S \\ref{sec:offsum} we summarize and discuss our findings. ",
        "conclusions": "\\label{sec:offsum}  We have determined the observed offset distribution of GRBs by astrometrically comparing localizations of GRB afterglow with optical images of the field surrounding each GRB. In all cases, the GRB location appears ``obviously'' associated with a galaxy---either because the position is superimposed atop a galaxy or very near ($\\ale 1.2\\arcsec$) a galaxy in an otherwise sparse field.  In fact, irrespective of the validity of individual assignments of hosts, the offset distribution may be considered a distribution of GRB positions from the nearest respective galaxy at least as bright as $R \\approx 28$ (note that in most cases the host galaxies are much brighter, typically $R = 24$--26 mag). We find that at most a few of the 20 GRBs could be unrelated physically to their assigned host and about a 50\\% chance that all GRBs are correctly assigned to their hosts (see \\S \\ref{sec:angoffs}).  We then compare the distribution of GRB locations about their respective hosts with the {\\it predicted} radial offset distribution of merging binary remnants. This comparison is complicated by an unknown projection factor for each burst: if a GRB occurs near an edge-on disk galaxy there exists no model-independent manner to determine the true 3-D radial offset of the GRB from the center of the host.  Indeed, in a few cases (e.g., GRB 980519, GRB 991216) even the ``center'' of the host is not well defined and we must estimate a center visually.  In all other cases, we find the centers using a luminosity-weighted centroid surrounding the central peak of the putative host.  To compare the GRB offsets with those predicted by the NS--NS and NS--BH binary models, we make a general assumption about the projection factor and, to facilitate a comparison in physical units (that is, offsets in kiloparsec rather than arcseconds), we assign an angular diameter distance to the 5 hosts without a confirmed distance (\\S \\ref{sec:physproj}).  We have shown that the conversion of an angular offset to physical projection is relatively insensitive to the actual redshift of the host. We estimate that the probability that the observed GRB offset distribution is the same as the predicted distribution of NS--NS and BH--NS binaries is $P \\ale 2 \\times 10^{-3}$.  Insofar as the observed distribution is representative (see below) and the predicted distribution is accurate, our analysis renders BH--NS and NS--NS progenitor scenarios unlikely for long-duration GRBs.  Having cast doubt on the merging remnant hypothesis, we test whether the offset distribution is consistent with the collapsar (or BH--He) class.  Since massive stars (and promptly merging binaries) explode where they are born, we have compared the observed GRB offset distribution with a very simplistic model of massive star formation in late-type galaxies: an exponential disk.  After normalizing each GRB offset by their host half-light radius we compare the distribution with a KS test and find good agreement: $P_{\\rm KS} = 0.454$. We have shown that these KS results, based on the assumption of $\\delta$-function offsets, are robust even after including the uncertainties in the offset measurements.  Thus far we have neglected discussion of the observational biases that have gone into the localizations of these 20 GRBs.  The usual problems plaguing supernova detection, such as the brightness of the central region of the host and dust obscuration, are not of issue for detection of the {\\it prompt} high-energy emission ({\\it i.e.}, X-rays and $\\gamma$-rays) of GRBs since the high-energy photons penetrate dust.  If the intrinsic luminosity of GRBs is only a function of the inner-workings of the central engine (that is, GRBs arise from internal shocks and not external shocks) then the luminosity of a GRB is independent of ambient number density.  Therefore, prompt X-ray localizations from BeppoSAX and $\\gamma$-ray locations from the IPN should not be a function of the global properties of GRB environment; only intrinsic GRB properties such as duration and hardness will affect the prompt detection probability of GRBs.  The luminosity of the afterglows is, however, surmised to be a function of the ambient number density. Specifically, the afterglow luminosity will scale as $\\sqrt{n}$ where $n$ is the number density of hydrogen atoms in the 1--10 pc region surrounding the GRB explosion site \\citep[cf.][]{mrw98b}.  While $n \\approx$ 0.1--10 cm$^{-3}$ in the interstellar medium, the ambient number density is probably $n \\approx 10^{-4}$--10$^{-6}$ in the intergalactic medium.  Thus GRB afterglows in the IGM may appear $\\sim 10^{-3}$ times fainter than GRB afterglows in the ISM (and even more faint compared to GRBs that occur in star-forming regions where the number densities are higher than in the ISM). If only a small fraction of GRBs localized promptly in X-rays and studied well at optical and radio wavelengths were found as afterglow, the ambient density bias may be cause for concern. However, this is not the case. As of June 2001, 29 of 34 bursts localized by prompt emission were later found as X-ray, optical, and/or radio afterglow \\citepcf{fkw+00}; that is, almost all GRBs have detectable X-ray afterglow.  Therefore, {\\it no more than about $10\\%$ of GRBs localized by BeppoSAX could have occurred in significantly lower density environments} such as in the IGM; thus, we do not believe that our claim against the delayed merging binaries is affected by this bias.  What about the non-detection of GRB afterglow at optical/radio wavelengths?  Roughly half of GRBs promptly localized in the gamma-ray or X-ray bands are not detected as optical or radio afterglow \\citep{fkw+99,lcg01}.  While many of these ``dark'' GRBs must be due to observing conditions (lunar phase, weather, time since burst, etc.) at least some fraction may be due to intrinsic extinction local to the GRB.  If so, then these GRBs are likely to be centrally biased since the optical column densities are strongest in star-forming regions and giant molecular clouds.  Therefore, {\\it any optically obscured GRBs which do not make it in to our observed offset sample will be preferentially located in the disk}. We do not therefore believe the ambient density bias plays any significant role in causing GRBs to be localized preferentially closer to galaxies; in fact, the opposite may be true.  The good agreement between our simplistic model for the location of massive stars and the observed distribution is one of the strongest arguments yet for a collapsar (or promptly bursting binaries) origin of long-duration GRBs. However, the concordance of the predicted and observed distributions are necessary to prove the connection, although not sufficient.  We may now begin to relate the offsets to the individual host and GRB properties. For instance, of the GRBs which lie in close proximity to their host centers (GRB 970508, GRB 980703, and GRB 000418), there is a striking similarity between their hosts---all appear compact and blue with high-central surface brightness suggesting that these hosts are nuclear starburst galaxies (none show spectroscopic evidence for the presence of an AGN).  In fact, the closeness of some GRBs to their host centers signifies that our simplistic model for star-formation may require modification. This is not unexpected since, in the Galaxy, star formation as a function of Galactocentric radius does not follow a pure exponential disk, but is vigorous near the center and is strongly peaked around $R \\sim 5$ kpc \\citepcf{ken89}.  As more accurate offsets are amassed, these subtle distinctions in the GRBs offset distribution may be addressed."
    },
    "0010/astro-ph0010495_arXiv.txt": {
        "abstract": "Using a periodic box calculation it is shown that, owing to helicity conservation, a large scale field can only develop on a resistive timescale. This behaviour can be reproduced by a mean-field dynamo with $\\alpha$ and $\\eta_{\\rm t}$ quenchings that are equally strong and `catastrophic'. ",
        "introduction": "After the original papers by Vainshtein \\& Cattaneo (1992) and Cattaneo \\& Vainshtein (1991) about `catastrophic' $\\alpha$ and $\\eta_{\\rm t}$ quenching, respectively, there has been a continuous debate about the existence and relevance of $\\alpha$-effect dynamos in astrophysics, i.e.\\ in the high $R_{\\rm m}$ regime. Their results were supported by calculations that did not however have dynamo action (see also Cattaneo \\& Hughes 1996). In the present paper we show, using a periodic box calculation that, owing to helicity conservation, a large scale field can only develop on a resistive timescale, and that this behaviour can be reproduced by a mean-field dynamo with $\\alpha$ and $\\eta_{\\rm t}$ quenchings that are equally strong and `catastrophic' in the sense that quenching sets in once the large scale field reaches the fraction $R_{\\rm m}^{-1/2}$ of the equipartition value. ",
        "conclusions": ""
    },
    "0010/astro-ph0010389_arXiv.txt": {
        "abstract": "Discrepancies have emerged between the predictions of standard cold dark matter (CDM) theory and observations of clustering on sub-galactic scales. Warm dark matter (WDM) is a simple modification of CDM in which the dark matter particles have initial velocities due either to their having decoupled as thermal relics, or having been formed via non-equilibrium decay. We investigate the nonlinear gravitational clustering of WDM with a high resolution N-body code, and identify a number of distinctive observational signatures. Relative to CDM, halo concentrations and core densities are lowered, core radii are increased, and large halos emerge with far fewer low mass satellites. The number of small halos is suppressed, and those present are formed by `top down' fragmentation of caustics, as part of a `cosmic web' connecting massive halos. Few small halos form outside this web. If we identify small halos with dwarf galaxies, their number, spatial distribution, and formation epoch appear in better agreement with the observations for WDM than they are for CDM. ",
        "introduction": "Standard cold dark matter (CDM) theory appears to be in conflict with observations of clustering on sub-galactic scales. First, the theory predicts  a very large number of low mass halos. For example, of order 500 satellites of mass greater than $10^8 M_\\odot$ are expected within a halo as massive as that of the Milky Way whereas only 11 candidate dwarf galaxies with comparable mass are observed \\citep{Moore}. In the whole Local group, out to $\\sim 1.5$ Mpc there are only $36$ known dwarf galaxies \\citep{mateo}, where several thousand would be expected in a straightforward interpretation of the theory. The second problem with CDM is that simulations produce halos with more concentrated cores than those inferred from the rotation curves of real galaxies \\citep{Moorecore,Ghigna99}. There are a number of related but more complex problems. For example, the predicted satellites would lead to a thickening of the disks of galaxies which is not observed. And the dense cores,  which form very early, would cause baryonic cooling to be too efficient, leading to disks with an order of magnitude too little angular momentum \\citep{Dolg,Navarro}.  These problems may well have complex astrophysical solutions. Feedback processes such as heating and supernovae winds might inhibit star formation in low mass halos \\citep{bullock,silk}, although it is not clear why such processes would suppress the number of satellites without destroying them all. No convincing mechanism has been devised to explain the fact that most observed dwarf systems have characteristic densities which are considerably lower than those of dark matter halos of the same mass predicted from $\\Lambda$CDM theory, and it certainly seems implausible that feedback processes would preferentially inhibit star formation in the denser halos. Further, dwarf spheroidals are believed to be completely dominated by dark matter, but their stellar profiles show no sign of a central density cusp \\citep{irwin}.  The observational situation regarding galaxy cores has been rapidly evolving. It was recently claimed that beam smearing effects were significant, bringing observations of the inner rotation curves of low surface brightness galaxies closer to the predictions of CDM \\citep{vdbswat00,swaters} (although there remained a discrepancy with the $\\rho \\propto r^{-1.5}$ central profile indicated  by high resolution CDM simulations \\citep{Moorecore,swaters,klypin}). But very recently de Blok {\\it et al.} \\citep{deblok} reanalised the same data and added considerably more new data, only to reach the opposite conclusion that beam smearing effects were not important and the data are seriously discrepant even with the milder $\\rho \\propto r^{-1}$ behavior of the NFW profile \\citep{NFW97}.  There is another potential problem with CDM, noted some time ago by \\citet{pjep00,peeb01}. The puzzle is that faint dwarf galaxies as measured in our vicinity (for a picture of the local volume see \\citet{peeb} p. 26 and \\citet{SRE97}) do not fill the voids in the bright galaxy distribution. Rather they tend to follow an even sharper distribution than the bright galaxies, tracing out surprisingly thin sheets around the voids. This effect has not as far as we know been systematically quantified, but is interesting in connection with the satellite problem of CDM. That is, observationally, dwarf galaxies seem both more strongly correlated with bright galaxies than in CDM, whilst at the same time being suppressed within large galaxy halos. In CDM theory both features are hard to understand, since the smaller halos formed earlier and should not be strongly correlated with later forming, larger scale structure. They should fill both the voids and massive structures alike, as we shall verify in our $\\Lambda$CDM simulations presented below.  Even if supernova winds or heating processes inhibit star formation in dwarf halos, and thereby suppress dwarf galaxies, it is hard to see why these processes would act preferentially in the voids. Another clue of potential significance is that observations \\citep{metc} point to the formation of dwarf galaxies at $3<Z<4$, later than bright galaxies.  This is not what one expects in a hierarchical theory like CDM where small objects should form first.  If these problems with CDM are real, they represent a remarkable opportunity. The observed pattern of gravitational clustering may be revealing the physical properties of the dark matter. If so, this will be an invaluable clue to physics beyond the standard model. Deciphering this clue represents an exciting challenge in which both more refined observations and numerical simulations will be needed. Even if the conclusion of this work is that CDM theory survives, it will have been strengthened in the process. Alternative theories are valuable foils against which the successes of the better theory may be judged.  The small scale problems mentioned above do not negate the remarkable successes of the $\\Omega_{M}\\sim 0.3$, $\\Omega_{\\Lambda} \\sim 0.7$ $\\Lambda$CDM cosmology on larger scales \\citep{BOPS99}. These include the abundances and observed evolutionary properties of Lyman $\\alpha$  clouds, large galaxies, clusters and of course the normalization and shape of the power spectrum of cosmic microwave anisotropies. Therefore it seems sensible to seek a modest modification of the scenario whose sole effect is to damp small scale structure.  Free streaming due to thermal motion of particles, or Landau damping, is the simplest known mechanism for smearing out small scale structure. This process was central to the hot dark matter (massive neutrino) scenario for structure formation, which failed because the particle speeds were too high and erased perturbations up to tens of Megaparsec scales. Structure formation was `top down', with galaxies forming only through fragmentation of pancakes, and at redshifts too low to be compatible with observation.  Warm dark matter (WDM) is just hot dark matter cooled down. We shall review below some of the many ideas in the literature as to how WDM might be produced. A particularly interesting possibility is that if the reheating temperature following inflation is low, then standard $\\mu$ or $\\tau$ neutrinos can be warm dark matter. Historically, WDM  was considered (and quickly rejected) as a means of reconciling computed and observed cluster abundances while maintaining a critical density for the dark matter \\citep{widrow}. That problem is resolved in the $\\Lambda$CDM model by simply lowering the dark matter density. But the new problems faced by CDM motivate reconsideration of WDM (reincarnated now as $\\Lambda$WDM) with a view to damping structure on sub-Megaparsec  scales \\citep{Dolg,hogdal}, requiring much lower particle velocities.  It is conventional to discuss WDM in terms of a canonical candidate, a light fermion with two spin degrees of freedom like the neutrino, assumed to decouple adiabatically and whilst still relativistic. The number of relativistic degrees of freedom at decoupling determines the effective temperature of the warm particles today,  and the dark matter density determines the WDM particle mass $m_X$. We shall discuss WDM in terms of this mass, but it should be remembered that what really matters is the particle streaming speed, given for the canonical particle in equation (\\ref{eq:veq}) of the Appendix.  We discuss linear perturbation theory relevant to warm particles in the Appendix, where we derive the following formula for the smoothing scale, defined as the comoving half-wavelength of the mode for which the linear perturbation amplitude is suppressed by two: \\begin{equation} R_S\\approx 0.31\\, \\left({\\Omega_X\\over.3}\\right)^{.15} \\left({h\\over .65}\\right)^{1.3} \\left({{\\rm keV}\\over m_X}\\right)^{1.15} h^{-1} {\\rm Mpc}. \\label{eq:rsup} \\end{equation} The scaling given here is different from that given in the previous literature on WDM (which mostly refers to \\citet{bs} and \\citet{bbks}). The scaling usually given does not properly take into account streaming prior to matter-radiation equality.  The outline of this paper is as follows. In Section 2 we review particle physics mechanisms through which WDM may be formed. In Section 3, we discuss the lower bound imposed on $m_X$ by the requirement of early structure formation and in particular of reionizing the universe (the Gunn-Peterson constraint). In Section 4 we discuss the phase space density (Tremaine-Gunn) constraint, and show that the initial thermal velocities of the particles are only relevant to the inner cores of halos, scales of hundreds of parsecs. The following sections are devoted to numerical results, revealing several potential observational signatures of warm dark matter.  We find that replacing cold with warm dark matter has the following effects:  \\begin{enumerate} \\item    Smoothing of massive halo cores, lowering core densities and increasing core radii. \\item    Lowering greatly the characteristic density of low mass halos. \\item    Reduction of the overall number of low mass halos. \\item    Suppression of the number of low mass satellite halos in high mass halos. \\item    Formation of low mass halos almost solely within caustic pancakes or ribbons connecting larger halos in a `cosmic web'. Voids in this web are almost empty of small halos, in contrast to the situation in CDM theory. \\item    Late formation ($Z<4$) of low mass halos, in a `top down' process. \\item    Suppression of halo formation at high redshift ($Z>5$), and increased evolution of halos at lower redshifts relative to CDM. \\end{enumerate}  The first four findings indicate that WDM  holds some promise as a solution to the satellite and core density problems of CDM. The fifth is interesting as it may solve the  problem raised by Peebles. The sixth and seventh items point to observational tests at high redshift, where there is the possibility of actually seeing the pancake formation as it first occurs. Recent HST observations probing galaxies at high redshift seem to  indicate the formation of dwarf galaxies after bright galaxies \\citep{metc}, which is an argument in favor of WDM. It is perhaps worth emphasising that {\\it WDM succeeds both in suppressing the number of satellites in large halos, whilst producing a distribution of dwarf halos in a cosmic web connecting higher mass galaxies.} All of these effects need to be better quantified in both observational data and simulations, so that precise statistical tests are possible.  It is a challenging problem at the limit of current numerical codes to simultaneously  represent the modes responsible for structure formation (tens of Megaparsecs at least) whilst resolving objects as small as dwarf galaxies $< 10^8 M_{\\odot}$. We chose the compromise of studying large boxes of $20 h^{-1}$ Mpc with $256^3$ particles in which we compare $\\Lambda$CDM with $\\Lambda$WDM, with WDM particle masses of $m_X=$ 175eV and 350 eV corresponding to smoothing scales $R_S$ of 2.3 and 1.0 $h^{-1}$ Mpc respectively. These values for $m_X$ are almost certainly excluded by the Gunn-Peterson limit which we detail below, or by the limit from Lyman $\\alpha$  clouds, pointing to $m_X>$ 750 eV \\citep{MaSper}. Nevertheless they are helpful for illustrating the qualitative differences between WDM and CDM clustering on large scales. We also performed a simulation with a more realistic value of $m_X=$ 1.5 keV, corresponding to a smoothing scale of 0.19 $h^{-1}$ Mpc, in smaller $3 h^{-1}$ Mpc boxes with $128^3$ particles. Finite size effects are important in this box, since long wavelength modes in the power spectrum are not properly represented. Nevertheless the simulation shows that the effects found at low $m_X$ persist at higher values. A second limitation is that with the current code, computation time becomes prohibitive for the $256^3$ boxes at redshifts below  unity. Therefore our largest simulations end at $Z=1$. The signatures we identify deserve more detailed study, and larger simulations  will be carried out in future work.\\footnote{ Movies of the simulations were found to be very instructive in showing how structure formation in WDM proceeds. We have made these, higher resolution versions of the figures and some other results  not included here available on the web, at \\url{http://\\allowbreak astro.\\allowbreak Princeton.\\allowbreak EDU/\\allowbreak $\\sim$bode/\\allowbreak WDM/}. }  Simulations have also been recently carried out by \\citet{ACVDF01}, \\citet{CoAvVa00}, and \\citet{colin}. They found that WDM results in a reduction of the number of satellites in Milky-Way sized galactic halos, and in satellite halos with less centrally concentrated density profiles. The work presented here is in agreement with these findings. \\citet{Moorecore} simulated a $\\sim 10^{12} M_\\odot$ halo and found little difference in the density profile between warm and cold dark matter. \\citet{WhCr00} used simulations to examine the nonlinear power spectrum in WDM models. ",
        "conclusions": "We have identified several key observational features of WDM clustering which serve to distinguish it from cold dark matter. The principal difference is due to the `top down' formation of structure below a characteristic mass scale, which we can roughly quantify as \\begin{equation} M_{S} =  1.0\\times 10^{10} \\left({\\Omega_X\\over.3}\\right)^{1.45} \\left({h\\over .65}\\right)^{3.9} \\left({{\\rm keV}\\over m_X}\\right)^{3.45} h^{-1} M_{\\odot}. \\label{eq:masssc} \\end{equation} Below this mass scale, objects are formed primarily by the fragmentation of pancakes and ribbons. They are rarer and considerably less dense than halos of the same mass in CDM. And their spatial distribution is very different - they are concentrated in sheets and ribbons running between the massive halos, an effect which has been noted for some time for dwarf galaxies in the local universe \\citep{pjep00,peeb01}. Finally, the dwarf galaxies form later than bright galaxies, contrary to the hierarchical picture of clustering in CDM theory. Again WDM seems to be in better agreement with the observations on this point \\citep{metc}.   This work points to the need for the following observational tests to verify or refute the predicted signatures, namely:  $\\bullet$ Observations revealing the ionization history and Lyman $\\alpha$ cloud distribution at redshifts higher than 6.  $\\bullet$ Observations revealing the history of galaxy formation and checking whether low mass galaxies formed later than higher mass galaxies, as in WDM, or earlier as in CDM.  $\\bullet$ Determination of the abundance, masses, density profiles   and clustering pattern of dwarf galaxies relative to large galaxies. A recent study \\citep{zab} indicates that the dwarf to giant ratio increases with increasing density, consistent with our findings here. Likewise the apparent absence of dwarf systems in the voids noted by \\citet{pjep00,peeb01} needs to be be carefully quantified and compared with CDM and WDM simulations.  $\\bullet$ High accuracy observations measuring the inner parts of galaxy rotation curves and comparing them with detailed N body simulations of $\\Lambda$WDM and $\\Lambda$CDM.  $\\bullet$ Observations of the abundances of satellites in galaxies other than the Milky way. Especially important is a determination of the characteristic densities of dwarf dark matter dominated systems, as compared to dwarf halos in simulations.  It is clear that a great deal also needs to be done in simulating WDM, to better quantify the qualitative features we have noted here. This will require higher resolution codes and considerable amounts of computation time."
    },
    "0010/astro-ph0010340_arXiv.txt": {
        "abstract": "We have used archival ultraviolet (UV) imaging of 50 nearby star-forming galaxies obtained with the Ultraviolet Imaging Telescope (UIT) to derive integrated near-UV and far-UV magnitudes, and have combined these data with \\hans, far-infrared, and thermal radio continuum measurements to explore the consistency of UV and \\ha star formation rates (SFRs). In agreement with previous studies, we find that the UV and \\ha SFRs are qualitatively consistent, even before corrections for extinction are applied.  The uncorrected UV SFRs are systematically lower by a factor of 1.5 (with a factor of two scatter) among luminous galaxies with SFR $\\ga 1$ \\msun$\\,{\\rm yr}^{-1}$, indicating a higher effective attenuation of the far-UV radiation. Among less luminous galaxies there is no significant offset between the \\ha and far-UV SFR scales. This behavior is consistent with those of higher-redshift samples observed by Sullivan et al., Glazebrook et al., and Yan et al. for comparable ranges of galaxy luminosities and absolute SFRs.  Far-infrared and thermal radio continuum data available for a subset of our sample allow us to estimate the attenuation in the UV and at \\ha independently.  The UV and \\ha attenuations appear to be correlated, and confirm systematically higher attenuations in the UV.  Although the galaxies in our sample show modest levels of attenuation (with median values of 0.9 mag at \\ha and 1.4 mag at 1550 \\AA), the range across the sample is large, $\\sim$4 mag for \\ha and $\\ga$5 mag in the far-UV (1550 \\AA). This indicates that the application of a single characteristic extinction correction to \\ha or UV SFRs is only realistic for large, well-defined and well-studied galaxy samples, and that extinction bias may be important for UV or emission-line selected samples of star-forming galaxies. ",
        "introduction": "Until recently most of our information about the systematic behavior of star formation rates (SFRs) in normal nearby galaxies has been based on measurements of the \\ha emission line \\citep[e.g.][]{kk83,gallego95}.  In contrast, SFRs for $z > 1$ galaxies are primarily based on observations of the redshifted ultraviolet (UV) continuum \\citep[e.g.][]{madau98,steidel99}.  Comparable UV observations for nearby galaxies are being accumulated \\citep[e.g.][]{donas87,tresse98,buat99,sullivan00}, and these have made it possible to derive the cosmic evolution of the volume-averaged SFR in a self-consistent manner.  However, questions remain about the systematic accuracy of the UV-derived SFRs, both in terms of the absolute SFR scale and possible redshift-dependent biases in that scale. Similar questions also apply to the \\hans-based SFRs, and with the extension of this technique to high-redshift galaxies by several groups \\citep[e.g.][]{glaze99,yan99,moorcroft00} it is important to understand the limitations of this technique also.  Although \\ha and UV measurements are now available for samples of hundreds of galaxies in each case, there have been only a few direct comparisons of the respective SFRs for galaxies in common.  Comparisons of three samples of high-redshift galaxies ($z = 0.7 - 2.2$) by \\citet{glaze99}, \\citet{yan99}, and \\citet{moorcroft00} show that the uncorrected \\ha fluxes yield SFRs that are $\\sim$3 times higher than derived from the 2800 \\AA\\ UV flux.  The simplest explanation for this difference would be a systematically higher extinction in the UV.  However, similar comparisons using nearby samples present a less clear picture.  \\citet{buat87} and \\citet{buat92} compared balloon-based UV and ground-based \\ha measurements of small samples of galaxies and found a general consistency between the extinction-corrected SFR scales.  Furthermore, \\citet{buat96} estimated UV extinctions for a large sample of star-forming spirals and estimated typical extinctions that are comparable to or lower than those at \\hans.  The most comprehensive comparison to date of UV and \\hans-based SFRs was carried out by \\citet{sullivan00}, who analyzed balloon-based UV (2000 \\AA) and \\ha fluxes for 273 galaxies in the redshift range $0 < z < 0.5$.  The galaxies in this sample span nearly 4 orders of magnitude in SFR, ranging from dwarfs to massive spirals; many galaxies are undergoing intense star formation bursts.  Although they observed a strong correlation between UV and \\hans-derived SFRs, they also found that the UV/\\ha flux ratio decreases significantly with increasing SFR.  In low-SFR systems the UV fluxes consistently yield {\\it higher} SFRs than \\hans, while the reverse is true for galaxies with the highest SFRs.  They attribute most of the differences to rapid variability in the SFR over the lifetimes of the starbursts.  These results underscore the need for a complimentary comparison of UV and \\ha SFRs for a sample of well-resolved nearby galaxies with a fuller range of star formation histories. In this paper we primarily compare far-UV (\\fuvns) and \\ha SFRs for a sample of 50 nearby star-forming galaxies imaged using the Ultraviolet Imaging Telescope \\citep[UIT;][]{stecher97}. This sample is much smaller than that studied by \\citet{sullivan00} but it has several advantages which complement the strengths of their study.  The UIT sample is far from complete: rather, it was selected to span a full range of morphological types, luminosities, SFRs, and star formation properties, with the bulk of the sample composed of high-luminosity spiral galaxies. This virtually eliminates the problem of rapidly variable SFRs that are important when analyzing starburst galaxy samples, and allows us to compare the zeropoints of the UV and \\hans-based SFR scales on a self-consistent basis.  Another advantage of this sample is the availability of truly integrated UV and \\ha fluxes, which eliminates any possible effects of aperture undersampling at \\hans.  This paper concentrates on a comparison of the integrated fluxes and SFRs of the UIT/\\ha sample.  Our main objectives are to: 1) compare the SFRs derived from \\fuv and \\ha luminosities, with and without corrections for dust attenuation;  2) compare the behavior of our UV/\\ha SFRs with those derived for other samples at low and high redshift; and 3) obtain insight into the amount of attenuation suffered by \\fuv and \\ha in this sample of star-forming galaxies. Future papers will compare the UV and emission-line properties of individual star-forming regions in these galaxies.  In this paper we follow the convention of \\citet{fluxratio} in using the term `attenuation' to refer to the net loss of radiation from a galaxy at a given wavelength.  For a point source such as a star this would be the equivalent to the extinction (itself a combination of dust absorption and scattering), but for a galaxy the radiation reaching the observer is the result of a complicated radiative transfer across multiple lines of sight with enormous variations in local optical depth.  In the specific context of this paper it will generally refer to the correction factor that must be applied to the observed integrated UV or \\ha flux to derive the actual SFR due to the effects of interstellar dust (and exclusive of other physical effects such as escape of ionizing photons from a galaxy, etc.).  The remainder of the paper is organized as follows. In \\S \\ref{sec:data}, we present the data: near-UV (NUV) and \\fuv magnitudes, \\ha line luminosities, far-infrared (FIR) luminosities and thermal radio continuum fluxes at 1.4 GHz.  We also present the SFR calibrations for \\ha and \\fuv luminosities. In \\S \\ref{sec:sfr}, we compare the raw \\ha and \\fuv SFRs, and interpret the discrepancies in terms of variations in star formation histories and differences in dust attenuation. In \\S \\ref{sec:atten}, we examine the role of attenuation in more detail. Finally, in \\S \\ref{sec:conc}, we present our conclusions. ",
        "conclusions": "\\label{sec:conc}  We have performed photometry on UIT \\nuv and \\fuv images of a sample of 50 star-forming galaxies.  Comparison with literature \\fuv magnitudes suggests that our magnitudes are accurate to better than 0.4 mag. Combining these \\nuv and \\fuv magnitudes with literature \\hans, \\firns, and thermal radio continuum measurements, we have found the following. \\begin{itemize} \\item Raw \\fuv and \\ha SFR estimates are consistent to within a factor of two.  At low luminosities, \\ha SFR estimates are around 25\\% lower then \\fuv SFR estimates, indicating either inconsistencies in the SFR calibrations, or the effects of low-level bursts of star formation.  At higher luminosities, \\ha SFR estimates are $\\sim\\,$1.5 times higher than \\fuv SFR estimates, with a factor-of-two scatter. This indicates that \\fuv attenuation is, on average, larger than \\ha attenuation.  Furthermore, the difference in \\ha to \\fuv SFR ratio between high and low luminosity galaxies suggests that there are differences in extinction properties which correlate at least loosely with luminosity. \\item For subsets of our sample of galaxies, we have constructed \\fuv attenuation estimates using the UV spectral slope as an attenuation indicator \\citep{calzetti94}, and \\ha attenuation estimates by comparing \\ha and thermal radio continuum luminosities. The median \\fuv attenuation is somewhat larger than the median \\ha attenuation, although with marginal significance. \\item Comparison of the \\ha attenuation with the \\firfuv for our galaxies demonstrates that there is a broad range in \\ha attenuation.  If \\firfuv is driven mainly by attenuation, there are suggestions that the \\ha attenuation correlates broadly with the \\fuv attenuation, and is around a factor of two lower than the \\fuv attenuation. \\item A characteristic \\ha and \\fuv attenuation for our sample of galaxies is $\\sim$ 0.9 mag and $\\sim$ 1.4 mag respectively, although there is a large range in attenuations in our sample, ranging from 0 to 4 magnitudes at \\hans. \\end{itemize}"
    },
    "0010/astro-ph0010448_arXiv.txt": {
        "abstract": "We present a series of systematic abundance measurements for a group of hot DA white dwarfs in the temperature range $\\approx 20,000-75,000$K, based on far-UV spectroscopy with STIS on \\hs , \\iu \\ and \\fuse . Using our latest heavy element blanketed non-LTE stellar atmosphere calculations we have addressed the heavy element abundance patterns for the hottest stars for the first time, showing that they are similar to objects like G191$-$B2B. The abundances observed in the cooler ($<50,000$K) white dwarfs are something of a mystery. Some of the patterns (e.g. REJ1032) can be explained by self-consistent levitation-diffusion calculations but there is then a serious difficulty in understanding the appearance of the apparently pure H atmospheres. We also report the detection of photospheric HeII in the atmosphere of WD2218+706. ",
        "introduction": "The existence of two distinct groups of hot white dwarfs, having either hydrogen-rich or helium-rich photospheres is now, qualitatively at least, understood to arise from the number of times the progenitor star ascends the red giant branch and the amount of H (and He) lost through the successive phases of mass-loss. Consequently, it seems clear that each group descends from their respective proposed progenitors, the H-rich and He-rich central stars of planetary nebulae (CSPN). Nevertheless, there remain several features of the white dwarf cooling sequence that cannot be readily explained. For example, while the hottest H-rich DA white dwarfs outnumber the He-rich DOs by a factor 7 (Fleming \\etal \\ 1986), the relative number of H- and He-rich CSPN is only about 3:1. In addition, there exists a gap in the He-rich track between $\\approx 45000$K and 30000K between the hot DO and cooler DB white dwarfs, confirmed by a detailed spectroscopic analysis of all the then known hot He-rich objects by Dreizler and Werner (1996).   To understand white dwarf evolution, we need to know accurately several physical parameters for each star. For example, a measurement of effective temperature ($T_{eff}$) establishes how far along its evolutionary sequence the star has progressed. A key result has been the establishment of the temperature scale for DA stars through the determination of \\teff \\ from the Balmer line profiles in optical spectra. The reliability of this work depends on the assumption that the Balmer line profile technique is a reliable estimator of \\teff \\ in all cases. However, Barstow Holberg \\& \\ Hubeny (1998) have shown that the presence of substantial blanketing from photospheric heavy elements does significantly alter the Balmer line profiles at a given effective temperature. Hence, the temperature scale of the hottest, most metal-rich DA white dwarfs, realised by studies using only pure H photospheric models, cannot be viewed as reliable and must be revised taking into account the photospheric composition of each star.  Until recently very few DA white dwarfs were known to have effective temperatures above $\\approx 70000$K. Therefore, the proposed direct evolutionary link between H-rich CSPN and white dwarfs has hardly been explored. In particular, there have been no measurements of the photospheric composition of what might be termed super-hot DAs, to distinguish them from the cooler ranges studied in detail. Using a new grid of non-LTE, heavy element-rich model atmospheres and a combination of \\hs \\ STIS and \\fuse \\ spectra, we have made the first measurements of photospheric abundance for a sample of very hot DA white dwarfs. For comparison, we have also re-examined the archive data of the cooler DA stars. ",
        "conclusions": "We have reported a series of systematic abundance measurements for a group of hot DA white dwarfs in the temperature range $\\approx 20,000-75,000$K, based on far-UV spectroscopy with STIS on \\hs , \\iu \\ and \\fuse . Using our latest heavy element blanketed non-LTE stellar atmosphere calculations we have addressed the heavy element abundance patterns for the hottest stars for the first time, showing that they are similar to objects like G191$-$B2B. The abundances observed in the cooler ($<50,000$K) white dwarfs are something of a mystery. Some of the patterns (e.g. REJ1032) can be explained by self-consistent levitation-diffusion calculations but there is then a serious difficulty in understanding the appearance of the apparently pure H atmospheres. New observations of other stars lying in this temperature range are need to test the self-consistent model calculations."
    },
    "0010/astro-ph0010481_arXiv.txt": {
        "abstract": "We present Johnson $B$ surface photometry for the UCM Survey galaxies. One-dimensional bulge-disk decomposition is attempted, discussing on fitting functions and computational procedures. The results from this decomposition, jointly with concentration indices and an asymmetry coefficient, are employed to study the morphological properties of these galaxies. We also compare our results with the previous morphological classification established using Gunn $r$ imaging data and with other samples of galaxies. No major statistical differences in morphology are found between red and blue data, although some characteristics such as size and luminosity concentration vary. We find a correlation between luminosity and size. Several parameters are used to segregate the objects according to their morphological type. ",
        "introduction": "The Universidad Complutense de Madrid Survey (UCM Survey List I, Zamorano et al. \\cite{Zam94}; List II, Zamorano et al. \\cite{Zam96}; List III, Alonso et al. \\cite{Gar99}) constitutes a representative and fairly complete sample of current star-forming galaxies in the Local Universe (Gallego \\cite{Gal99}). Its main purposes are to identify and study new young, low metallicity galaxies and to quantify the properties of the current star formation in the Local Universe. Another key goal is also to provide a reference sample for the studies of high-redshift populations, mainly dominated by star-forming galaxies (see, e.g., Hu et al. \\cite{Hu98} and Steidel et al. \\cite{Ste99}).  Specific details of the UCM sample concerning spectroscopic properties were presented in Gallego et al. (\\cite{Gal95}, \\cite{Gal96}, \\cite{Gal97}). Photometric properties in the Gunn $r$ band can be found in Vitores et al. (\\cite{Alv96a}, \\cite{Alv96b}) and near-infrarred data are available in Alonso-Herrero et al. (\\cite{Alm96}) and Gil de Paz et al. (\\cite{Gil99}).  In P\\'erez-Gonz\\'alez et al. (\\cite{yor}, hereafter Paper I), integrated Johnson $B$ photometry for the whole sample was presented. In this paper we will study the spatial concentration of the $B$ luminosity in these objects. This will allow us to obtain information about morphology and the different structures embedded in the galaxies. The effects of the photometric band on the determination of galaxy morphology will be evaluated through the comparison of the results presented here with those achieved with the Gunn $r$ data (Vitores et al. \\cite{Alv96a}, \\cite{Alv96b}).  The study of large-scale characteristics of the galaxies involves the quantitative measurement of structural parameters and light distribution. These parameters, which must describe the different components of galaxies (i.e., bulges, disks, bars), may be used to perform a morphological classification of the objects.  In this sense, we will attempt bulge-disk decomposition in one dimensional surface photometry radial profiles. Calculation of concentration indices and an asymmetry coefficient will also be done. All these data will be utilized to classify the galaxies according to their Hubble type.  The paper is structured as follows: we introduce the sample of galaxies and the Johnson $B$ observations in Sect.~\\ref{obs}. In Sect.~\\ref{data} the method used in this surface photometry study is detailed, including the explanation of the procedures followed in the bulge-disk decomposition and in the calculation of concentration indices and the asymmetry coefficient. The results and discussion about morphology are located in Sect.~\\ref{results}. Finally, Sect.~\\ref{correlations} presents the correlations found between several photometric parameters. A Hubble constant $H_{0}$=50\\,km\\,s$^{-1}$\\,Mpc$^{-1}$ and a deceleration parameter $q_0$=0.5 have been used throughout this paper. ",
        "conclusions": "We have carried out a morphological study of the UCM Survey galaxies based on Johnson $B$ imaging. This paper, jointly with Paper I (P\\'erez-Gonz\\'alez et al. \\cite{yor}), have analyzed the main features of the UCM sample concerning integrated and surface photometry in the $B$ bandpass.  Paper I presented integrated apparent and absolute $B$ luminosities as well as isophote 24 mag$\\cdot$arcsec$^{-2}$ magnitudes. Effective radii and $B-r$ colours were also calculated in this first release. In the present paper we have outlined the main results concerning bulge-disk decomposition values, ellipticities, position angles, concentration indices, mean photometric radii and surface brightnesses and asymmetry coefficients.  All the above information has been used to perform the morphological classification of the UCM galaxies. The sample is dominated by late Hubble type objects (65\\% being Sb or later). We have not found a great difference between this classification and the results achieved with the Gunn $r$ data (Vitores et al. \\cite{Alv96a}, \\cite{Alv96b}).  Our galaxies are characterized by shorter disks than those of normal spirals. Besides, they seem to be objects with a high luminosity concentration. A preliminary comparison between the characteristics of the sample in the red and blue bandpasses, yields to the result that emission-line galaxies have a higher concentration of blue light than red light in the inner parts of the objects; on the contrary, they also seem to be more extended in $B$ than in $r$.  Finally, a size-luminosity correlation has been outlined. We have also presented several plots where a morphological segregation is patent. These plots involve information about luminosity, concentration of light and asymmetry.  In next papers we will face a most exhaustive comparison between bands, including colour gradients and a stellar population study. Likewise, we will likened our sample to high-redshifts surveys searching for clues to their nature and galaxy evolution."
    },
    "0010/astro-ph0010162_arXiv.txt": {
        "abstract": "We present an analysis of \\beppo observations of three clusters of galaxies which are amongst the most luminous in the Universe: RXJ1347-1145, Zwicky~3146 and Abell~2390. Using data from both the Low Energy (LECS) and Medium Energy (MECS) Concentrator Spectrometers, and a joint analysis with the Phoswich Detection System (PDS) data above 10 keV, we constrain, with a relative uncertainty of between 7 and 42 per cent (90 per cent confidence level), the mean gas temperature in the three clusters. These measurements are checked against any possible non-thermal contribution to the plasma emission and are shown to be robust.  We confirm that RXJ1347-1145 has a gas temperature that lies in the range between 13.2 and 22.3 keV at the 90 per cent confidence level, and is larger than 12.1 keV at $3 \\sigma$ level. The existence of such a hot galaxy cluster at redshift of about 0.45 implies an upper limit on the mean mass density in the Universe, $\\Omega_{\\rm m}$, of 0.5.  Combining the \\beppo estimates for gas temperature and luminosity of the three clusters presented in this work with \\asca measurements available in the literature, we obtain a slope of 2.7 in the $L-T$ relation once the physical properties are corrected from the contamination from the central cooling flows. ",
        "introduction": "The X-ray emitting gas in clusters of galaxies cools by the emission of energy on a time scale which depends on the temperature and density of the intracluster medium (ICM). Where the density is higher, in the cores of the clusters, the cooling is more efficient and the temperature falls. To support the outer layers of gas, a subsonic flow of gas occurs towards the central region, producing a {\\it cooling flow} that appears in X-rays as an enhancement of the central peak in the surface brightness profile and with a spectrum that presents multiple temperature components.  The stronger the cooling flow, the higher the X-ray luminosity which arises from both the thermal energy lost by the cooling gas and the gravitational work done to maintain the gas at constant pressure. Therefore, to compare global cluster gas properties with theoretical and numerical predictions, that, at present, cannot model the radiative cooling in detail, we need to account fully for the effects of cooling flows on the temperature and luminosity measurements (Fabian et al. 1994, Allen \\& Fabian 1998, Markevitch 1998).  In this paper, we present data collected with the instruments onboard of the Italian-Dutch satellite \\beppo for three highly luminous ($L_{\\rm bol}>10^{45}$ erg s$^{-1}$), hot ($T_{\\rm gas} \\ga 10$ keV), distant ($z>0.2$) clusters of galaxies with large ($\\ga 1000 M_{\\odot}$ yr$^{-1}$) mass deposition rates in their cores (Allen 2000): RXJ1347-1145, Zwicky~3146 (Z3146) and Abell~2390 (A2390). Using data from both the Low Energy (LECS) and Medium Energy (MECS) Concetrator Spectrometer and a joint analysis with the Phoswich Detection System (PDS) data above 10 keV, we show that it is possible to constrain at a high level of significance the hard tail of the bremsstrahlung emission.  All quantities quoted in this paper refer to the following cosmological parameters: $H_0 = 50 h_{50}^{-1}$ km s$^{-1}$ Mpc$^{-1}$, $\\Omega_{\\rm m} = 1$, $\\Omega_{\\Lambda} = 0$. ",
        "conclusions": "From the \\beppo observation, we can constrain the ambient temperature in the clusters of galaxies RXJ1347-1145, Z3146 and A2390 with a relative uncertainty of (17/41), (11/41), (7/9) per cent in the (lower/upper) boundary (at the 90 per cent confidence level) respectively, when a cooling flow model is included in the spectral analysis.  Although the instrument characteristics of \\beppo do not allow us to constrain firmly any absorption intrinsic to the cluster atmosphere and/or the normalization of the emission from the central cooling gas, the measurements of the ambient gas temperature are in agreement, within the 90 per cent confidence level, with the results from the analysis of \\asca data in Allen (2000). In particular, the plasma temperature in RXJ1347-1145 appears well defined between 13.2 and 22.3 keV (90 per cent confidence level) and above 12.1 keV at the $3 \\sigma$ level. This measurement of the temperature is in good agreement with an independent estimate of 16.2 ($\\pm$ 3.8, $1 \\sigma$ level) keV obtained from the observed X-ray gas density and the measured Sunyaev--Zeldovich (1972) distorsion of the cosmic microwave background in the direction of RXJ1347-1145 (Pointecouteau et al. 1999; see also Komatsu et al. 1999).  \\begin{figure} \\hspace*{-1.cm} \\psfig{figure=fig5.ps,width=.6\\textwidth} \\caption[]{The expected number density of clusters with gas temperature larger than 13.2, 15.9, 22.3 keV (from right to left in the figure, respectively; cf. Table~2 for the constraint on the gas temperature of RXJ1347-1145) in the redshift interval $[0.4,0.5]$. The dashed lines represent the distribution in a ``$\\Omega_{\\rm m} +\\Omega_{\\Lambda}=1$\" Universe (thick) and in a open Universe (thin). The error bars, around the observed value of 1 cluster with the given characteristics, are at the 90 per cent level of confidence, according to the Poisson single-sided distribution tabulated in Gehrels (1986). } \\label{fig:cosmo} \\end{figure}  Following the arguments of many authors (e.g. Bahcall \\& Fan 1998, Donahue et al. 1999, Henry 2000) the existence of such a hot cluster at redshift 0.45 is highly unlikely in an Einstein -- de Sitter universe. We can estimate this probability, calculating the expected evolution of collapsed structures with a given minimum mass according to the Press-Schechter (1974) formalism. Integrating the number of collapsed objects with gravitating mass above that corresponding to a temperature of 15.9 keV over the redshift range of $0.4-0.5$, and requiring {\\it at least} the existence of RXJ1347-1145, the single detection provides a stringent upper limit of $\\Omega_{\\rm m} \\la$ 0.43. Considering also the lower and upper end of the range of the acceptable gas temperature of 13.2 and 22.3 keV, with the 90 per cent uncertainty on the single detection, we constrain the cosmological parameter $\\Omega_{\\rm m}$ between 0.25 and 0.70 (cf. Fig.~\\ref{fig:cosmo}).  Future {\\it Chandra} and {\\it XMM} observations of these highly luminous clusters of galaxies at intermediate redshift will resolve spatially the profiles of the gas temperature, abundance and deposition rate."
    },
    "0010/astro-ph0010212_arXiv.txt": {
        "abstract": "The CCD Johnson $BV$ and Cousins $RI$ photometric magnitudes are determined for 20 stars in the field of GRB 000926. They are used to calibrate the present $R$ as well as published $BVRI$ photometric magnitudes of the GRB 000926 afterglow. Optical light curves of the afterglow emission are obtained in $B$ and $R$ passbands. They show a steepening of the flux decay as expected of an anisotropic fireball losing collimation with the fall of the bulk Lorentz factor. We derive the early and late time flux decay constants as $1.4\\pm0.1$ and $2.6\\pm0.06$ respectively. Steepening in the flux decay seems to have started around 1.7 days after the burst. Negligible Galactic but relatively large intrinsic extinction amounting $E(B-V) =$ 0.03 and 0.36$\\pm$0.02 mag respectively are derived in the direction of GRB 000926. The value of the spectral index in the \\mbox{X-ray}--optical--near-infrared region is $\\sim -0.9$. The determination of the redshift $z=2.0369$ indicates a cosmological origin of the burst at a luminosity distance of 16.6 Gpc. The observed fluence in the energy range 20--100 keV indicates, if isotropic, the release of $\\ge 10^{53}$ ergs of energy. Attributing the observed break in the light curve at 1.7 days to the onset of sideways expansion of a jet-like ejecta, we infer an initial jet opening angle of $\\sim$0.14~radian. This indicates a large anisotropy in the original emission and the amount of released energy is reduced by a factor of $\\sim$ 100 relative to the isotropic value, which can be understood in terms of the currently popular stellar death models. ",
        "introduction": "We have used the data published in GCN circulars in combination with the present measurements to study the optical flux decay of GRB 000926 afterglow. Fig. 2 shows a plot of the photometric measurements as a function of time. The X-axis is log ($t-t_0$) where $t$ is the time of observation and $t_0$ (= 2000 September 26.9927 UT) is the time of GRB trigger. All times are measured in unit of day.   \\begin{figure*} \\begin{center}\\vspace*{-1.5cm}\\hspace*{-1cm} \\epsfig{file=light.ps,width=15.0cm} \\end{center}\\vspace*{-0.5cm} \\label{light} \\caption{Light curve of GRB 0000926 afterglow in optical $B$ and $R$ photometric passbands.  Flux decay can not be fitted by a single power-law. Solid lines represent the least square non-linear fits to the observed data for a jet model (see text).} \\end{figure*}  The emission from GRB 000926 OT is fading in both $B$ and $R$ passbands. The flux decay of most of the earlier GRB afterglows, barring a few, is generally well characterized by a single power law $F(t) \\propto (t-t_0)^{-\\alpha}$, where $F(t)$ is the flux of the afterglow at time $t$ and $\\alpha$ is the decay constant. In contrast, the optical light curves of GRB 000926 (Fig. 2) can not be fitted by a single power-law. Overall the OT flux decay seems to be described by a broken power-law as expected in GRB afterglows having jet-like relativistic ejecta (Sari et al. 1999; Rhoads 1999). Broken power-law in these light curves can be empirically fitted by functions of the form (see Sagar et al. 2000b for details)  $F(t) = 2F_0/[(t/t_b)^{\\alpha_1 s} + (t/t_b)^{\\alpha_2 s}]^{1/s} + F_g$,  \\noindent where $F_g$ is the constant flux from the underlying host galaxy, $\\alpha_1$ and $\\alpha_2$ are asymptotic power-law slopes at early and late times with $\\alpha_1 < \\alpha_2$.  The parameter $s (> 0)$ controls the sharpness of the break, a larger $s$ implying a sharper break. With $s = 1$, this function becomes the same as that used by Stanek et al. (1999) to fit the optical light curve of GRB 990510 afterglow. $F_0$ is the flux of afterglow at the cross-over time $t_b$. The function describes a light curve falling as $t^{-\\alpha_1}$ at $t << t_b$ and $t^{-\\alpha_2}$ at $t >> t_b$. In jet models, an achromatic break in the light curve is expected when the jet makes the transition to sideways expansion after the relativistic Lorentz factor drops below the inverse of the opening angle of the initial beam. Slightly later, the jet begins a lateral expansion which causes a further steepening of the light curve.  We use the observations in $B$ and $R$ to determine the parameters of the jet model using the above function. We find that the minimum value of $\\chi^2$ is achieved for $s \\ge 5$. This indicates that the observed break in the light curve is sharp, unlike the smooth break observed in the optical light curve of GRB 990510 (cf. Stanek et al. 1999; Harrison et al. 1999) but similar to the sharp break observed in the optical and near-IR light curves of GRB 000301C (cf. Sagar et al. 2000b).  In order to avoid a fairly wide range of model parameters for a comparable $\\chi^2$ due to degeneracy between $t_b, m_b$ and $m_g$ (magnitudes corresponding to $F_0$ and $F_g$ respectively), $\\alpha_1$, $\\alpha_2$ and $s$, we have used a fixed value of $s = 5$ in our further analyses.  The least square fit values of the parameters $t_b$, $m_b$, $\\alpha_1$, and $\\alpha_2$ then work out to be 1.74$\\pm$0.11~d, 21.27$\\pm$0.15, 1.45$\\pm$0.06 and 2.57$\\pm$0.10 respectively in $R$ band, with a corresponding $\\chi^2$ of 0.93 per degree of freedom ($DOF$).  The fit yields the host galaxy contribution to be $m_g$=24.41$\\pm$0.25 in $R$. This is slightly fainter than the value $\\sim$ 23.9 estimated by Fynbo et al. (2000e) from deep CCD imaging, but brighter than $R$=25.19$\\pm$0.17 derived from HST measurements (Price et al 2000b). The main reason for the discrepancy with the HST estimate is possibly the restriction of the HST photometry to within a 1.5 arcsec aperture, while no such cutoff has been applied to the photometric data used here. The derived values of the above parameters do not change significantly with larger adopted values of $s$, although a marginal decrease in $\\chi^2/DOF$ is noticed in doing so.  In $B$ band the available data is sparse, and a fit with fixed $s=5$ and $t_b=1.74$~d yields $\\alpha_1$=1.55$\\pm$0.11, $\\alpha_2$=2.72$\\pm$0.48 and $m_g(B)$=24.7$\\pm$1.0.  Note the poor constraint on $\\alpha_2$ and B-magnitude of the host galaxy, because of the scarcity of measurements in the late light curve.  One can improve the constraints on the decay indices somewhat by assuming a colour index for the host galaxy, and thereby fixing its B magnitude from the R magnitude determined above.  The restricted aperture HST photometry reported by Price et al (2000b) indicate a $B-R$ colour index of 1.04$\\pm$0.53 for the host galaxy.  Fixing the value of $m_g(B)$ at 25.5 using this mean colour index, the fit yields $\\alpha_1$=1.53$\\pm$0.11 and $\\alpha_2$=2.52$\\pm$0.12.  On the other hand, if the host galaxy has a flat spectral distribution similar to that for GRB~990123 (Castro-Tirado et al.\\ 1999), then $B-R$ colour will be $\\sim 0.5$ (which is allowed within the errors of the HST measurements and our light curve fits), and $m_g(B)$ will be 24.9.  If we use this value for $m_g(B)$, then the best fit values of the decay constants become $\\alpha_1$ = 1.54$\\pm$0.10 and $\\alpha_2$ = 2.64$\\pm$0.12.  We can therefore say with confidence that within the range of host galaxy colours allowed by the light curves and the HST measurements, the decay indices in $B$ and $R$ bands are consistent with each other, and that the break in the light curve displays the achromaticity expected in the case of a lateral expansion of collimated ejecta.  The best fit light curves for the $R$ and the $B$ bands, with an assumed host galaxy colour index of $B-R$=1.04, are displayed in Fig.~2.  The decay parameters derived here are consistent with those obtained by Price et al (2000b) from their measurements.  \\subsection{Spectral index of the GRB 000926 afterglow }  We have constructed the GRB 000926 afterglow spectrum on 2000 September 29.25 UT and 29.76 UT using the available optical, near-IR, radio and \\mbox{X-ray} observations. These epochs were selected for the long wavelength coverage possible at the time of observations of \\mbox{X-ray} flux by Piro \\& Antonelli (2000) and Garmire et al. (2000). Optical fluxes at the wavelengths of $BVR$ and $I$ passbands have been derived using the slope of the fitted light curve shown in Fig. 2 for the epochs under consideration. The fluxes for the desired epoch at 8.46 GHz and $J$ wavelength are taken from Frail \\& Berger (2000) and Di Paola et al. (2000) respectively assuming that the flux decay is similar to optical one. We used the reddening map provided by Schlegel, Finkbeiner \\& Davis (1998) for estimating Galactic interstellar extinction towards the burst and found a small value of $E(B-V) = 0.03$ mag. We used the standard Galactic extinction reddening curve given by Mathis (1990) in converting apparent magnitudes into fluxes and used the effective wavelengths and normalisations by Fukugita et al. (1995) for $BVR$ and $I$ and by Bessell \\& Brett (1988) for $J$. The fluxes thus derived are accurate to $\\sim$ 10\\% in optical and $\\sim$ 30\\% in $J$. Fig. 3 shows the spectrum of GRB 000926 afterglow from \\mbox{X-ray} to radio region. It is observed that as the frequency decreases the flux increases from \\mbox{X-ray} to radio wavelengths. For a chosen frequency interval, we describe the spectrum by a single power law: $F_{\\nu}\\propto\\nu^{\\beta}$, where $F_{\\nu}$ is the flux at frequency $\\nu$ and $\\beta$ is the spectral index. In the \\mbox{X-ray} to optical region at $\\Delta t$ = 2.26 day, the value of $\\beta$ is $-0.85\\pm$0.02, while in the \\mbox{X-ray} to $J$ region, the value is $-0.88\\pm$0.04. At $\\Delta t$ = 2.77 day the value of $\\beta$ is $-0.94\\pm$0.04 for \\mbox{X-ray} to $J$ region. These values agree very well with the spectral slopes derived for the \\mbox{X-ray} spectrum of the GRB 000926 afterglow by Piro \\& Antonelli (2000) and Garmire et al. (2000). During $\\Delta t$ = 1 to 4 days, the values of $(B-R) $ and $(B-I)$ are almost the same indicating no change in the optical spectral index before and after the time $t_b$. However, the values of $\\beta$ in optical ($-1.54\\pm0.15$ and $-1.61\\pm0.14$ derived at $\\Delta t= 0.9$ and 3.9 day respectively) and in optical to $J$ region ($-1.68 \\pm0.07$ at $\\Delta t= 2.26$ day) are much steeper than those derived for the broadband (X-ray to optical) spectrum. This may indicate presence of intrinsic extinction in the GRB 000926 afterglow. The amount of intrinsic extinction is determined in the following way assuming that the OT spectrum follows a single spectral slope from \\mbox{X-ray} to near-IR. The observed flux at \\mbox{X-ray} and $J$ wavelengths are used to determine $\\beta$ as the extinction affects them least. This slope predicts the fluxes of GRB 000926 OT at $BVRI$ wavelengths. The differences between these predicted and the corresponding Galactic extinction corrected observed fluxes are used to evaluate the intrinsic extinction. From these values, intrinsic extinction at $J$ is determined assuming that it follows the Galactic extinction law given by Mathis (1990). The extinction corrected flux at $J$ and the observed \\mbox{X-ray} flux are used to derive modified value of $\\beta$. This procedure is iterated till self consistent values of intrinsic extinction at $BVRI$ and $J$ wavelengths are obtained. This yields $E(B-V) = 0.36\\pm0.02$ mag and $A_J = 0.31\\pm0.04$ as the values for the intrinsic extinction. These are in agreement with the values determined independently by Price et al (2000b) from their optical data. The observed fluxes corrected for both Galactic and intrinsic extinction are also shown in Fig. 3. The values of $\\beta$ derived from them are $-0.94\\pm0.02$ and $-1.02\\pm0.02$ at $\\Delta t =$ 2.26 and 2.77 days respectively.  In the light of above, we conclude that the value of $\\beta$ is $\\sim -0.95$ in  \\mbox{X-ray} to near-IR spectral region and it has not changed before and after $t_b$ during $\\Delta t=$ 1 to 4 days.  Fig. 3 indicates that the peak frequency appears to lie in the millimeter region. This peak frequency is thus similar to that of GRB 970508 (cf. Galama et al. 1998) and GRB 000301C (cf. Sagar et al. 2000b) but different from that of GRB 990123 (Galama et al. 1999) where the peak is in radio region and that of GRB 971214 for which the peak is in optical/near-IR waveband (Ramaprakash et al. 1998). From this, one may infer that the synchrotron peak frequency may span a large range in GRB afterglows.  \\begin{figure*} \\begin{center}\\vspace*{-0.5cm}\\hspace*{-1cm} \\epsfig{file=flux.ps,width=10.0cm,angle=270} \\end{center}\\vspace*{-0.6cm} \\label{spec} \\caption{The spectral flux distribution of the GRB 000926 afterglow at $\\sim$ 2.26 and 2.77 days after the burst.  The least square linear relations derived using fluxes at \\mbox{X-ray}, $BVRI$ optical and $J$ near-IR are shown by solid lines for both epochs. Effect of intrinsic extinction on the spectral slope can be seen from the two plots. } \\end{figure*} ",
        "conclusions": ""
    },
    "0010/astro-ph0010024_arXiv.txt": {
        "abstract": "We explore the predictions of the standard hierarchical clustering scenario of galaxy formation, regarding the numbers and metallicities of PopIII stars likely to be found within our Galaxy today. By PopIII we shall be referring to stars formed at large redshift ($z>4$), with low metallicities ($[Z/Z_{\\odot}]<-2.5$) and in small systems (total mass $\\simlt$ $2\\times 10^{8} M_{\\odot}$) that are extremely sensitive to stellar feedback,  and which through a prescribed merging history (Lacey \\& Cole 1993) end up becoming   part of the Milky Way today. An analytic, extended Press-Schechter formalism is used to get the mass functions of halos which will host PopIII stars at a given redshift, and which will end up in Milky Way sized systems today. Each of these is modeled as a mini galaxy, with a detailed treatment of the dark halo structure, angular momentum distribution, final gas temperature and disk instabilities, all of which determine the fraction of the baryons which are subject to star formation. Use of new primordial metallicity stellar evolutionary models allows us to trace the history of the stars formed, give accurate estimates of their expected numbers today, and their location in $L/L_{\\odot}$ vs. $T/K$ HR diagrams. A first comparison with observational data suggests that the IMF of the first stars was increasingly high mass weighted towards high redshifts, levelling off at $z\\simgt  9$ at a characteristic stellar mass scale $m_s=10-15 M_\\odot$. ",
        "introduction": "The formation and the properties of the first stars in the universe are probably among the most fascinating problems in present-day cosmology. Numerical simulations and analytical studies in a cosmological context (Tegmark \\etal 1997, Abel \\etal 1998, Bromm, Coppi \\& Larson 2000, Machacek, Bryan \\& Abel 2000) have reached a consensus on the fact that the fragments potentially leading to star formation were already available inside very small dark matter halos at redshift $z\\approx 30$ in most cosmological models. After that point the requirements concerning the dynamic range and number of physical processes to be treated become so demanding that one has to turn to simplified and/or lower dimensional studies. Although considerably refined, these studies (Omukai \\& Nishi 1999, Susa \\& Umemura 2000, Nakamura \\& Umemura 2000, Ripamonti \\etal 2000) have not yet reached an agreement on the mass of the stellar seed (core) and on the accretion rate by which the former grows into  a star. Determining if and when the accretion is quenched seems to hold the key for the understanding of the properties of the first stars. However, the large variety  of physical effects to be considered (for a review see Ferrara 2000) again complicate the problem tremendously.  On general grounds some authors have speculated that the IMF of the first stars can be skewed towards high stellar masses (Larson 1998) or even be bimodal (Nakamura \\& Umemura 2000). One of the consequences of this assumption is an increase of the relative number of massive, short-lived stars. These stars    will soon end their life as supernovae, leaving either a compact remnant (a neutron stars or a black hole) or nothing at all (for masses above $ \\approx 120 M_\\odot$, see Umeda \\etal 2000 and Fryer, Woosley \\& Heger 2000), and possibly releasing detectable bursts of gravitational waves (Schneider \\etal 2000). In turn, this implies a strong decrease of the amount of zero- or very low-metallicity stars presumably  included in the present-day Milky Way (MW), according to the hierarchical paradigm. Given the above discussion it appears that number counts of very low-metallicity MW stars can not only provide evidences for ``living fossils\", \\ie stars formed immediately after the end of Dark Ages, but also probe the primordial IMF and hierarchical models to cosmic epochs ($z\\simgt 6$) currently unreachable to essentially all experiments.  We propose here that the first stars are born inside the so-called PopIII objects (Haiman \\& Loeb 1996, Gnedin \\& Ostriker 1997, Ciardi \\etal 2000) \\ie systems with total mass $\\simlt  2\\times 10^8 M_\\odot$, which strongly rely on molecular hydrogen cooling in order to collapse, given their low virial temperatures; in short, we define as PopIII stars those formed inside such mini galaxies. As Ciardi \\etal have pointed out, these objects are extremely fragile to the energy release by their own supernovae; as a result, their gas is quickly evacuated and they can only witness a single burst of star formation.  The remnant is a \"naked stellar object\" (Ciardi \\etal 2000), \\ie a tiny agglomerate  of long lived low-mass stars, essentially retaining the metallicity of the gas from which they formed. These are the stars that we are considering as candidates for the very low-metallicity stars observed in the MW halo.  Our paper is structured as follows: the next section presents the Press-Schechter formalism used to calculate the mass functions of MW sub-galactic clumps at different redshifts. In Section 3 we describe the treatment of PopIII halos  as mini galaxies to estimate the number of stars each contributes to the final MW. Section 4 gives a comparison with recent observational results of the numbers and metallicity distribution of extremely low metallicity halo stars. Finally, a brief summary is given in Section 5. ",
        "conclusions": "From our study we conclude that:  \\begin{itemize} \\item If the standard hierarchical merger scenario is correct, in the MW one should not expect to find stars having metallicities lower than $[Z/Z_{\\odot}]=-4.3$, consistent with current observational limits.  \\item We infer the IMF of PopIII stars to have been strongly weighted towards high masses, and increasingly more so in the range $6<z<9$, this levelling off at $z\\simgt 9$ at a characteristic mass scale $m_{s}=10-15 M_{\\odot}$.  \\item Observational limits on the numbers and metallicity distribution of very low metallicity stars in the MW can place sharp restrictions on the IMF of the first stars. This can be achieved through detailed modelling of the full range of phenomena giving rise to such stars, of the type attempted here, thus necessarily establishing a connection with the underlying structure formation scenario.  \\end{itemize}"
    },
    "0010/hep-th0010280_arXiv.txt": {
        "abstract": "We present the canonical and quantum cosmological investigation of a spatially flat,  four-di\\-men\\-sio\\-nal Friedmann-Robertson-Walker (FRW) model that is derived from the M-theory effective action obtained originally by Billyard, Coley, Lidsey and Nilsson (BCLN). The analysis makes use of two sets of canonical variables, the Shanmugadhasan gauge invariant canonical variables and the ``hybrid'' variables which diagonalise the Hamiltonian. We find the observables and discuss in detail the phase space of the classical theory. In particular, a region of the phase space exists that describes a four-dimensional FRW spacetime first contracting  from a strong coupling regime and then expanding to  a weak coupling regime, while the internal space ever contracts. We find the quantum solutions of the model and obtain the positive norm Hilbert space of states. Finally, the correspondence between wave functions and classical solutions is outlined. ",
        "introduction": "} The search for a theory of quantum gravity constitutes one of the uppermost challenges in theoretical high energy physics. The need for quantum gravity finds its roots within Einstein general relativity. Powerful general theorems \\cite{EH} imply that our universe must have started from an initially singular state with infinite curvature. In such circumstances, where the laws of classical physics break down, it is unclear how any boundary conditions necessary for a description of a dynamical system could have been imposed at the initial singularity. Quantum corrections could induce a modification of classical general relativity and strongly influence the evolution of the very early universe \\cite{QG}.  In the last two decades superstring theory \\cite{strings} has emerged as a successful candidate for the theory of quantum gravity. In cosmology, most of the modifications to general relativity induced by superstring theory are originated by the inclusion of the dilaton, axion and various moduli fields, together with higher curvature terms that are present in the low-energy effective actions. Each of these novel ingredients do indeed lead to new cosmological solutions. A remarkable example is given by the so-called pre-big bang cosmologies \\cite{prebb} that are derived by the low-energy effective string action. Different branches of the solution are related by time reflection and internal transformations -- $O(d,d)$ and, in particular, scale factor duality --  that follow from the $T$-duality property of the full superstring theory \\cite{duality}. According to the pre-big bang scenario the universe evolves  from a weak coupled string vacuum state to a radiation-dominated and then matter-dominated FRW geometry through a region of strong coupling and large curvature. Although the pre-big bang model has not yet proven able to solve all its pitfalls, such as the existence of a singular boundary that separates the pre- and post-big bang branches \\cite{exit}, it provides a good starting point to investigate high-energy cosmology.  Recently, it has become evident that the five consistent, anomaly free, perturbative formulations of ten-dimensional superstring theories are connected by a web of duality transformations \\cite{strings} and constitute special points of a large, multi-dimensional moduli space of a more fundamental (non-perturbative) theory, designated as M-theory \\cite{M-th}. Quite interestingly, it happens that another point of the M-theory moduli space corresponds to eleven-dimensional supergravity, the low-energy limit of M-theory \\cite{M-th,Townsend}. Assuming that M-theory is the ultimate theory of quantum gravity, it is natural to begin exploring its cosmological implications. Although our understanding of M-theory is still incomplete, there are hopes that some of the obstacles of dilaton driven inflation in string theory could be overcome within the new theory. The underlying idea is to investigate the dynamics at the extreme weak- and strong-coupling regimes of superstring theory from a M-theory perspective, where the existence of eleven dimensions seems mandatory.  Several different approaches to M-theory cosmology have already been explored in the literature \\cite{LOW},\\cite{RS},\\cite{DH}-\\cite{BCLN}. In the framework of the Ho{\\v r}ava-Witten model \\cite{HW}, M-theory and cosmology have been combined in the works of Lukas, Ovrut and Waldram \\cite{LOW}. A somewhat related line of research is the brane world by Randall and Sundrum \\cite{RS}, where our four-dimensional universe emerges as the world volume of a three-brane in a higher-dimensional spacetime. From a different point of view, Damour and Henneaux have investigated chaotic models \\cite{DH} and Lu, Maharana, Mukherji and Pope have discussed classical and quantum M-theory models with homogeneous graviton, dilaton and antisymmetric tensor field strengths \\cite{LMMP}. Different classes of cosmological solutions that reduce to solutions of string dilaton gravity have been discussed in Ref.\\ \\cite{cosm}-\\cite{KKO}. In particular, a global analysis of four-dimensional cosmologies derived from M-theory and type $IIA$ superstring theory has been presented by BCLN \\cite{BCLN}. Using the theory of dynamical systems to determine  the qualitative behaviour of the solutions, the authors find that form-fields associated with the NS/NS and R/R sectors play a rather crucial role in determining the dynamical behaviour of the solutions: the R/R fields cause the universe to collapse but the NS/NS fields, such as the axion, have opposite effect. Quite interestingly, for flat FRW models the boundary of the classical physical phase space is a set of invariant submanifolds, where either the axion field is trivial or the four-form field strength is dynamically unimportant. This interplay leads to important consequences, as the orbits in the phase space are dominated by the dynamics associated with one, or the other, or both invariant submanifolds in sequence, shadowing trajectories in the invariant submanifold \\cite{BCLN}.  In this paper we discuss the main scenario introduced by BCLN from a canonical perspective. This approach allows us to analyse in depth the physical properties of the classical solutions and to obtain a consistent quantum description of the model. We consider the bosonic sector of eleven-dimensional supergravity which consists of a graviton and an antisymmetric three-form potential. The theory is compactified to four dimensions by assuming a geometry of the form $M^4 \\times T^6 \\times S^1$, where $T^6$ is a six-dimensional torus and $M^4$ corresponds to a spatially flat FRW spacetime. The effective theory in four dimensions bears a dilaton $\\phi$, a modulus field $\\beta$ identifying the internal space, a pseudo-scalar axion field $\\sigma$ and a potential term induced by a four-form field. [See Eq.\\ (\\ref{b5}) below.] A brief derivation of the previous steps is presented in Section 2, where the two invariant submanifolds are identified. In Section 3 we analyse the first invariant submanifold, where the four-form field is negligible and the axion field dominates. In particular, in Subsection 3.1 we discuss the parameter space of the classical theory and in Subsection 3.2 we find the Hilbert space of states of the quantum theory. This programme is performed by using two sets of canonical variables, the so-called Shanmugadhasan variables \\cite{Shan} (forming a maximal set of gauge invariant variables)  and the ``hybrid\" variables that diagonalise the Hamiltonian. In Section 4, we repeat the analysis of Section 3 for the invariant manifold with negligible axion field and dominant four-form field. Finally, our conclusions are presented in Section 5. Two appendixes contain technical details and complete the paper. ",
        "conclusions": "} In this paper we have analysed a spatially flat, four-dimensional cosmological model derived from the M-theory effective action [Eq.\\ (\\ref{b1})]. The eleven-dimensional metric is first compactified on a one-dimensional circle to obtain the type IIA superstring effective action and then on a six-torus to obtain the effective four-dimensional theory. In our investigation we concentrated the attention on the boundary of the physical phase space of the theory which is constituted by two invariant submanifolds, where either the axion field (from the NS/NS sector) is negligible or the four-form R/R field strength is irrelevant. In our discussion we have heavily employed the canonical formalism. This approach makes the analysis of the features of the classical solution extremely simple and allows a straightforward quantization of the theory.  We found regions in the moduli space where a four-dimensional FRW universe evolves from a strong coupling regime towards a weak coupling regime, both internal six-volume and eleven-dimension contracting. In some cases, the dynamics is characterized by an early accelerated (inflationary) expansion with the spacetime eventually approaching a standard FRW decelerated expansion. Such scenario happens when the pseudo-scalar axion field dominates and the R/R field strength is negligible. (See also Ref.\\ \\cite{CLW}.) When the axion is subdominant w.r.t. the R/R field, different cosmological scenarios appear on the scene. In this case an ever expanding external spacetime is characterized at large times by an infinite ten-dimensional coupling constant (the size of eleven dimension). This seems to rule out this model from a physical point of view. On the contrary, for bouncing external spacetimes, where the external dimensions recollapse after initially expanding, geometries with early accelerated expansion, decreasing ten-dimensional coupling constant and decreasing internal space are found.  The quantization of the two invariant submanifolds can be performed exactly and the Hilbert space of states is derived. In the quantum framework our analysis allows to identify the quantum states that correspond to the different classical behaviours. In the hybrid representation we have identified regions in the space of parameters where the wave function of the universe is either oscillating or exponentially decaying. These regions are determined by  the inverse exponential function of the four-dimensional (unshifted) dilaton, i.e., by the four-dimensional string coupling, and correspond to classically allowed and classically forbidden regions, respectively. Starting from the Hilbert space of states, the quantum mechanics of M-theory cosmology can be constructed with aid of usual elementary quantum mechanics techniques.  The results summarized in the preceeding paragraphs strengthen and extend previous results obtained within different approaches \\cite{LMMP}-\\cite{BCLN}. Although the models investigated in this paper represent a severe, yet consistent, truncation of the original eleven-dimensional action Eq.\\ (\\ref{b1}), new interesting physical output can be extracted. In particular, the quantum mechanics of M-theory that has been presented here may constitute the basis for further insights in high-energy (quantum) M-theory cosmology. {\\em If} M-theory is the ultimate theory of quantum gravity, M-theory quantum cosmology is a subject which is certainly worth exploring. From this point of view related topics of investigation which deserve attention are the study of inhomogenous perturbations of the fields and a complete analysis of FRW models in the full phase space of the theory, when all form fields are excited. In the latter case, as Damour and Henneaux have recently observed \\cite{DH}, the dynamics may be qualitatively different and lead to potentially new theoretical and observational effects."
    },
    "0010/astro-ph0010268_arXiv.txt": {
        "abstract": "We discuss the space-and-time-dependent Monte Carlo code we have developed to simulate the relativistic radiation output from compact astrophysical objects, coupled to a Fokker-Planck code to determine the self-consistent lepton populations. We have applied this code to model the emission from a magnetized neutron star accretion shell near the Alfv\\'en radius, reprocessing the radiation from the neutron sar surface. We explore the parameter space defined by the accretion rate, stellar surface field and the level of wave turbulence in the shell. Our results are relevant to the emission from atoll sources, soft-X-ray transient X-ray binaries containing weakly magnetized neutron stars, and to recently suggested models of accretion-powered emission from anomalous X-ray pulsars. ",
        "introduction": "The high energy radiation from compact astrophysical objects is emitted by relativistic or semi-relativistic thermal and nonthermal leptons (electrons and pairs) via synchrotron, bremsstrahlung, and Compton processes, plus bound-bound and bound-free transitions of high-Z elements.  Since Compton scattering is a dominant radiation mechanism in this regime, the most efficient and accurate method to model the transport of high energy radiation is the Monte Carlo (MC) technique.  During the past decade we have developed a versatile state-of-the-art space-and-time-dependent MC code to model the radiative output of compact objects (see e.g. Liang et al. \\markcite{liang00}2000). Recently we have added the self-consistent evolution of the leptons using a Fokker-Planck scheme.  The lepton evolution is then coupled to the photon transport. Since this is the first time that we report on results obtained with this code, the first part of the present paper is devoted to a detailed description of the capabilities of the code (\\S 2) and its verification in comparison with previous work (\\S 3).  In the second part of this paper (\\S 4), we apply this coupled MC-FP code to the study of the reprocessing of blackbody radiation from a magnetized neutron star surface by a magnetized spherical shell at the Alfv\\'en radius. Recent observations of weak-field neutron star binaries, such as low-luminosity, X-ray bursters (atoll sources, e.g., 4U~1608-522: \\cite{zhang96}, 4U~1705-44: \\cite{barret96}), bursting soft X-ray transients (e.g., Aql~X-1: \\cite{harmon96}), or pulsar binary systems (e.g., PSR~B1259-63: \\cite{tavani96}) indicate that many of them exhibit soft (photon index $\\gtrsim 2$) power law tails extending beyond $\\sim 100$~keV, at least episodically, in addition to the thermal component at temperatures of $\\sim$~a few keV, which presumably originates from the stellar surface. The luminosity of this high-energy tail appears to be anti-correlated with the soft X-ray luminosity (\\cite{bv94,tl96}).  The origin of this high energy tail is unexplained at present. It could be due to thermal Comptonization by a hot coronal plasma, or it could be due to nonthermal emission.  Tavani and Liang (\\markcite{tl96}1996) examined systematically the possible sites of nonthermal emissions and concluded that the Alfv\\'en surface is the most likely candidate since the dissipation of the rotation energy of the disk is strongest there, due to magnetic reconnection and wave turbulence generation.  Here we first focus on particle acceleration by wave turbulence.  Because of the 1-D nature of our code at this stage, the neutron star is assumed to emit isotropic blackbody radiation, the reprocessing magnetospheric plasma is idealized as part of a spherical shell at the Alfv\\'en radius, and the magnetic field is taken to be nondirectional, so that the radiation output is isotropic.  Even though this is not a perfect representation of a quasi-dipolar magnetosphere and accretion flow, we believe that, except for very special viewing angles, our output spectra should be a reasonable approximation to the angle-averaged output of the reprocessing by a hybrid thermal/nonthermal magnetosphere. We assume that the leptons are energized by Coulomb collisions with virial ions and accelerated nonthermally by Alfv\\'en and whistler wave turbulence, and cooled by cyclotron/synchrotron, bremsstrahlung, and inverse Comptonization of both internal soft photons and blackbody photons from the stellar surface.  The primary focus of the parameter study presented in \\S 4 is the application to weakly magnetized neutron stars with surface magnetic fields of $B_{\\rm surf} \\lesssim 10^{11}$~G. In particular, we will show that the anti-correlation of the hardness and luminosity of the hard X-ray emission with the soft X-ray luminosity is a natural consequence of the energetics of particle acceleration and cooling near the Alfv\\'en radius. We predict that the nonthermal tails in the hard X-ray spectra of accreting, weakly magnetized neutron stars may extend up to $\\sim 1$~MeV. A solid detection and the measurement of the cutoff energy of these high-energy tails by the INTEGRAL mission, scheduled for launch in 2002, will provide important constraints on accretion-based models for the hard X-ray emission from accreting neutron stars.  In this context, it is interesting to note that Chatterjee, Hernquist, \\& Narayan (\\markcite{chat00}2000; see also \\cite{ms95,wang97,ch00}) have recently proposed a similar type of accretion-powered emission for anomalous X-ray pulsars (AXPs), as an alternative to models based on magnetic-field decay (\\cite{td96}) or residual thermal energy (\\cite{hh97}). According to Chatterjee et al. (\\markcite{char00}2000) the X-ray emission from AXPs (which generally consists of a soft, thermal component with $kT \\sim 0.3$ -- 0.4~keV plus a hard X-ray tail with photon index $\\Gamma \\sim 3$ -- 4) is powered by accretion of material from the debris of the supernova which had formed the neutron star, onto a the surface of the neutron star, which possesses a typical pulsar magnetic field of $B_{\\rm surf} \\sim 10^{12}$~G. Therefore, we extend our parameter study to parameter values relevant to accreting pulsars. However, we point out that in the case of a magnetic field as high as $B_{\\rm surf} \\sim 10^{12}$~G, the assumed shell geometry and the quasi-isotropy of the emission from the neutron star surface may be a gross over-simplification. However, although consequently the precise parameter values used in this region of the parameter space should not be taken at face value, our parameter study might still provide interesting insight into the dependence of the equilibrium electron and photon spectra on the various input parameters in the high-magnetic-field case. ",
        "conclusions": "We have reported on the development of a new, time-dependent code for radiation transport and particle dynamics. The radiation transport, accounting for Compton scattering, bremsstrahlung emission and absorption, cyclotron and synchrotron emission and absorption, and pair processes, is done using a Monte-Carlo method, while the electron dynamics, including radiative cooling, Compton heating/cooling, and stochastic acceleration by resonant interaction with Alfv\\'en/whistler wave turbulence, are calculated using an implicit Fokker-Planck scheme, coupled to the Monte-Carlo radiation transfer code.  In the second part of this paper, we have applied our code to the static situation of a shell at the Alfv\\'en radius of a magnetized neutron star. This situation is representative for accretion onto weakly magnetized neutron stars (bursting atoll sources or soft X-ray transients containing wekaly magnetized neutron stars) as well as to recently suggested models of accretion-powered emission from anomalous X-ray pulsars.  The main results of our parameter study are:  (1) The lepton thermal temperature increases, and the hard X-ray photon spectrum becomes harder as the accretion rate is decreasing. At the same time, the normalization of the hard X-ray power-law, relative to the thermal blackbody from the neutron star surface, becomes smaller.  (2) The nonthermal tails in the electron and photon spectra become more dominant and harder as the turbulence level is increasing. At the same time, the quasi-thermal electron temperature decreases, leading to a softer hard-X-ray spectrum.  (3) For low accretion rates ($\\lesssim$~a few \\%), the photon spectra are only very weakly dependent on the magnetic moment of the neutron star. For higher accretion rates, an increasing neutron star magnetic moment leads to a moderate hardening of the hard X-ray spectrum.  Our results are in good agreement with the non-detection of hard X-ray tails during soft X-ray high states of systems containing weakly magnetized neutron stars. However, we predict that a hard X-ray excess beyond $\\sim 20$~keV should exist even in the high/outburst state. This excess should have a cutoff energy of $E_{\\rm c} \\sim 100$ -- 200~keV. In the low/quiescent state, the predicted spectral indices of $\\Gamma \\sim 2$ -- 3 are in excellent agreement witht he observed hard X-ray excesses observed from sources believed to contain weakly magnetized neutron stars. We predict that these spectra should extend out to $E_{\\rm c} \\lesssim 1$~MeV.  In the strong-field case, representative of a recently suggested model for anomalous X-ray pulsars, we expect a strong correlation between the hard X-ray spectral hardness and the soft X-ray luminosity only in the case of very low accretion rates and high magnetic turbulence level. The predicted hard X-ray spectral indices are generally in very good agreement with the observed values of $\\Gamma \\sim 3$ -- 4. Cutoff energies of $E_{\\rm c} \\sim 100$ -- 500~keV are predicted.  Dedicated, deep observations by the upcoming INTEGRAL mission should be able to establish the existence of the hard power-law tails predicted in the models discussed here, and to constrain the high-energy cutoff of these tails. These measurements will be essential ingredients for a more detailed modeling of the physical conditions governing the accretion onto magnetized neutron stars."
    },
    "0010/astro-ph0010097_arXiv.txt": {
        "abstract": "The Wisconsin H-Alpha Mapper (WHAM) has completed a survey in Balmer $\\alpha$ of the entire Northern sky above declination $-30\\deg$. This survey provides the first calibrated, velocity-resolved map of the H$\\alpha$ emission from the Galaxy. With one-degree spatial resolution, 12 km s$^{-1}$ velocity resolution, and sensitivity to features as faint as 0.1 R (EM $\\sim$ 0.2 cm$^{-6}$ pc), this survey provides the deepest maps of the ionized content of the Galaxy to date. In addition to probing the detailed kinematic structure of the Warm Ionized Medium and the vertical structure of the ionized content in spiral arms, initial results include the discovery of several faint, extended ($d > 1\\deg$) H {\\sc ii} regions and the first map of the ionized component of an intermediate velocity cloud. ",
        "introduction": "\\label{sec:intro}  Past studies have shown the Warm Ionized Medium (WIM) to be a significant component of the ISM, especially in the halos of disk galaxies (Reynolds 1993; Hoopes, Walterbos, \\& Rand 1999; Rand 1996; Rossa \\& Dettmar 2000; Dettmar 1998). In the Milky Way, the WIM has a mass surface density about one-third that of H I, an ionization power requirement equal to the supernovae rate, and a characteristic scale height of 1 kpc (Reynolds 1993; Haffner, Reynolds, \\& Tufte 1999). Although the details of how the WIM is ionized and heated are not well understood yet, early-type stars seem to be the biggest contenders for providing the bulk of the ionizing photons. Dove, Shull, \\& Ferrara (2000); Dove \\& Shull (1994); and Miller \\& Cox (1993) have studied how O-star radiation can leak from the plane to ionize the large expanse occupied by the WIM. However, we will show below that B-stars may also be a significant contributer to the halo ionization field. The problem of Lyman continuum propagation through neutral gas is somewhat easier if these more widely distributed stars can contribute.  Aided by modern detector technology, several groups have initiated intensive programs to characterize the global details of the ionized content of galaxies (also called Diffuse Ionized Gas---DIG). In addition to the velocity-resolved all-sky survey described here, three large-area imaging surveys are also in progress. A southern survey with arc-minute resolution has been completed by Gaustad, McCullough, \\& Van Buren (1996; McCullough, these proceedings). Dennison, Simonetti, \\& Topasna (1998) have been imaging the northern sky with similar resolution. A high-resolution ($\\sim 1\\arcsec$) survey of the Galactic plane and Magellanic clouds is also underway by Parker \\& Phillips (1998).  Here we describe the results of the Wisconsin H-Alpha Mapper (WHAM) Northern Sky Survey, the first deep ($I_{\\rm H\\alpha} \\sim 0.1$ R; 1 R = $10^6/4\\pi$ photons cm$^{-2}$ s$^{-1}$ ster$^{-1}$), velocity-resolved survey of the WIM in our Galaxy. ",
        "conclusions": "\\label{sec:summary}  WHAM has completed a full spectral survey of the northern sky ($\\delta > -30\\deg$) in H$\\alpha$. Similar to the neutral component, filamentary substructure abounds in the ionized halo of the Galaxy. Exciting results from the faint H$\\alpha$ sky are abundant and we have presented two specific examples here. We have presented the first ionized map of an HVC or IVC. Complex K contains an ionized component with $I_{H\\alpha} \\approx 0.1$ -- $0.2$ R that follows the shape of the neutral gas extremely well. We have also presented several new, faint H\\ {\\sc ii} regions, many of which appear to be ionized by early B stars. In addition, one of these regions is very likely an example of an isolated H\\ {\\sc i} cloud being ionized by a nearby sdO star."
    },
    "0010/astro-ph0010574_arXiv.txt": {
        "abstract": "Dynamical instability is shown to occur in differentially rotating polytropes with $N = 3.33$ and $T/|W| \\gtrsim 0.14$.  This instability has a strong $m=1$ mode, although the $m=2, 3,$ and $4$ modes also appear.  Such instability may allow a centrifugally-hung core to begin collapsing to neutron star densities on a dynamical timescale. The gravitational radiation emitted by such unstable cores may be detectable with advanced ground-based detectors, such as LIGO II. If the instability occurs in a supermassive star, it may produce gravitational radiation detectable by the space-based detector LISA. ",
        "introduction": "Rotational instabilities are potentially important in the evolution of massive stellar cores.  For example, the core of a massive star that has been prevented from collapsing to neutron star densities by centrifugal forces may rotate rapidly enough for rotational instabilities to develop \\citep{HEH98,HEH99}. Massive cores spun up by accretion from a binary companion \\citep{wagoner84} and the remnants of compact binary mergers \\citep{RS92,ZCM94} may also reach fast enough rotation rates to become unstable. Rotational instabilities may produce detectable gravitational radiation \\citep{schutz89,thorne96}.  If enough angular momentum is shed, full collapse to a neutron star or black hole may occur.  We focus here on global rotational instabilities that arise in fluids from growing azimuthal modes $e^{i m \\phi}$ \\citep{tassoul}.  Dynamical instabilities are driven by hydrodynamics and gravity, and grow on the order of the dynamical timescales of the system. In objects with Maclaurin-like rotation laws,  dynamical rotational instabilities are generally expected to arise at fairly high values of $\\beta \\equiv T/|W| \\gtrsim 0.27$ in Newtonian gravity (see New, Centrella, \\& Tohline and references therein), and $\\beta  \\gtrsim 0.25 - 0.26$ in general relativity \\citep{SBS00}.  Here, $T$ is the rotational kinetic energy, and $W$ is the gravitational potential energy. Dynamical instability may set in at lower values of $\\beta$ for thick, self-gravitating disks \\citep{PDD,WTH} and in tori \\citep{TH}.  Secular instabilities develop on longer timescales, and arise from dissipative processes such as viscosity and gravitational radiation reaction.  They are expected to set in for relatively low values of $\\beta \\gtrsim 0.14$, although the rotation law, polytropic index, and general relativity can affect the value of the instability limit \\citep{ITDPY95,SF98}.  Studies of scenarios for the formation of centrifugally-hung cores have generally concluded that these objects will have $\\beta < 0.27$ \\citep{tohline84, EM85, ME85}. Direct numerical simulations of axisymmetric stellar collapse by \\citet{ZM97} also indicate that, for the cases studied, it is difficult for collapsing cores to reach and exceed $\\beta \\sim 0.27$.   Overall, it is generally assumed that centrifugally-hung cores will evolve secularly, as in the ``fizzler'' scenario \\citep{HEH98,HEH99}.  In this {\\it Letter}, we report on new 3-D numerical simulations of rotating stellar cores that exhibit dynamical rotational instability at relatively low values of $\\beta \\gtrsim 0.14$. These models are computed using Newtonian gravity with no backreaction; this is a reasonable approximation for a stellar core of mass $M \\sim 1.4 M_{\\odot}$ hung up at a radius $R \\gtrsim 100$km, so that $(GM/Rc^2) \\lesssim 0.02$. Our results show that compact objects can become dynamically unstable in a region of parameter space previously thought to be solely the province of secular instabilities. ",
        "conclusions": "Our results demonstrate that dynamical instability can occur in differentially rotating polytropes with low values of $\\beta \\gtrsim 0.14$, and that this instability has a strong, $m=1$ character.  Figure~\\ref{init-mods} shows that the unstable axisymmetric equilibria each have a torus of dense material centered on the rotation axis within the model.  This suggests that this instability may be related to the ones found by \\citet{WTH} and \\citet{TH} in their studies of polytropic tori with $N=3/2$. We note that \\citet{PDD} also found a dominant $m=1$ instability that set in at a relatively low $\\beta \\gtrsim 0.20$, in a centrally condensed model (without an off-center density maximum). Their model was an $N=3/2$ polytrope with an $n'=2$ rotation law.  The $n'=2$ rotation law produces configurations with a strong concentration of angular momentum in the outer regions of the model. We plan to carry out more detailed studies to investigate the character of the unstable modes seen in our simulations and their properties for various values of $N$ and the $j$-constant rotation law parameter $d$.  If such instability occurs in a centrifugally-hung core, collapse to neutron star densities may result.  Our simulations do show that the density is increasing at the end of the unstable runs.  Further studies are needed to see how dense the remnant actually becomes, and whether the $m=1$ mode will result in the dense region moving at a velocity comparable to those of actual neutron stars \\citep{popov}.  Dynamical instability will also produce gravitational radiation. Although longer runs are needed to obtain the full gravitational waveforms, we can estimate the properties of the gravitational wave emission from the initial stages of the instability.  For a massive stellar core with $M \\sim 1.4 M_{\\odot}$ and $R_{\\rm E} \\sim 200$km, the peak amplitude will occur at a frequency of roughly $f \\sim 200$Hz.  This peak amplitude will be $h \\sim 10^{-24} r_{20}^{-1}$ for $\\beta = 0.14$, and $h \\sim 10^{-23} r_{20}^{-1}$ for $\\beta = 0.18$.  Here, $r_{20}$ is the distance to the source in units of 20 Mpc. Emission from such unstable cores may be detectable with advanced ground-based interferometers, such as LIGO-II. An even more optimistic scenario for detectable gravitational radiation occurs if these instabilities arise in supermassive stars\\footnote{ We note that the appropriate equation of state for supermassive stars is the $n=3$ polytropic equation of state.} \\citep{NS}. For example, if the instability occurs when a supermassive star contracts to the point that $(GM/Rc^2) \\sim 1/15$, an approximate value for uniformly rotating stars (Baumgarte \\& Shapiro 1999), we estimate the frequency of the gravitational radiation to be $f \\sim 3.5 \\times 10^{-3}$Hz, with amplitude $h \\sim 10^{-18}r_{20}^{-1}$ for $\\beta = 0.14$ and $h \\sim 10^{-17}r_{20}^{-1}$ for $\\beta = 0.18$. Such signals would be easily detectable by the space-based LISA detector."
    },
    "0010/astro-ph0010432_arXiv.txt": {
        "abstract": "We use the semi-analytical approach to analyze gravitational lensing of remote quasars by foreground dark halos in various cold dark matter (CDM) cosmologies, in order to determine the sensitivity of the predictions for probabilities of images separations to the input assumptions regarding the properties of halos and cosmological models. The power spectrum of primordial fluctuations is normalized by the cluster abundance constraints. The mass function of dark halos is assumed to be given by the Press-Schechter function. The mass density profile of dark halos is alternatively taken to be the singular isothermal sphere (SIS), the Navarro-Frenk-White (NFW) profile, or the generalized NFW profile. The cosmologies being considered include: the Einstein-de Sitter model (SCDM), the open model (OCDM), and the flat $\\Lambda$-model (LCDM). As expected, we find that the lensing probability is extremely sensitive to the mass density profile of lenses (dark halos), and somewhat less so to the mean mass density in the universe, and the amplitude of primordial fluctuations. NFW halos are very much less effective in producing multiple images than SIS halos. For NFW lenses, the SCDM model produces lensing events fewer than the OCDM/LCDM models by two orders of magnitude. For SIS lenses, the SCDM model produces more lensing events with small splitting angles but produces fewer lensing events with large splitting angles than the OCDM/LCDM models, which is due to the fact that for large mass halos the Press-Schechter function is very sensitive to the amplitude of primordial fluctuations. In all cases the difference between the OCDM model and the LCDM model is not dramatic. None of these models can completely explain the current observations: the SIS models predict too many large splitting lenses, while the NFW models predict too few small splitting lenses. Essentially, the observed high ratio of small splitting to large splitting lenses is not predicted correctly. This indicates that there must be at least two populations of halos in the universe: small mass halos with a steep inner density slope and large mass halos with a shallow inner density slope. A combination of SIS and NFW halos can reasonably reproduce the current observations, if we choose the mass for the transition from SIS to NFW to be $\\sim 10^{13}$ solar masses as might plausibly occur due to baryonic cooling and contraction in lower mass systems. Additionally, there is a tendency for CDM models to have too much power on small scales, i.e. too much mass concentration. From our sensitivity studies it appears that the cures proposed for other apparent difficulties of CDM would help here as well, an example being the warm dark matter (WDM) variant which is shown to produce large splitting lenses fewer than the corresponding CDM model by one order of magnitude. ",
        "introduction": "Gravitational lensing directly probes the mass distribution in the universe, so the investigation of lensing events of quasars at high redshifts can provide us with important information about cosmology and the matter distribution in the universe. A number of groups have attempted to test cosmological models by comparing lensing probabilities predicted by various cosmological models with those obtained from observations \\citep{tur84,nar88,cen94,wam95,mao97,wam98,bar98,mor00,por00,kee01}. In the statistical study of cosmological gravitational lensing, two different approaches are used: one is to study the lensing probability of splitting angles of multiple images \\citep{tur84,nar88,cen94,koc95,wam95,wam98,por00,kee01}, the other is to study the lensing probability of the length-to-width ratio of arcs formed by gravitational lensing \\citep{bar94,bar98,coo99,kau00,men00}. The two approaches reflect two different aspects of gravitational lensing and are complimentary to each other, both of them are directly related to observations and deserve detailed investigations. In this paper we focus on the splitting angles of multiple images caused by gravitational lensing. The case of length-to-width ratio of arcs will be considered in another paper. Results of the work done in testing cosmological scenarios to date are inconclusive with no clear preference found for any of the current models, and, overall, one could conclude that none of the studied models provides a good fit to the rapidly improving observations. Perhaps the reason for this indeterminacy is that lensing also probes small scale structure. While this is known, its importance has perhaps not been appreciated sufficiently. Gravitational lensing provides an exquisitely sensitive test to the high frequency part of the power spectrum, because it is this part which establishes the central profiles of dark matter structures. Since lensing is comparably sensitive to input assumptions concerning the properties of lenses and of cosmological models, it is hazardous to use observed lensing statistics to draw inferences with regard to cosmology before determining the sensitivity to other factors. The primary purpose of this paper is to quantitatively assess the sensitivity of lensing expectations to the various input parameters so that this tool can be used to greatest effect.  To determine the probability for gravitational lensing, we need to know the mass density profile of lenses (dark halos), the mass function of lenses (which gives the number distribution of lenses over masses), the structure and global features of the universe (the universe is flat, closed, or open; the matter contents in the universe, etc), and the positions and shapes of the source objects (quasars). For studying the splitting angles of multiple images, it is enough to assume the source objects are points at high redshifts. In the framework of the Friedmann-Robertson-Walker models, there are a number of cosmologies as candidates for describing our universe, among which the most popular ones are the Einstein-de Sitter universe, which assumes a zero curvature and a zero cosmological constant; the open universe, which assumes a negative curvature and a zero cosmological constant; and the flat $\\Lambda$-universe, which assumes a zero curvature and a positive cosmological constant \\citep{pee93,ost95}. The matter contents in the universe are usually assumed to be dominated by dark matter: cold dark matter, warm or hot dark matter, or the mixture of them \\citep{kol90,pee93}. Different kinds of dark matter give rise to different mass density profiles for dark halos. If the matter contents in the universe are dominated by cold dark matter (CDM), the three cosmological models are denoted respectively as SCDM (standard CDM = Einstein-de Sitter + CDM), OCDM (open + CDM), and LCDM (cosmological constant $\\Lambda$ + CDM).  The primordial fluctuations determine the mass distribution and the number density of dark halos in the universe. In studying gravitational lensing, the cosmological models, the matter contents in them, the power spectrum of the primordial fluctuations, and the source object positions are pre-assumed. Then the mass density fluctuations evolve according to the Einstein equations, which can be traced with numerical simulations. As time goes on, gravitationally bound objects (dark halos) are formed, which play the role of lenses. The mass density profile, and the mass function of dark halos in the universe are automatically derived from numerical simulations. Then, tracing the trajectories of light rays, we can obtain the number of lensing events and thus the lensing probability. This is the basic spirit of numerical simulations for gravitational lensing \\citep{wam95}. With this approach, the mass density profile of lenses and the mass function of lenses appear as the results of simulations, they do not need to be pre-assumed. However, numerical simulations have their own limitations. First, numerical simulations usually take a great deal of computer time, which makes it inconvenient to test several different models at a time. Second, in every numerical simulation there are two limits on the spatial scales: the size of box, which is the maximum scale and determines the presence or absence of intrinsically rare events, and the size of fundamental cell (i.e. the resolution), which is the minimum scale. Correspondingly this gives rise the upper and lower bonds on the masses of halos formed in the simulation, which causes the result that the number of lensing events with small splitting angles (caused by halos with small masses) and  large splitting angles (caused by rare halos with large masses) are underestimated. This last deficit is seen by comparing the simulation results of \\citet{wam95} with the analytical results of \\citet{koc95}, which will also be confirmed by the results in this paper. Furthermore, the lensing cross-sections are very sensitive to inner most parts of the density profiles which are most subject to problems of numerical resolution \\citep{ghi00,kly00}. Ultimately, numerical simulations are required to best determine the expected halo properties. The current generation of numerical simulations has insufficient dynamical range to simultaneously address accurately both large and small scale structures, but forthcoming $1024^3$-particle simulations may remedy this defect. For the present, semi-analytical methods may be best adapted to explore the sensitivity of the results to the input parameters.  In the semi-analytical approach, the mass density profile and the mass function of lenses are pre-assumed and can be adjusted to fit various dark matter scenarios. Though this makes us rely on an assumed density profile and mass function, it has the advantage that the calculation of the lensing probability can be handled with semi-analytical methods, which takes much less computer time than numerical simulations, so several different models can easily be tested at a time. And, in this approach, there are not the limitations in spatial scales and mass scales necessitated by numerical simulations; the lensing probability for small splitting angles and large splitting angles can be obtained accurately for any assumed halo density profile, the most critical feature of which is the steepness of the central density profile which may be parameterized by the central slope, $-\\alpha$. The mass density profile is often taken to be the singular isothermal sphere (SIS, $\\alpha = 2$) \\citep{got74}; the mass function is usually taken to be the Press-Schechter function \\citep{pre74}. Though the SIS profile is simple, it does not fit the CDM simulation results well. \\citet{nav96,nav97} have proposed a ``universal'' density profile (the so-called NFW profile, $\\alpha=1$) which has been shown to fit the simulation results better. A generalized NFW (GNFW) profile with somewhat steeper inner slopes intermediate between NFW and SIS appears to provide a still better fit to simulated profiles \\citep{zha96,sub00,ghi00,jin00,wyi00} for which typically $\\alpha = 1.4 \\pm 0.1$. With N-body simulations of the SCDM scenario \\citep{efs90,kau93}, or comparison with observations \\citep{bah93,gir98,gir00}, the Press-Schechter mass function has been shown to describe the mass distribution in the universe quite accurately.  In this paper we use the semi-analytical approach to investigate the gravitational lensing of quasars by foreground dark halos. We work with three kinds of cosmological models: SCDM, OCDM, and LCDM models. The mass function of dark halos is taken to be the Press-Schechter function. The power spectrum of primordial fluctuations is normalized by the cluster abundance constraints \\citep{wan98,wan00}. We alternatively consider three kinds of density profiles: SIS, NFW, and GNFW profiles, where the last model allows us to estimate (by varying the free parameters in the GNFW profile) variants of the CDM scenario. For the various models we will calculate the lensing probability of splitting angles of multiple images, and test the sensitivity of the lensing probability to the input parameters regarding the properties of halos and cosmological models.  The paper is organized as follows: In section \\ref{section2} we outline the cosmological models to be calculated in the paper, and introduce the power spectrum of primordial fluctuations and the Press-Schechter function. In section \\ref{section3} the most simple case is investigated -- SIS halos as lenses. We find that, the SCDM model produces fewer lensing events with large splitting angles, but more lensing events with small splitting angles than the OCDM model and the LCDM model. This is due to the fact that for large splitting angles, which are produced by large mass halos, the Press-Schechter function is exponentially sensitive to the amplitude of primordial fluctuations and the SCDM model has the smallest amplitude of primordial fluctuations; but for small splitting angles, which are produced by small mass halos, the Press-Schechter function is not sensitive to the amplitude of primordial fluctuations but the mean mass density in the universe plays a more dominant role. In section \\ref{section4} we consider the case for the NFW profile and the GNFW profile. As expected, we find that, the lensing probability is very sensitive to the mass density profile of the lenses. The more concentrated toward center the density distribution is, the higher lensing probability it produces. Compared to a SIS halo, a NFW halo is very ineffective in producing multiple images. For the NFW profile, the SCDM model produces the least lensing events: the OCDM model and the LCDM model produce a comparable number of lensing events which exceed the number of lensing events produced by the SCDM model by two orders of magnitudes. This is mainly due to two facts: a NFW model in a SCDM universe has the smallest concentration parameter, and a SCDM universe has the smallest amplitude of fluctuations. In section \\ref{section5}, we compare our predictions with observations. We find that, none of the models can explain the current observations: the SIS models predict too many lensing events with large splitting angles, while the NFW models predict too few lensing events with small splitting angles. Essentially, the observed high ratio of small splitting to large splitting events is not predicted correctly. By a combination of them -- using the SIS profile for small mass halos and the NFW profile for large mass halos, we show that the observation results can be reasonably re-produced if we choose the mass at the transition point between the two halo populations to be $\\sim 10^{13}$ solar masses -- the mass of halos below which cooling of the corresponding baryonic component will lead to concentration of the baryons to the inner parts of the mass profile \\citep{ree77,blu86,por00}. Our results indicate that the halos in the universe cannot be described by a single universal density profile, there exist at least two halo populations: small mass halos with a steep inner density slope (the inner parts dominated by the baryonic component), and large mass halos with a shallow inner density slope (within which baryons and dark matter have essentially the same profiles). In section \\ref{section6}, we draw our conclusions.  The failure of the SIS, and GNFW ($\\alpha = 1.5$) models at large splitting angles appears to be another manifestation of the tendency for CDM models to have too much power on small scales, i.e. too much mass concentration. From our sensitivity studies it appears that the cures proposed for other difficulties of CDM would help here as well, an example being the warm dark matter (WDM) variant, for which $\\alpha$ tends to be smaller \\citep{bod00}. A model very roughly designed to match a WDM variant has a greatly reduced probability of large splitting angles: lower than the corresponding probability for the CDM model by one order of magnitude. ",
        "conclusions": "\\label{section6} With the semi-analytical approach we have calculated the probability for forming multiple images of a remote source object gravitationally lensed by foreground dark halos. The mass density profile of a halo is alternatively taken to be described by the SIS profile, the NFW profile, and the GNFW profile. The mass function of halos is assumed to be given by the Press-Schechter function. The cosmological model is alternatively the SCDM model, the OCDM model, and the LCDM model. Our results show that the lensing probability is very sensitive to the density profile of lenses (dark halos). The more the mass distribution concentrates toward the center, the higher lensing probability the density profile gives rise, which is clearly seen in Fig. \\ref{fig7} -- Fig. \\ref{fig10}. Compared to SIS lenses, NFW lenses are extremely inefficient in producing multiple images. For example, the lensing probability $P(>5^{\\prime \\prime})$ for NFW halos is lower than the corresponding probability for SIS halos by more than two orders of magnitudes. For GNFW halos, as $\\alpha$ decreases, the differences among different cosmological models become prominent (Fig. \\ref{fig8}), especially the difference between the SCDM model and the OCDM/LCDM model: when $\\alpha = 1$ (i.e. for NFW halos) the lensing probability for the SCDM model is lower than the lensing probability for the OCDM/LCDM model by more than two orders of magnitudes. This dramatic effect is dominantly caused by the fact that halos in a SCDM universe have the smallest concentration parameter \\citep{bar98} and for smaller $\\alpha$ the lensing probability is more sensitive to the concentration parameter $c_1$: for the NFW case the lensing probability is extremely sensitive to $c_1$, especially for small $c_1$ (Fig. \\ref{fig9}). Though not so dramatic as the dependence on the inner slope, the lensing probability also shows dependence on the outer slope of halos. Comparing the right panel of Fig. \\ref{fig8} with Fig. \\ref{fig3}, the lensing probability produced by GNFW halos with $\\alpha = 2$ is somewhat higher than the lensing probability produced by SIS halos. This mild difference is attributed to the different slopes of the $\\alpha = 2$, GNFW profile and the SIS profile at large radii (the $\\alpha = 2$, GNFW profile and the SIS profile have the same slope at small radii). Similar conclusions are obtained by \\citet{wyi00} and \\citet{kee01}. \\citet{wyi00} find that the optical depth to multiple imaging is a very sensitive function of the profile parameters and the GNFW profile exhibits degeneracies between profile parameters with respect to lensing statistics. \\citet{kee01} demonstrate that the lensing probability is determined almost entirely by the fraction of the halo mass that is contained within a fiducial radius that is $\\sim 4\\%$ of the virial radius.  Our results also show the dependence of the lensing probability on the cosmological parameters, which is particularly manifested in the SIS case. In Fig. \\ref{fig3}, for small splitting angles the lensing probability in the SCDM universe is higher than the lensing probability in the OCDM/LCDM universe, while for large splitting angles the lensing probability in the SCDM universe is much lower than the lensing probability in the OCDM/LCDM universe. This is explained by the fact that the Press-Schechter function is sensitive to $\\sigma_8$ only for large mass halos (Fig. \\ref{fig1}). Since small splitting angles are produced by small mass halos, and for small mass halos the Press-Schechter function is not sensitive to $\\sigma_8$, the lensing probability is dominantly determined by the cosmic mean mass density $\\Omega_m$; thus for small splitting angles the lensing probability is highest for SCDM since the SCDM universe has the highest mean mass density. Since large splitting angles are produced by large mass halos and for large mass halos the Press-Schechter function is (exponentially) sensitive to $\\sigma_8$, the lensing probability is dominantly determined by $\\sigma_8$, thus for large splitting angles the lensing probability is lowest for SCDM since the SCDM universe has the smallest $\\sigma_8$. For the NFW case, in Fig. \\ref{fig6} we see that to produce the same small splitting angle the required NFW halo mass is significantly larger than the required SIS halo mass. Thus, for the NFW case $\\sigma_8$ takes effect for all $\\Delta\\theta\\ge 1^{\\prime\\prime}$, which together with the concentration parameter $c_1$ overtakes the effect of $\\Omega_m$ (see Fig. \\ref{fig8}). The studies of cluster abundances proposed a correlation between $\\Omega_m$ and $\\sigma_8$ as given by equation (\\ref{s8w}). We have also tested the sensitivity of the lensing probability to $\\Omega_m$ when equation (\\ref{s8w}) is satisfied. We see that in general the lensing probability varies with $\\Omega_m$; and for the NFW case the lensing probability is very sensitive to $\\Omega_m$ (Fig. \\ref{fig10}).  Our numerical results are summarized in Table \\ref{tab1}, where we have not included the effect of magnification bias (i.e. $P$ is the intrinsic probability), the source object is assumed to be at $z_s=1.5$, the Hubble constant is assumed to be $h=0.7$. To show the sensitivity of the lensing probability to the parameters $\\sigma_8$, $\\Omega_m$, $\\alpha$, and $c_1$, we have evaluated the differentiation of lensing probability $P(>\\Delta\\theta)$ with respect to $\\sigma_8$, $\\Omega_m$, $\\alpha$, and $c_1$ for the LCDM model at $\\sigma_8 = 0.95$, $\\Omega_m = 0.3$, $\\Omega_{\\Lambda} = 0.7$, $\\alpha = 1.5$, and $c_1(z=0) = 4$ for $\\Delta\\theta =5^{\\prime\\prime}$ and $\\Delta\\theta= 10^{\\prime\\prime}$ respectively. The results are \\begin{eqnarray} {\\delta P\\over P} &=& 4.9 \\left({\\delta \\sigma_8\\over \\sigma_8} \\right)_{\\Omega_m,\\alpha,c_1} + 1.7\\, \\left({\\delta \\Omega_m\\over \\Omega_m}\\right)_{\\sigma_8,\\alpha,c_1} +11 \\left({\\delta \\alpha \\over \\alpha}\\right)_{\\sigma_8,\\Omega_m,c_1}\\nonumber\\\\ &&+3.1\\left({\\delta c_1\\over c_1}\\right)_{\\sigma_8, \\Omega_m,\\alpha}\\,, \\label{diff1} \\end{eqnarray} for $\\Delta\\theta=5^{\\prime\\prime}$, and \\begin{eqnarray} {\\delta P\\over P} &=& 5.8 \\left({\\delta \\sigma_8\\over \\sigma_8} \\right)_{\\Omega_m,\\alpha,c_1} + 1.8\\, \\left({\\delta \\Omega_m\\over \\Omega_m}\\right)_{\\sigma_8,\\alpha,c_1} +10 \\left({\\delta \\alpha \\over \\alpha}\\right)_{\\sigma_8,\\Omega_m,c_1}\\nonumber\\\\ &&+3.4\\left({\\delta c_1\\over c_1}\\right)_{\\sigma_8,\\Omega_m, \\alpha}\\,, \\label{diff2} \\end{eqnarray} for $\\Delta\\theta=10^{\\prime\\prime}$. Note the extreme sensitivity of the lensing probability to the slope of the inner profile: $(d\\ln P/ d\\ln \\alpha)_{\\sigma_8,\\Omega_m, c_1} \\approx 10$. Note also that, if $\\Omega_m$ and $\\sigma_8$ enter the lensing probability as a combination given by equation (\\ref{s8w}), then we should expect $(d\\ln \\sigma_8/ d\\ln \\Omega_m)_{P,\\alpha, c_1} \\approx \\gamma \\approx 0.53$, but equation (\\ref{diff1}) and equation (\\ref{diff2}) show rather smaller values of $(d\\ln \\sigma_8/ d\\ln\\Omega_m)_{P,\\alpha, c_1}$ ($0.34$ and $0.31$ respectively). Thus, in lensing statistics the degeneracy between $\\Omega_m$ and $\\sigma_8$ in equation (\\ref{s8w}) is broken. This is important since it allows us to separately determine the values of $\\Omega_m$ and $\\sigma_8$ in principle.  With the magnification bias being considered, none of the simple models can completely explain the JVAS/CLASS observations. The SIS models produce too many large splitting lenses, but the JVAS/CLASS observations have detected no lenses with $\\Delta\\theta > 3^{\\prime\\prime}$ \\citep{hel00,phi00}. The null result for detecting large splitting lenses are true not only for JVAS/CLASS \\citep{phi00}, but also for ARCS (Arcminute Radio Cluster-lens Search) which is aimed at looking for gravitational lensing events with images separation between $15^{\\prime\\prime}$ and $60^{\\prime\\prime}$ \\citep{phi00a}. While the NFW models produce very rare large splitting lenses which is consistent with observations, they produce too few small splitting lenses which is against observations since in the JVAS/CLASS survey at least $17$ lenses with $0.3^{\\prime\\prime}<\\Delta\\theta<3^{\\prime\\prime}$ have been discovered \\citep{hel00}. None of the models can explain the observable large ratio of the number of small splittings to the number of large splittings. This strongly suggests that there are at least two populations of halos in the universe: small mass halos with a steep inner density slope, and large mass halos with a shallow inner slope. We have constructed a very simple two-population halo model to test this conjecture: the mass density of halos with mass $< 10^{13}\\,h^{-1} M_{\\odot}$ is given by the SIS profile, while the mass density of halos with mass $> 10^{13}\\,h^{-1}M_{\\odot}$ is given by the NFW profile. We find that the results for this model are reasonably consistent with the JVAS/CLASS observations. In particular, the number of lenses with $\\Delta \\theta>0.3^{\\prime\\prime}$ predicted by the LCDM model is $\\approx 16$, which is remarkably close to the number of lenses observed by the JVAS/CLASS survey which is $17$. The SIS model predicts too many lenses ($\\approx 37$), while the OCDM model predicts somewhat too few lenses ($\\approx 12$). A similar compound model has also been considered by \\citet{por00} to explain the results from CASTLE (CfA-Arizona Space Telescope Lens) survey.  In summary, with the semi-analytical approach we have shown that the gravitational lensing probability is very sensitive to the mass density profile of lenses (especially in the central region), the mean mass density in the universe, and the amplitude of primordial fluctuations. Compared with the observation results of JVAS/CLASS, our calculations indicate that the halos in the real universe cannot be described by a single universal universal density profile, there are at least two populations of halos in the universe: small mass halos with a steep inner density slope and large mass halos with a shallow inner density slope. Ultimately, of course, very accurate address of the question is left open by this study."
    },
    "0010/astro-ph0010118_arXiv.txt": {
        "abstract": "We explore how the growth rate of spiral disks can be measured via analyses of the scatter in the Tully-Fisher (TF) relation of local and intermediate redshift galaxies. As an initial step, we show it is possible to construct a low-dispersion TF relation for nearly face-on, nearby spirals. We find these spiral disks are non-circular, (a mean ellipticity of 6\\%), which accounts for $\\sim$0.1 mag of the intrinsic scatter in the nearby TF relation of even the most ``normal'' looking spirals. If this ellipticity is induced by matter accretion, we expect to find greater disk ellipticity or disturbance in the past. We find evidence that more extreme outliers of the intermediate-redshift TF relation are more morphologically and kinematically disturbed. Whether this effect reflects a redshift trend or selection bias of local samples needs to be addressed.  \\vskip -0.25in ",
        "introduction": "\\vskip -0.1in  Dynamically cold, large galaxy disks seen today appear to have formed early or gradually; they are easily destroyed through strong interactions or mergers. For example, Wyse (these proceedings) argues the Milky Way has not suffered a major merger in the last 10 Gyr. The nearly constant co-moving number of large disks out to $z \\sim 1$ (e.g. Lilly et al. 1998) and the subtle changes with redshift in the scaling relations for large, field spirals corroborate a picture of early formation or slow, quiescent growth, but infrequent death.  The mass accretion rate onto disks is not known directly, however, at any epoch. Direct measurements of disk mass as a function of time, e.g., from measurement of disk scale-heights and stellar velocity dispersions, are desirable, but currently unavailable -- except for a handful of local systems. Photometric estimates of disk mass based on colors are uncertain and incomplete. Another approach is to use the evolution of the zeropoint and scatter in disk scaling-relations as a diagnostic of {\\it changes} in mass accretion rates. Here we focus on the correlations with, and cause of the scatter in the Tully-Fisher relation.  For gradual formation, we assume that discrete, but small accretion events perturb otherwise axisymmetric systems. Were we to obtain a complete census of galaxies at difference epochs, the relative number of perturbed systems should then reflect changes in the minor-merger rate. Zaritsky \\& Rix (1997) estimated the minor-merger rate for nearby galaxies by comparing the amplitude of photometric $m=1$ modes to $B-R$ and offsets in luminosity from a fiducial TF relation. Kinematic measures of disturbance would be preferred, however, since the long-term morphological perturbation may be subtle (Haynes 2000), and photometric effects may evolve with look back-time due to changing gas fractions and star-formation rates (Mihos, these proceedings). Here we give two examples of how spatially resolved kinematics can directly link disk asymmetries to scatter in the TF relation. This, in turn, points to a tractable method for measuring the growth rate of spiral disks.    \\begin{figure} \\plotfiddle{bershady1.ps}{2in}{0}{55}{55}{-210}{-75} \\vskip 0.15in \\caption{\\hsize 5.2in \\baselineskip 0.165in (a) The contribution to the TF scatter ($\\sigma_{TF}$) due to inclination errors produced by intrinsic disk ellipticity ($\\epsilon$), for two TF slopes (labeled). The mean, standard deviation and median ellipticity of our sample are indicated. (b) The $R$-band TF-relation for our nearly face-on sample, compared to Courteau's (1997) zeropoint and scatter (assuming $r-R=0.36$). The one outlier has an AGN and W$_{50}$/W$_{20}$ line-width ratios consistent with a dynamically disturbed system (Conselice et al. 2000).} \\vskip -0.15in \\end{figure}  \\vskip -0.35in \\null ",
        "conclusions": ""
    },
    "0010/astro-ph0010143_arXiv.txt": {
        "abstract": "We describe a simple algorithm for classifying orbits into orbit families.  This algorithm works by finding patterns in the sign changes of the principal coordinates.  Orbits in the logarithmic potential are studied as an application; we classify orbits into boxlet families, and examine the influence of the core radius on the set of stable orbit families. ",
        "introduction": "The dynamical structure of an elliptical galaxy may be represented by a collection of orbits (e.g.~Schwarzschild 1979).  In this representation the time-average of each orbit yields an invariant density distribution, and a weighted sum of such distributions equals the overall mass density of the galaxy.  Orbits are categorized into different families depending on their shapes; certain orbit families become stable or dominant in certain potentials.  The orbits available in a given potential may be characterized in several interrelated ways.  Classifying orbits into families says something about their geometry and thus about the contribution they can make to the mass distribution.  This paper describes a simple algorithm for classifying orbits in elliptical potentials, and illustrates its application to the well-studied logarithmic potential (see Miralda-Escud\\'{e} \\& Schwarzschild 1989, hereafter MES89).  The orbital content of triaxial potentials has long been of interest to dynamical astronomers.  Integrable potentials have fairly simple structures, with one family of `box' orbits and three of `tubes' (de~Zeeuw 1985).  In nonintegrable potentials the situation is much more complex; chaotic orbits may appear (Goodman \\& Schwarzschild 1981; Valluri \\& Merritt 1998), and the monolithic tube and box families are partly replaced by a plethora of minor orbit families (Binney \\& Spergel 1982; MES89; Lees \\& Schwarzschild 1992).  This rich structure is created by a pattern of resonances between fundamental orbital frequences.  In $d$ dimensions there are $d$ such frequencies $\\omega_i$, and an orbit family corresponds to a resonance condition of the form \\begin{equation} {\\bf n} \\cdot {\\bf \\bomega} = 0 \\, , \\end{equation} where ${\\bf n}$ is a vector of integers, and $|{\\bf n}| > 0$; thus each orbit family may be characterized by the corresponding vector ${\\bf n}$.  Several different approaches to orbit classification exist.  Spectral methods (Binney \\& Spergel 1982, 1984; Papaphilippou \\& Laskar 1996, 1998; Carpintero \\& Aguilar 1998; Valluri \\& Merritt 1998) classify orbits by measuring their fundamental frequencies $\\omega_i$.  These methods are powerful and general but also somewhat complex to implement.  Alternately, orbits may be classified by heuristic methods based on known characteristics of specific orbit families.  For example, tube orbits maintain a definite sense of circulation about a principal axis (e.g.~Barnes 1992).  Likewise, the `boxlet' orbits found in nonintegrable potentials may be classified by measuring extrema along the principal coordinates (e.g.~Schwarzschild 1993).  We have developed a classification algorithm that captures the geometry of both loop and boxlet orbits in a simple unified scheme. This algorithm is easily implemented and generalized to find higher-order resonances.  In its present form the algorithm is best suited for two-dimensional potentials; while it provides a partial characterization of orbits in three-dimensional potentials, it cannot recognise the full range of orbit families present in such systems. Nonetheless, the generality and simplicity of our method may make it attractive when a full-blown spectral treatment is unnecessary or prohibitively expensive.  In this paper we describe the algorithm and present some sample results.  Section~2 develops the theoretical basis for the algorithm and discuss some aspects of its implementation.  Section~3 illustrates the use of this algorithm in exploring the orbits in the logarithmic potential.  Conclusions and further applications of the algorithm are discussed in Section 4. ",
        "conclusions": "Our algorithm works by reducing orbital trajectories to sequences of symbols, and matching these sequences against a set of known patterns. The idea of describing trajectories symbolically is not new; it has been applied to geodesics on negatively-curved surfaces, to forced van der Pol oscillators, and to the restricted three-body problem (Moser 1973 and references therein).  Symbolic representation of orbits may have other applications in galactic dynamics.  Below we briefly discuss a few related issues and some possible future developments.  \\subsection{Start spaces}  The reader may feel somewhat reassured to learn that the three 1-D start spaces considered above are probably both necessary and {\\it sufficient\\/} to find all boxlet families.  We can show this by using the properties of the parametric curves (3).  First, the orbits which have stationary points are those for which $k = 0$.  Second, centrophilic orbits are those for which the quantity $\\ell + m + k$ is an even number.  Third, orbits which cross the $y$ axis at an angle of $90^\\circ$ are those for which $\\ell + 2m + k$ is even.  Now if an orbit has no stationary point and is centrophobic then it has $k = 1$ and $\\ell + m + k$ is odd, and because $\\ell$ and $m$ are relatively prime it follows that $m$ is odd and $\\ell + 2m + k$ is even -- so the orbit crosses the $y$ axis at $90^\\circ$.  Thus {\\it any\\/} orbit must fall into at least one of these three categories.  Orbits with $k = 0$ can be accessed from the stationary start space, orbits with even $\\ell + m + k$ are reachable using the central start space, and orbits with even $\\ell + 2m + k$ result from the perpendicular start space. With a few exceptions due to misclassification of high-order resonances, these rules accurately predict which orbit families appear in each of Figs.~2, 3, and~4.  The stationary and perpendicular start spaces suffice for the singular logarithmic potential ($R_{\\rm c} = 0$) since all centrophilic orbits are unstable in this potential (e.g.~Kuijken 1993).  These results have some bearing on orbit surveys in three-dimensional potentials.  At a given energy, the manifold of possible orbits can be covered by a 4-D start space constructed by analogy with the 2-D start space described in Section~3.1 (Levison \\& Richstone 1987).  Since this start space is awkward to sample and to display, many workers favor 2-D start spaces analogous to the stationary and perpendicular start spaces described above (Schwarzschild 1993).  But these 2-D start spaces don't allow access to centrophilic orbits which lack a stationary point, and these orbits may be stable in nonsingular potentials.  Thus a full survey of orbits in a three-dimensional nonsingular potential requires {\\it at least\\/} three different 2-D start spaces: the stationary and perpendicular start spaces must be supplemented by a central start space (Merritt 1999).  It still remains to be shown that these 2-D spaces really are sufficient to access all orbits.  \\subsection{Noninteger $k$?}  We've assumed, following MES89, that the {\\it only\\/} allowed values of the phase parameter are $k = 0$ and $k = 1$.  This assumption is critically important for the discussion in the previous subsection. However, D.~Merritt (private communication) has pointed out that $k$ may have values between $0$ and $1$; for example, in a 2-D harmonic oscillator the $x$ and $y$ motions are independent and neutrally stable orbits with {\\it any\\/} real value of $k$ exist.  For nonintegrable potentials the situation is unclear; $k$ is probably limited to a discrete set of values, but it's possible that values between $0$ and $1$ are permitted.  Our algorithm can't distinguish a resonant orbits with noninteger $k$ from a centrophobic orbit with the same $(\\ell,m)$, since both yield the same sign-change pattern.  Such orbits would be excluded from the 1-D start spaces used in Section~3.2, though in principle they are accessible from a 2-D start space.  Visual inspection may be the easiest way to discover orbits with noninteger $k$; such orbits, while following the same sign-change patterns as `normal' centrophobes, would have different symmetry properties.  No such orbits have been seen in a cursory inspection of candidates, but more work is needed.  \\subsection{Orbit families}  The results presented in Section~3 emphasize the remarkable variety and structure of orbit families in the logarithmic potential.  Of the $46$ families listed in Table~1, some $29$ can be found in Figs.~2, 3, and~4.  Almost half of these families are centrophilic.  While centrophilic orbits have been described before (MES89; Lees \\& Schwarzschild 1992; Barnes 1998, Fig.~42), most earlier studies treated these families as isolated curiosities.  MES89's bifurcation diagrams showed that boxlets and antiboxlets exchange parental roles as $R_{\\rm c}$ changes; we have made the consequences of this role-reversal visible by routinely finding the centrophilic counterparts of unstable centrophobic orbits.  Centrophilic orbits may be destabilized by central mass concentrations in real galaxies, but their recognition still helps in understanding the mechanisms which determine the orbital content of a potential.  We note that some families did not appear in any of our classification diagrams.  Most of these missing families are fairly high-order resonances which may be difficult to find because they occupy very small regions of the start spaces.  Two low-order centrophilic families, however, are conspicuously absent.  One is the orbit generated by the $(\\ell,m) = (1,-1)$ resonance with $k = 0$, and the other is the antibanana, generated by the $(2,-1)$ resonance with $k = 1$.  The former resonance is stable in some anharmonic potentials (de Zeeuw \\& Merritt 1983), while the latter may occur in highly flattened logarithmic potentials (Papaphilippou \\& Laskar 1996, Fig.~10).  We would not expect these centrophilic orbits to appear in our plots since their centrophobic counterparts, the loop and banana orbits, do appear and are stable for a wide range of parameters.  \\subsection{3-D orbits}  It's somewhat disappointing to admit that the technique described here doesn't trivially generalize to three-dimensional potentials.  The reason for this failure is fairly straightforward.  In two dimensions the resonance condition (1) fixes a unique relation between the fundamental frequencies, but in three dimensions it does not; as a result, the parenting orbits of many three-dimensional orbit families are thin membranes rather than closed curves (Merritt \\& Valluri 1999).  Such parenting orbits don't generate unique periodic sign-change sequences, so our present algorithm can't recognise them. In contrast, spectral methods work quite well in three dimensions (Carpintero \\& Aguilar 1998; Papaphilippou \\& Laskar 1998; Valluri \\& Merritt 1998).  Nonetheless, we can partly classify orbits in three-dimensional potentials by projecting their trajectories onto the principal planes and using our two-dimensional algorithm to analyse the resulting sign-change sequences.  This approach recognises resonances in which one component of the integer vector ${\\bf n}$ in (1) is zero; such resonances, though not the rule, are fairly common in many potentials (Papaphilippou \\& Laskar 1998; Merritt \\& Valluri 1999).  Some orbits obey two independent resonance conditions; such `doubly-degenerate' orbits are parented by closed three-dimensional curves.  But even doubly-degenerate orbits don't always yield strictly periodic sign-change sequences.  For example, the centrophobic `banana-pretzel' orbit, combining a $(2,-1)$ resonance in the $x$--$z$ plane with a $(4,-3)$ resonance in the $x$--$y$ plane, yields a `ZYXZ[YZ]X[ZY]ZXYZ' semi-pattern containing not one but {\\it two\\/} ambiguously-ordered pairs of symbols.  This example suggests that more general pattern-matching methods may be able classify the crossing-sequences generated by arbitrary three-dimensional resonances.  \\subsection{Future developments}  There are several ways in which this algorithm could be improved or extended.  As just mentioned, one line of attack is to generalize the pattern-scanning process.  It seems fairly easy to extend the routine which searches for arbitrary periodic patterns (Section~3.2) to recognise centrophilic as well as centrophobic semi-patterns; this would provide a better safety-net for high-order resonances.  It's also straightforward, in principle, to couple the trajectory-following and pattern-matching routines together so that the latter can call the former as needed for more data to resolve ambiguous cases.  This would be more efficient and reliable than first integrating the orbit for a fixed number of sign-changes and then analysing the results.  Another line of attack is to adapt the algorithm to analyse N-body simulations.  A simplified version has already been used to classify orbits in merger remnants (Barnes 1998; Bendo \\& Barnes 2000); however, this version just followed individual trajectories in a frozen approximation of the N-body potential.  A more interesting project would be to classify orbits directly from the computed trajectories in a simulation.  This could be accomplished by using an N-body code to generate separate sign-changes sequences for all $N$ bodies; these sequences could be quickly scanned by pattern-matching algorithms like the one described here.  Such techniques could attach a concise historical record to each body in an N-body simulation, nicely complementing the instantaneous snapshots of positions and velocities which currently form the basis for most N-body analysis.  \\vspace{0.9cm} \\noindent  We thank the referee, Dr.~P.~Palmer, for an open and constructive report, and L.~Aguilar, T.~de Zeeuw, D.~Merritt, and E.~Roberts for valuable and encouraging comments on a preliminary draft of this paper.  This work was supported in part by NASA grant NAG 5-8393 and STScI grant GO-06430.03-95A.  \\newpage"
    },
    "0010/astro-ph0010469_arXiv.txt": {
        "abstract": "Current searches for extrasolar planets have concentrated on observing the reflex Doppler shift of solar-type stars. Little is known, however, about planetary systems around non-solar-type stars.  We suggest a new method to extend planetary searches to hot white dwarfs.  Near a hot white dwarf, the atmosphere of a Jovian planet will be photoionized and emit hydrogen recombination lines, which may be detected by high-dispersion spectroscopic observations. Multi-epoch monitoring can be used to distinguish between non-LTE stellar emission and planetary emission, and to establish the orbital parameters of the detected planets. In the future, high-precision astrometric measurements of the hot white dwarf will allow the masses of the detected planets to be determined.  Searches for Jovian planets around hot white dwarfs will provide invaluable new insight on the development of planetary systems around stars more massive than the Sun and on how stellar evolution affects these systems. We present high-dispersion spectroscopic observations of the white dwarf Feige~34 to demonstrate the complexity and feasibility of the search method. ",
        "introduction": "Searches for extrasolar planets have employed four techniques: direct imaging of the star and planet, photometric monitoring for transits, and astrometric and Doppler measurements of the reflex motion of the star due to planets.  The imaging technique is difficult because a Jupiter-like planet is $\\sim$10$^{9}$ and $\\sim$10$^{4}$ times fainter than a solar-type host star at the visible and thermal IR wavelengths, respectively.  The photometric method searches for occultations, but can only detect systems with the orbital plane seen edge-on, which is an exceedingly rare circumstance. For example, a Jupiter-sized planet orbiting a solar-type star at 1~AU has only $\\lesssim$0.5\\% chance of being in the proper alignment for photometric detection. The reflex motion of a star due to planets is typically small.  To date, only the Doppler measurements have been successful in discovering planets around a pulsar \\citep{Wol92} and solar-type stars \\citep[e.g.,][]{MQ95,MB96,BM96,CH97}. With the exception of the planets around the pulsar PSR1257+12, all other known extrasolar planets are Jupiter-sized, 0.2--11 Jupiter masses (M$_{\\rm J}$), with orbital semi-major axis ranging from 0.04 AU to 3 AU.  Current searches for extrasolar planets do not explore whether intermediate-mass (2 -- 7 M$_\\odot$) stars also possess planetary systems, because these stars, of MK spectral types A and B, have fewer stellar absorption lines available for accurate velocity measurement.  Furthermore, because they are more massive than solar-type stars, the stellar reflex motion is even smaller.  These difficulties can be circumvented, if one searches for planets around the descendants of these intermediate-mass stars: hot white dwarfs with $T_{eff} > 50,000$ K. A hot white dwarf's UV radiation will photoionize the planetary atmosphere and recombination lines will be emitted by the planet. The contrast between the planet's recombination line emission and the stellar spectrum will be the limiting factor for detecting the planet.  The strength of the planet's recombination line emission is proportional to the UV flux where the stellar emission peaks, but the recombination line is emitted at a longer wavelength where the stellar emission is lower; therefore, the contrast between the planet's recombination line emission and the stellar spectrum is not as extreme as the contrast between the planet and star both in continuum.  Thus, it may be possible to find planets around hot white dwarfs by searching for a periodically Doppler-shifted recombination line emission superposed on the stellar spectrum.  We have carried out calculations to demonstrate the feasibility of this search, taking into account the stellar evolutionary effects.  In this letter we report the results of our calculations, discuss practical considerations, and present observations of a white dwarf to illustrate observational considerations. ",
        "conclusions": "We suggest that high-dispersion spectroscopic observations of hot white dwarfs can be used to search for H$\\alpha$ emission from Jovian planets within a few AU of the star.  H$\\alpha$ emission from the ionized outer envelope of a Jovian planet as weak as 1/100 the stellar emission can be detected with modern large ($\\gtrsim$4-m) telescopes. Non-LTE effects in the hot white dwarf atmosphere may produce stellar H$\\alpha$ emission.  Stellar and planetary H$\\alpha$ emission can be disentangled by multi-epoch observations if changes in the H$\\alpha$ emission line profile due to the orbital motion of the planet are detected.  Monitoring observations to establish the orbital parameters are required to distinguish between Jovian planets and a low-mass dwarf companion. High-dispersion spectroscopic observations of Feige~34 demonstrate the complexity and feasibility of this method to search for these planets.  Such a search will provide targets for future high-precision astrometric observations (e.g., SIM, adaptive optics, etc.), and valuable insight can be obtained on the development and evolution of planetary systems around stars more massive than the Sun."
    },
    "0010/astro-ph0010375_arXiv.txt": {
        "abstract": "The contribution of extragalactic objects with HI masses below $10^8 M_\\odot$ to the HI mass function remains uncertain. Several aspects of the detection of low-mass sources in HI surveys are not always considered, and as a result different analysis techniques yield widely different estimates for their number density. It is suggested at one extreme that the number density of galaxies follows a shallow Schechter power-law slope, and at the other extreme that it follows a steep faint-end rise like that found for field optical sources. Here we examine a variety of selection effects, issues of completeness, and consequences of LSS. We derive results for the large Arecibo Dual Beam Survey which indicate that the field mass function does rise steeply, while within the Virgo Cluster environs, the slope appears to be much shallower. Dependence on the local density of galaxies may partially explain differences between surveys. ",
        "introduction": "The shape of the HI mass function, particularly the behavior below $10^8 M_\\odot$, has been the subject of considerable controversy. As paramaterized by a Schechter function slope $\\alpha$, several studies seem to indicate a fairly shallow slope of $\\alpha\\approx-1.2$, including the HI masses of: optically-selected samples (e.g., Briggs \\& Rao 1993; Huchtmeier, this volume); galaxies in high-density regions like the Canes Venatici group (Kraan-Korteweg et al.~1999), Centaurus A (Banks et al. 1999), and the Ursa Major cluster (Verheijen et al.~this volume); even some ``blind'' (optically unbiased) surveys.  On the other hand, several studies suggest the slope may be considerably steeper in field samples. Initial reports at this meeting of results from the Parkes survey suggest a slope of $\\alpha=-1.5$ (Webster et al.~this volume). Interestingly, Kraan-Korteweg et al. (1999) found $\\alpha=-1.4$ when they did not restrict their sample to the Canes Venatici group. Our analysis of two earlier Arecibo surveys (Schneider et al. 1998; SSR hereafter) also suggested a steeper slope or a faint-end rise similar to the rise found for an optical sample of field objects found by Loveday (1997) and Driver \\& Phillipps (1996).  There are several possible causes for different conclusions to be drawn about the population of low-mass HI sources. Some of these have to do with astrophysical characteristics of the sample, like the effects of LSS and distance uncertainties. However, a significant cause of the differences is analysis problems: (1) a poor understanding of survey sensitivity; (2) small number statistics; and (3) selection effects.  To address these problems, we have completed the largest blind HI survey to date at the Arecibo Observatory. The Arecibo Dual-Beam Survey (ADBS; Rosenberg \\& Schneider 2000 and this volume) covers 430 sq. deg. in drift scans at 30 separate declinations between  $9^\\circ < \\delta < 28^\\circ$. We were able to achieve an rms sensitivity of 3-4 mJy per 3.3 arcmin beam along the drift scans. We also repeated most of the scans to provide internal confirmation of source detections. The advantages of using Arecibo for this type of survey are discussed by Rosenberg \\& Schneider (2000).  The most important new feature of the ADBS is the insertion of hundreds of ``synthetic'' HI signals prior to applying the data reduction procedures. These sources allow us to test  our actual sensitivity and completeness directly. We discuss these issues in detail in \\S 2. The resulting mass function for the field population of galaxies in the ADBS is derived in \\S 3.  Large-scale structure (LSS) may also have an effect on the shape of the mass function. Density differences also affect the derivation of the mass function, as do large scale flows and velocity dispersion relative to the Hubble flow. We examine these potential effects in \\S 4. Our analysis of the ADBS suggests that field samples of HI-selected sources {\\it do} show a steep slope, while optically-selected samples and cluster samples show a shallower slope. We conclude with a discussion of these results and future prospects in \\S 5. ",
        "conclusions": "The difference seen in and out of the Virgo Cluster region suggests a strong environmental dependence of the HI mass function. It is perhaps not surprising that low-mass HI sources are rare in a cluster where a deep gravitational potential is needed to retain a galaxy's gas against the effects of ram-pressure stripping. This might explain why the mass function appears much more similar in shape to the luminosity function of optically-selected galaxies.  The shallow mass function seen in moderate-density groups and clusters may be explained by related effects. For example, even though Ursa Major has a long crossing time and no evidence of a hot intracluster medium, the fact that the mean density is an order of magnitude higher than the field implies that the region has been locally converging with respect to the Hubble flow. This in turn suggests that the rate of interactions and mergers between low-mass HI sources and nearby galaxies will be greatly enhanced, again suppressing the low-mass end of the mass function.  A density dependence of the HI mass function could be important for the earlier Arecibo surveys. They covered a small area of the sky, and are therefore subject to localized differences in the HI mass function. In particular, both derived a substantial fraction of their high-mass counts from the Pisces-Perseus supercluster region, while the low-mass sources were detected nearby in regions where the density of sources is more similar to the norm for field galaxies. This contrast in LSS density for different mass sources might explain the differences from the ADBS survey, which covered a much larger area of the sky more shallowly. As a result the ADBS survey volume (excluding the Virgo Cluster) is more representative of the field.  The effects of large scale structure do not explain all of the differences between surveys, however. We comment again on the importance of producing testable measures of the completeness of HI surveys. There are too many subtle effects in the data-processing stream to simply declare a limit based on what is expected or detected. As we have shown, some low $S/N$ sources are likely to creep into a sample and may suggest a better sensitivity than is warranted. There are also clearly peculiar effects on the completeness as a result of source linewidth that may differ depending on the precise analysis procedure.  An additional effect that we have not explored, but which may be especially important for higher-resolution synthesis mapping, is the effect of source size. For example, many of the low-mass sources detected in the ADBS appear to have much larger HI diameters than their optical sizes; some appear extended even relative to the Arecibo beam. Spreading the emission over such a large area will suppress such sources' detectability in synthesis mapping, but may enhance the chance of their detection in sparsely sampled surveys like the ADBS.  The competing effect of confusion mainly influences single-dish measurements of the mass function. Since the two-point correlation function indicates that galaxies tend to cluster around other galaxies, the challenge for a large-beam 21 cm instrument is to separate sources and determine which ``should have been'' detected based on the survey sensitivity.  It will do little good to conduct larger surveys if the potential biases caused by selection effects are not well understood. The good news is that with modern computational resources, it is relatively simple to model false sources and to insert them into raw survey data. Effects of source size and confusion can be established directly, and then a more useful comparison of results can be carried out."
    },
    "0010/astro-ph0010005_arXiv.txt": {
        "abstract": "A century of observations of \\om allows us to accurately follow the period changes of a large sample of variable stars. Although the period changes are composed of monotone and irregular changes, period increases dominate in most of the RRab stars. The typical period increase rates are of the same order of magnitude as model calculations predict for the evolved, redward phase of horizontal branch evolution. For the 44 well observed RRab stars which show monotonic period changes, a mean rate of 0.15 d/Myr has been found. The period changes of the first overtone RRc type stars show a much more complex, irregular behavior. ",
        "introduction": "One of the principal goals of globular cluster photometric studies carried out during the past century was to determine period changes of RR Lyrae stars. The hope to reveal evolutionary changes motivated many extensive works. (For details see e.g. \\citet{sm95} and \\citet{rs97}.) However, attempts to connect the observed period changes directly with the different phases of horizontal branch (HB) evolution have not led to indisputable conclusions. Neither the large period decreases nor the random and/or abrupt period changes that have been detected in many clusters can be explained on the basis of steady evolutionary effects.  The basic tool in the determination of period changes of variable stars is the \\oc diagram. This plots the difference between the observed times ($O$) of a particular phase (e.g. maximum) of the light-curve and the predicted times of the same phase ($C$) calculated according to an accepted ephemeris. A nonlinear \\oc indicates period changes while a linear \\oc with a non-zero slope points to inaccuracy of the adopted period.  \\citet{ste34} showed that period-noise can have a cumulative effect in the \\oc of variable stars, thus mimicking true period changes. However, in case of RR Lyrae stars \\citet{bdd65} concluded that their \\oc diagrams cannot be wholly explained as the result of period-noise. Nevertheless, \\citet{lee91} showed that the mean period change rates observed in globular clusters were consistent with synthetic HB models if the HB types of the clusters were also considered. However, accurate determinations of period changes are available for only a small sample of clusters and for many of these, the derived period change rates vary from one study to another (see e.~g. the $0.07$ d/Myr difference in the mean period change rate for NGC~7006 given  by \\citet{we92} and \\citet{we99}). Thus it is still an important task to define accurate period change rates in globular clusters by using not only the most recent data, but also any earlier observations, even if they can only be reconstructed from published light curves or epochs of maximum light. Searching for unpublished cluster observations and making them accessible for future studies is very worthwhile.   The preference for period increases in \\om was first noticed by \\citet{ma38}. \\citet{be64} strengthened this result and raised the possibility of an evolutionary explanation. The recent CCD observations of the variables \\citep{kal97} within the framework of the OGLE (Optical Gravitational Lensing Experiment) project have lengthened the time-base of the observations to more than one hundred years, thus making it possible to extend the earlier period change studies. ",
        "conclusions": "We have shown in Sect.~\\ref{ab} that most of the RRab stars have increased their periods by some $10^{-11}-10^{-9}$ days/day rate during the past century. This is in agreement with evolutionary model predictions for the late, redward phase of the horizontal branch evolution. The period changes along the post zero age horizontal branch, as derived from the \\citet{do92} models and from the linear pulsation period formulae of \\citet{kb94}, range typically from $- 10^{-10}$ to $+ 10^{-9}$ days/day. Within the instability strip there is no metallicity and mass value for which the evolutionary period decrease rate is larger than some $ 10^{-11}$ days/day, but period increases of the order of $10^{-9}$ days/day do occur in the redward phase of the blue loops in the HB evolution, especially for the lower mass stars.  The physical parameters derived from the light-curves (Jurcsik 1998b, and Table~\\ref{paramab}) indicate that most of the RRab stars in \\om belong to a very homogeneous population, and are already in the evolved (redward) phase of their HB evolution. Thus, their comparison with model predictions of unique chemical composition models give adequate results, in spite of the fact that large chemical inhomogeneities have been detected in \\om stars.  In Fig.~\\ref{fig7} horizontal branch evolutionary tracks are shown for oxygen enhanced models with [Fe/H]$=-1.48$ dex \\citep{do92}. Tracks for four different masses ($0.66, 0.64, 0.62, 0.60 M_{\\odot}$) are shown. The position of the 40 RRab stars which were found to have [Fe/H]$=-1.54\\pm0.08$ dex \\citep{j98b} are also given. The luminosities and temperatures of these variables (given in Table~\\ref{paramab}) were determined by using the empirical relations between the physical parameters and the Fourier coefficients of the light curves (see \\citet{j98a} and references therein). The empirical $\\log L$ and $\\log T$ scales are shifted by 0.10 and 0.016, respectively, in order to reach agreement with evolutionary results \\citep{jk99}. This transformation yields a mean pulsational mass of $0.64M_{\\odot}$ with $0.02M_{\\odot}$ rms scatter for these 40 variables.  Linear pulsational periods have been calculated for different locations along the tracks where the fundamental mode period falls within the $0.4 -1.0$~d range. Then, evolutionary period changes have been calculated using these periods and the evolutionary time scales of the \\citet{do92} models. They are given in units of $10^{-10}$ days/day in Fig.~\\ref{fig7}. The distribution of the observed $\\beta$ values shown in Fig.~\\ref{fig4} for those RRab stars which have well defined monotonic period change rates is in very good agreement with the evolutionary $\\beta$ values calculated within the region occupied by these stars.  Summarizing our results, we have found that, for those \\om RRab stars that exhibit dominantly monotonic period changes, the period change rates can be brought into very good agreement with horizontal branch evolutionary model predictions. Thus the observed period changes of these stars may directly indicate evolutionary effects. However, irregular period changes and one very large period decrease rate (similar to those observed in other globular clusters) also occur in \\om and no satisfactory explanation for these phenomena has yet been given. The much less regular period change behaviour of the first overtone RRc stars is also not yet understood."
    },
    "0010/astro-ph0010233_arXiv.txt": {
        "abstract": "We have collected pulsar flux density observations and compiled spectra of 281 objects. The database of Lorimer et al. (\\cite{lorimer}) has been extended to frequencies higher than 1.4 GHz and lower than 300 MHz. Our results show that above 100 MHz the spectra of the majority of pulsars can be described by a simple power law with average value of spectral index $<\\alpha> = -1.8\\pm0.2$. A rigorous analysis of spectral fitting revealed only about 5\\% of spectra which can be modelled by the two power law. Thus, it seems that single power law is a rule and the two power law spectrum is a rather rare exception, of an unknown origin, to this rule. We have recognized a small number of pulsars with almost flat spectrum ($\\alpha \\geq -1.0$) in the wide frequency range (from 300 MHz to 20 GHz) as well as few pulsars with a turn-over at unusually high frequency ($\\sim$~1~GHz). ",
        "introduction": "One of the main observables of pulsar emission is its flux density ${S_{\\nu }}$, measured at a given central frequency $\\nu $ of the receiver bandwidth. Flux measurements are crucial for deriving, the so called, pulsar luminosity function and therefore the birth rate of the Galactic population of neutron stars. The first spectrum of a pulsar taken at 5 frequencies was published by Robinson et al. (\\cite{robinson}). The flux density variations and spectra for frequencies between 0.15 and 1.6 GHz were later reported for a number of pulsars by McLean (\\cite{mclean}). Spectra of 27 pulsars were published by Sieber (\\cite{sieber73}) who used pulse energy values obtained with the 100-m and 25-m telescopes of the Max-Planck-Institute for Radioastronomy (MPIfR) as well as other values published in the literature. He was the first to show a turn-over behaviour at low frequencies around 100 MHz and a break in spectrum at high frequencies (about 1 GHz). Sieber et al. (\\cite{sieber75}) published pulse shapes and their energies for 35 pulsars at 2.7 and 4.9 GHz, and for 7 pulsars at 10.7 GHz. Izvekova et al. (\\cite{izvekova}) and Slee et al. (\\cite{slee}) presented an analysis of flux measurements for pulsars at meter wavelengths ($\\sim$~80 MHz). A compilation of spectra of 45 pulsars over a wide frequency range was published for the first time by Malofeev et al. (\\cite{malofeev94}). Seiradakis et al. (\\cite{seiradakis}) published a collection of high frequency data on pulsar profiles and flux densities for 183 pulsars at 1.4 GHz, 46 pulsars at 4.85 GHz and 24 pulsars at 10.5 GHz. The catalogue of pulsar flux density measurements for 281 pulsars at frequencies ranging from 0.4 GHz to 1.6 GHz was published by Lorimer et al. (\\cite{lorimer}). The first spectra for four millisecond pulsars were published by Foster et al. (\\cite{foster}). Kramer et al. (\\cite{kramer98}) presented the results of flux density measurements for 23 millisecond pulsars at frequencies 1.4 and 1.7 GHz and an analysis of their spectra. Van Ommen et al. (\\cite{ommen}) presented polarimetric data together with flux density measurements for a large number of southern pulsars at frequencies of 800 MHz and 950 MHz. Most recently, Toscano et al. (\\cite{toscano}) have presented flux density measurements for southern millisecond and slow pulsars at frequencies between 0.4 GHz and 1.6 GHz, and Kramer et al.~(\\cite{kramer99}) studied the emission properties of millisecond pulsars up 4.85 GHz. A first large sample of flux densities of weak pulsars at 4.85 GHz was published by Kijak et al. (\\cite{kijak}).  It has been long believed that flux density spectra for most of the pulsars have been well described by a simple power law $S \\propto \\nu ^{\\alpha }$ with spectral index of $\\sim {-1.6}$ (Sieber \\cite{sieber73} and Lorimer et al. \\cite{lorimer}). However, a considerable fraction of pulsars demonstrated spectra that required modelling by two power laws (Sieber \\cite{sieber73} and Malofeev et al. \\cite{malofeev94}). These were commonly called the broken-type spectra. In this paper we show that the two power law spectra are rare exception and that majority of pulsars can be modelled by a single power law. Our analysis shows that only 10\\% of pulsars requires two power law spectra.   Recently, pulsar flux density measurements have been extended to the mm-wavelengths region, which provided information about this newly explored spectral region. These measurements suggested that spectrum flattens out or even turns up at very high frequencies (Kramer et al. \\cite{kramer96}).  In this paper we present a new and most complete compilation of spectra for 281 pulsars in the frequency range from 39 MHz to 43 GHz. The flux density measurements for frequencies from 300 MHz to 1.4 GHz were done at Jodrell Bank (Lorimer et al. \\cite{lorimer}). For frequencies equal or above 1.4 GHz we have utilized flux measurements made by different authors at the Effelsberg Radiotelescope of the Max-Planck Institute for Radioastronomy (Sieber et al. \\cite{sieber75}, Bartel et al. \\cite{bartel}, Sieber \\& Wielebinski \\cite{sieber87}, Wielebinski et al. \\cite{wielebinski}, Seiradakis et al. \\cite{seiradakis}, Kramer et al. \\cite{kramer96}, Hoensbroech et al. \\cite{hoensbroech}, Kramer et al. \\cite{kramer97}, Kijak et al. \\cite{kijak}). We have also reduced a significant amount of the unpublished data available at the archives of the MPIfR. Most of these data is available throughout the European Pulsar Network Database (Lorimer et al. \\cite{lorimer98}). We have also included the data published by Izvekova et al. (\\cite{izvekova}) and Malofeev et al. (\\cite{malofeev2000}) containing observations made at low radio frequencies from 39 MHz to 102.5 MHz. Our own observations at 4.85 GHz made in Effelsberg in 1998 are also included. ",
        "conclusions": "The main conclusion of this paper is that the single power law spectrum is a rule and the two power law spectrum is a rare exception to this rule. One could therefore think that the nature of this exception is that the spectrum just becomes steeper starting from some, relatively high frequency. However, inspection of Fig.~\\ref{1} indicates that this might not be a case. The distribution of $\\alpha_1$ seems different from that of $\\alpha$, meaning that in pulsars with two power law spectra the average spectral index $<\\alpha_1>$ is typically much larger than the average $<\\alpha>$ (Fig.~\\ref{1}b) and of course, the high frequency index $<\\alpha_2>$ is much smaller than $<\\alpha>$ (Fig.~\\ref{1}c). Thus, it seems that the two power law spectra are qualitatively different from the typical single power law spectra. In this paper we obtained spectral index for a large sample of pulsars in a wide frequency range (form 400 MHz to 23 GHz) . The average spectral index of pulsars $<\\alpha>$ with simple power-law spectrum in our sample is $\\rm-1.8\\pm0.2$ which agrees with results obtained by other authors (see Table 4).  The distribution of spectral indices is symmetric and almost Gaussian. The average indices for the broken-type spectra are $<\\alpha_1> = -0.9\\pm0.5$ and $<\\alpha_2> = -2.2\\pm0.9$, respectively, with a break frequency of $<{\\nu}_b> \\approx 1.5$ GHz on the average.  We have not found any correlations between spectral index and rotation period $P$, spin-down rate $\\dot P$, characteristic age $\\tau$, polarization and profile type for pulsars with both simple power law spectra and two-power-law spectra. We have found 2 young, nearly fully polarized pulsars which indicate turn-over at unusually high frequency ($\\sim$~1~GHz). We have also found 15 pulsars which definitely require two-power-law spectra. The comparison of pulsar spectra analysis for slow and millisecond pulsars indicates that both groups have the same emission mechanism (Table~4).  \\begin{table} \\caption{Spectral indices obtained by different authors.} \\begin{tabular}{crcl} \\hline Spectral &\tNo. of & Freq. range & References\\\\ index & PSRs & [GHz]  &\\\\ \\hline -1.6 &\t  27 &\t0.1 -- 10 &\tSieber 1973\\\\ -1.9 &        20 &    0.1 -- 30          &               Malofeev et al. 1994\\\\ -1.6 &\t280 &\t0.3 -- 1.6 &\tLorimer et al. 1995\\\\ -1.7 &\t284&\t0.1 -- 10\t&\tMalofeev et al. 1996\\\\ -1.8 &\t  32 &\t0.3 -- 4.9 &\tKramer et al. 1998, 1999\\\\ &           &                         &              (millisecond PSRs)\\\\ -1.7 &\t216 &\t0.4 -- 1.5 &\tToscano et al. 1998\\\\ &           &                         &               (southern PSRs)\\\\ -1.9 &\t19 &\t0.4 -- 1.5 &\tToscano et al. 1998\\\\ &           &                         &               (millisecond PSRs)\\\\ -1.9 &\t144 &\t1.4 -- 4.9 &\tKijak et al. 1998\\\\ -1.8 &\t281 &\t0.4 -- 23 &\tthis paper\\\\ \\hline \\end{tabular} \\end{table}  Postscript files of spectra are available at http://astro.ca.wsp.zgora.pl/olaf/paper1 and our spectra are also presented in EPN Database at http://www.mpifr-bonn.mpg.de"
    },
    "0010/astro-ph0010555_arXiv.txt": {
        "abstract": "We searched for coherent X-ray pulsations from \\aql\\ in a series of \\rxte\\ observations taken shortly after a recent outburst. During the course of these observations, \\aql\\ passes through an apparent ``propeller'' phase as its luminosity fades to its quiescent value. No pulsations were detected, and we place upper limits (ranging from $0.52\\%$ to $9.0\\%$) on the fractional RMS amplitude of any periodic signal contained in the various data sets searched. This result has implications for the geometry of the system, if the quiescent luminosity is due to continued low-level accretion. Alternatively, our result supports the idea that the quiescent luminosity may be due to thermal emission. ",
        "introduction": "The origin of the quiescent emission observed from low-magnetic field transient neutron stars (NSs) has posed something of a puzzle.  When first discovered observationally (in the transient \\cenx; \\citenp{jvp87}), the X-ray emission, fit with a blackbody spectrum, was too faint to have originated over the entire surface of a 10~km NS.  It was suggested that the emission was due to accretion over a fraction ($\\sim \\rm{few}\\%$) of the NS surface.  As additional quiescent NS X-ray spectra were observed and fit with a blackbody spectrum \\citep{verbunt94,asai96a,asai96b}, the small implied emission areas seemed to confirm this view.  The cause of accretion anisotropy over the surface of a low B-field NS has been purported to be the ``propeller effect'' \\citep{ill75,stella86}, in which at low accretion rates, the magnetic field increases in importance relative to the gravitational field in determining the accretion geometry, perhaps expelling much of the accretion from the system when the effect of the magnetic field is comparable to or stronger than that of the gravitational field at the radius where the Keplerian frequency is equal to the spin frequency of the NS.  When this occurs, some fraction of the accretion may follow the magnetic field lines to the magnetic poles which, if these are offset from the rotational poles, could conceivably produce X-ray pulsations.  The efficiency of such a propeller has been estimated recently \\citep{menou99}.  However, some observations of decreasing torque with increasing luminosity may indicate deviations from this standard magnetic accretion scenario in low mass X-ray binaries \\citep{chakrabarty97}.  If, after an outburst, accretion continues into quiescence then pulsations might be expected, particularly if the NS can be shown to have gone through a ``propeller phase'', in which the accretion geometry in the vicinity of the NS is dominated by the magnetic field, rather than by the gravitational field. No pulsations from these objects in quiescence have yet been reported, although they have been searched for \\citep{verbunt94,asai96b}.  Thus, until recently, the view of the quiescent luminosity of low-B transient NSs has been as due to continued accretion over a fraction of the surface of the NS, perhaps caused by modified accretion geometry due to the effects of the magnetic field, which may therefore give rise to pulsations in the X-ray intensity.  A different view of the origin of the quiescent luminosity has recently been described \\citep{brown98}, in which nuclear reactions at the base of the NS crust keep the NS core heated to temperatures ($\\approx10^8\\, \\rm{K}$) sufficient to explain a large fraction, if not all, of the quiescent luminosity of these objects (\\ee{32-33} \\cgslum); in this manner, the quiescent luminosity can be produced in the absence of active accretion.  The discrepancy in the emission area can be explained as due to an incorrect blackbody assumption -- specifically, if accretion is shut-off, metals stratify in the NS atmosphere \\citep{alcock80,romani87}, producing a pure H photosphere, in which the free-free opacity, which decreases with increasing frequency, dominates, permitting higher E photons to originate at greater depth, thus at higher temperatures \\citep{rajagopal96,zavlin96}.  This produces an emergent spectrum which is spectrally harder than a simple blackbody, and -- when described as a simple blackbody -- requires a higher $T$, and subsequently lower emission area, to parameterize it.  Spectral fits with H atmosphere spectra show that the observed X-ray spectra from the three such objects for which data exists (\\cenx, \\aql, and 4U~1608-522) imply emission areas consistent with a 10 km NS \\citep{rutledge99}.  Thus, both their quiescent luminosities and their emergent X-ray spectra can be explained in this manner, without invoking active accretion during quiescence.  An attractive feature of this explanation is that it accounts for the similar luminosities of these three objects, without requiring their quiescent mass accretion rates (a product of the propeller efficiency and mass accretion rate through the accretion disk) to be serendipitously similar.  It remains an open question if accretion (at a level of $\\dot{M}$\\approxlt$10^{-12}$\\msun\\  yr$^{-1}$) occurs at all, contributing some fraction of the quiescent X-ray luminosity.  There are few observational means by which this may be investigated.  One such possibility is to search for metal lines in the quiescent X-ray spectrum, which, if present, would imply that metals are being fed into the atmosphere at a rate greater than gravity can stratify it (\\citenp{bildsten92}, see \\citenp{zavlin96} for X-ray spectra of metallic atmospheres).  Such spectra can be investigated by the latest generation of X-ray spectroscopy missions (\\axaf\\ and \\xmm). A second method is to search for intensity variability during quiescence.  In particular, as described above, if pulsational variability at a particular frequency were to be found in quiescence, this would support the scenario of active accretion along the magnetic field lines to the NS surface. Alternatively, if the NS magnetic field were strong enough to significantly affect the photospheric opacity ($>10^{10}$~G; cf. \\citenp{zavlin95}), this would produce apparent ``hot spots'' near the magnetic poles of the NS which, if offset from the rotation axis, could give rise to pulsations; however, it is thought that type-I X-ray bursting sources have magnetic fields below this value. Clearly discovery of pulsations or constraints on the pulsed fraction during X-ray quiescence (following the onset of a ``propeller'') would provide useful limits for models of quiescent NS emission.  The system \\aql\\ lends itself to such a search for pulsations.  During the X-ray decline of a recent outburst, the intensity at first decreased by one order of magnitude over 17 days, then abruptly (beginning at $L_x\\approx10^{36}\\ \\cgslum$) by 3 orders of magnitude over 3 days, which was coupled with the sudden spectral hardening of the X-ray intensity \\citep{campana98,zhang98a}.  The X-ray intensity stopped dropping at a luminosity $L_x\\sim10^{33}\\ \\cgslum$, comparable to the quiescent luminosity observed previously \\citep{verbunt94}. This was interpreted as evidence for a propeller-phase, though this remains to be confirmed through repeated observations of \\aql\\ in outburst. If indeed this behavior does indicate a propeller phase in \\aql, it offers the opportunity to search for pulsations during the period when the NS magnetic field substantially alters the accretion geometry.  Here we have searched for, and not found, pulsations in \\aql\\ during and immediately following its supposed propeller phase. In Sec.~\\ref{sec:analysis}, we describe our search method and the results of our analysis.  In Sec.~\\ref{sec:discuss} we discuss these results and our conclusions. ",
        "conclusions": "\\label{sec:discuss}  We find no evidence of pulsations from \\aql\\ as it fades into quiescence. The upper limits on the pulsed fraction for the eight data sets searched range from $0.52\\%$ to $9.0\\%$ RMS. If, indeed, during the period identified as the ``propeller phase'' \\citep{campana98,zhang98a} the NS magnetic field significantly modifies the accretion flow geometry in the vicinity of the NS, the apparent absence of pulsations does not support the hypothesis that the quiescent emission is due to continued accretion. Our results cannot completely rule out such accretion, however, since it is possible that the geometry of the system may not lead to detectable pulsations; for example, the magnetic axis may be very nearly aligned with the rotation axis, or the rotation axis may point directly towards the earth.  The non-detection of pulsations coupled with the observation of a thermal spectrum during quiescence \\citep{rutledge99} favors the interpretation that the quiescent luminosity is not due to accretion, but rather to a hot NS core \\citep{brown98}."
    },
    "0010/astro-ph0010249_arXiv.txt": {
        "abstract": "In models of hierarchical structure formation, interactions and mergers at high redshift play a key role in the formation and evolution of galaxies. Numerical modeling and observations of nearby systems explore the detailed physics of galaxy interactions, but applying these results to the high redshift universe remains complicated. Evolution in the properties of galaxies over cosmic history may lead to differences in interaction-induced star formation as a function of redshift. High redshift interacting galaxies may be more prone to extreme, disk-wide starburst activity than are low redshift systems. Also, while activity in nearby clusters seems to be largely diminished, at higher redshift we observe galaxies falling into clusters for the first time. A variety of processes, including interactions and mergers, can drive strong evolution in cluster galaxies at higher redshift. ",
        "introduction": "Most of our detailed knowledge of the effects interactions have on galaxies comes from studies grounded in the local universe. Well-studied samples of interacting galaxies such as the Toomre (1977) sequence extend only out to $\\sim 10,000$ km/s, and so much spatial information is lost at redshifts greater than a few tenths that we can only rely on statistical studies of interactions at higher redshift. Dynamical models, which have illuminated many of the physical processes involved during galaxy collisions, rely heavily on galaxy models constructed to resemble nearby galaxies in terms of their structural and kinematic properties.  Whether such models accurately portray interacting galaxies at high redshift is unclear -- structural differences between high and low redshift galaxies may translate to substantial differences in the response to an interaction.  Nonetheless, these dynamical models allow us to view the basic evolution of a merging encounter (Figure 1, from Mihos \\& Hernquist 1996 [MH96]). Shortly after the initial passage, the galaxies become strongly distorted, with disk self-gravity amplifying the perturbation into strong spiral arms or central bars. As the galaxies separate, dynamical friction between the dark halos slows the galaxies on their orbit, causing them to fall back and merge together.  The violent relaxation accompanying the merger destroys the disks and results in the creation of an elliptical-like remnant. During the interaction and ensuing merger, a large fraction of the galaxies' ISM is compressed and driven inwards due largely to gravitational torques during the encounter. This compression and inflow presumably leads to strong starburst and/or AGN activity during the encounter.  \\begin{figure} \\plottwo{mihos1a.eps}{mihos1b.eps} \\caption{Left: Evolution of an equal mass merger, from MH96. Time is given in model units; the sequence covers approximately 750 Myr. Right: Evolution of the inflowing gas. The subpanels show (top left) the angular momentum of the gas, (top right) the gravitational and hydrodynamic torques acting on the gas, (bottom left) the gravitational torques from the host and companion galaxies respectively, and (bottom right) the growth of nonaxisymmetric modes in the stellar disk.} \\label{fig1} \\end{figure}  The triggering of enhanced star formation by interactions is borne out in the observational record, but the effects, on average, are not extreme. Optically selected samples of interacting galaxies show star formation rates elevated only by factors of a few over normal galaxies, with inferred starburst mass fractions of only a few percent (e.g., Larson \\& Tinsley 1978; Kennicutt et al. 1987).  This enhanced SFR is preferentially found in the nuclear regions of galaxies (Keel et al. 1985; Kennicutt et al. 1987), suggesting that nuclear inflow is the preferred method for triggering starbursts in interacting systems at low redshift.  The most extreme examples of merger-induced activity are found in infrared samples; indeed, the most luminous starbursts in the local universe, ultraluminous infrared galaxies (ULIRGs) with luminosities $> 10^{12}$ L$_{\\sun}$ all show evidence of being interacting systems in the final throes of merging (Sanders \\& Mirabel 1996). Yet it is important to realize that not all merging systems are ultraluminous -- the link between mergers and luminous activity is not a simple one.  Given the wide range of star-forming properties exhibited by interacting systems, what determines how a galaxy responds to an interaction? Dynamical modeling has shown that stability of the host disk plays an essential role in the degree of star-forming response. Also shown in Figure 1 is the evolution of the gas in the prograde disk during the encounter.  Two distinct phases of inflow are seen -- one early, and a second late in the encounter when the galaxies ultimately merge.  By decomposing the torques which drive this gas inwards, we see that they are largely gravitational in origin, and that the first phase of inflow is driven by gravitational instabilities in the host disk rather than from the passing companion. The rapid inflow shown in Figure 1 is the result of bars and strong spiral features in the host disk; in these features, the gas tends to slightly lead the stellar component (Barnes \\& Hernquist 1996 [BH96]), resulting in a strong gravitational torque which drives the gas inwards. In this model, the first phase of inflow is halted by the presence of a central bulge which stabilizes the inner disk; when the galaxies then ultimately merge this stability is overwhelmed by the merger process and a second phase of inflow commences. In models without a bulge, instabilities are so strong that the gas is driven to the nucleus in a single phase at early times, while the galaxies are still well-separated.  Because nuclear gas inflows are so intimately tied to instabilities in the galactic disk, star formation in interacting galaxies should depend strongly on the stability of disks against strong perturbations.  This idea has been demonstrated in a number of theoretical studies. A variety of mechanisms have been shown to suppress instabilities and slow the inflow, including the presence of a dense central bulge (MH96), a lowered disk surface density relative to the dark matter content (Mihos et al. 1997), and retrograde encounter geometries, which reduce the spin-orbit coupling driving the instability (BH96). In these cases, early inflow is suppressed in favor of milder gas compressions (and, presumably, milder star formation enhancement) throughout the disk. ",
        "conclusions": ""
    },
    "0010/astro-ph0010280_arXiv.txt": {
        "abstract": "We  present  evidence of a direct accretion disk $-$ jet connection in the Galactic microquasar GRS~1915+105 based on  our analysis of RXTE/PCA data with a ``spike'' in X-ray light curves. We find that the radio emission increases as the hardness ratio increases during the low hard state. We suggest that the ``spike'' which separates the dips with hard and soft spectra marks the beginning of the burst phase when the luminosity of the soft X-rays (5$-$15 keV) increases by a large factor ($\\sim$ 10). This produces a major ejection episode  of the synchrotron - emitting plasma termed as ``baby jets'' which are associated with infrared (IR) and radio flares of about half an hour period widely reported in the literature. Subsequent short but frequent soft dips produce overlapping faint flares which result in an enhanced level of quasi-steady emission. We discuss the differences between ``baby jets'' and relativistic radio jets and especially investigate their signatures in X-rays. ",
        "introduction": "Two Galactic X-ray transient sources  GRS~1915+105 and GRO~J1655-40 are known to produce relativistic radio jets (\\cite{mira:94}; \\cite{ting:95}).  The combination of relativistic jets and a suspected central black hole has earned these two objects the name ``microquasars'' as they seem to be stellar mass analogs of the massive black hole systems in quasars and other active galactic nuclei (AGNs) (\\cite{morg:97}; \\cite{bell:97b}; \\cite{oros:97}).  Since microquasars are much smaller, closer and show faster variability than the extragalactic systems, they are  potential ``laboratories'' for the study of black hole accretion/relativistic jet systems. GRS~1915+105 has shown spectacular X-ray variability since its discovery in 1992 (\\cite{cast:94}). Recently, Belloni et al. (2000) have classified a large sample of  RXTE/PCA observations in 12 separate classes on the basis of their light curves and the color-color diagrams. Out of these, the $\\beta$ class is described as the most complex, has  all the three basic states of the source (\\cite{bell:00}), and is always accompanied by IR/radio flares. The presence of a ``spike'' in the X-ray light curve, which separates the dips with hard and soft spectra, clearly distinguishes this class from others. The ``spike'' coincides with the beginning of IR flares seen during simultaneous X-ray and IR observations, suggesting its role in initiating IR flares (\\cite{eike:98}). These flares are  termed as ``baby jets'' from energy consideration.  Another simultaneous observations of GRS 1915+105 in the X-ray, IR, and radio wavelengths confirm that the IR and radio flares are associated with the X-ray dips (\\cite{mira:98}).  \\begin{deluxetable}{llclccc} \\footnotesize \\tablecaption{SUMMARY OF THE SELECTED OBSERVATIONS OF GRS~1915+105  \\label{tbl-1}} \\tablewidth{0pt} \\tablehead{ \\colhead{Observation ID} & \\colhead{Date} &   \\colhead{Exposure} & \\colhead{PCU\\tablenotemark{a}}  &  \\colhead{Radio\\tablenotemark{b}} & \\multicolumn{2}{c}{Infrared (IR) or Radio flares}\\\\ \\colhead{} & \\colhead{} &   \\colhead{(s)} &  \\colhead{on}  &\\colhead{Flux (mJy)}& \\colhead{Period (min)}& \\colhead{Peak ampl.(mJy)} } \\startdata 10408-01-38-00& 1996 Oct 07 & 7000 &5&3& -& -\\nl 20186-03-03-01& 1997 Aug 14 & 10000&5&29& $\\sim$ 30& $\\sim$ 12 $^\\dag$\\nl 20402-01-45-03& 1997 Sep 09 & 10000&4&47& $\\sim$ 30& $\\sim$ 60 $^\\S$\\nl 20402-01-52-01& 1997 Oct 30 & 4000&5&200& $\\sim$ 25& $\\sim$ 200 $^\\ddag$\\nl 20402-01-53-00& 1997 Oct 31 & 10000&5\\tablenotemark{c}&170& $\\sim$ 30& $\\sim$ 150 $^\\ddag$\\nl \\enddata \\tablenotetext{ } { $^\\dag$ IR flares (2.2 $\\mu$m) \\cite{eike:98}, $^\\S$ Radio flares (8.3 GHz) \\cite{mira:98}, $^\\ddag$ Radio flares (15 GHz) \\cite{fend:99}.} \\tablenotetext{a}{Number of Proportional Counter Units (PCUs) operating at the time of observation.} \\tablenotetext{b}{Radio flux is obtained by  interpolation  of  8.3 GHz public domain data from NSF-NRAO NASA except on  1996 Oct. 7 for which data are  for 15.2 GHz estimated from figure 4 of Pooley \\& Fender (1997).} \\tablenotetext{c}{Except in the time range of 1.20887 $\\times 10^8$ to 1.20889 $\\times 10^8$ (MJD $\\sim$ 50752.6) when only 3 PCUs  were on.} \\end{deluxetable}  At present, the disk-jet connection is indirect at best for relativistic radio jets.   Harmon et al. (1997) have shown a long-term correlation between hard X-ray flux and jet activity in GRS~1915+105. Fender et al. (1999) have observed four relativistic radio ejections in 1997 October/November and have presented high quality MERLIN radio images of ejecta in 400$-$5000 AU scales which are consistent with ballistic motion. Recently, Eikenberry et al. (2000) have reported detection of faint IR flares and have classified the ejection events associated with IR and radio  flares into three classes: (A) the  relativistic events producing bright superluminal radio jets of $\\sim$ 1 Jy with decay times of several days (\\cite{mira:94} ; \\cite{fend:99}), (B) the ``baby jets'' associated with $\\sim$ 100$-$200 mJy IR and radio flares with decay times of several minutes (\\cite{eike:98}; \\cite{mira:98}), and (C) faint IR flares with peak amplitude of $\\sim$ 0.5 mJy and duration of 8-10 minutes (\\cite{eike:00}). In  this paper we present the results of our detailed analysis of RXTE/PCA observations spread over a year of $\\beta$ and $\\lambda$ classes which actually belong to the same class as shown here. We  investigate the role of the ``spike'' in initiating the ``baby jets'' and study the accretion disk corresponding  to IR/radio flares of varying peak amplitudes (Table 1). We present  evidence of a direct disk $-$ jet connection for radio ejections  of class A. We discuss the dissipation  of accretion energy in the inner part of the disk in terms of a  comptonising cloud and  the escaping  mass which produces synchrotron radiation. ",
        "conclusions": "We  present    evidence of a direct accretion disk $-$ jet connection for relativistic radio jets. We have provided here an explanation for the most complex light curves with a ``spike''  observed  to date in the Galactic microquasar GRS~1915+105. We suggest that the ``spike'' which separates the dips with hard and soft spectra marks the beginning of the burst phase, when the luminosity of the soft X-rays  increases by a large factor ($\\sim$ 10). This produces a major ejection episode of the synchrotron - emitting plasma termed as ``baby jets''  with the half-hour spacing  widely reported in the literature. Subsequent short but frequent soft dips produce overlapping faint flares which result in an enhanced level of quasi-steady emission.  The radio emission during ``baby jets'' is consistent with synchrotron emission from an adiabatically expanding small cloud (within a distance of few tens of AU of the inner accretion disk) with a flat radio spectrum  and a decay time of several minutes. On the other hand, the  radio emission during the relativistic radio jets is consistent with synchrotron emission from the extended radio cloud  with ballistic motion on a scale of a few hundreds to few thousands AU and a decay time of a few days. These jets require continuous replenishment of energetic particles. Our results support  the relativistic jet model with  quasi-continuous mass  ejection and strong variations in the jet speeds required for the formation of strong internal shocks to  continuously accelerate the particles."
    },
    "0010/astro-ph0010139_arXiv.txt": {
        "abstract": "Through studying the hard lags between the soft (3.3 -- 5.8 keV) and hard (13.0 -- 41.0 keV) photons of the 0.5 -- 10 Hz QPOs in GRS 1915+105, we have classified them into three types: 0.5 -- 2.0 Hz QPOs, 2.0 -- 4.5 Hz QPOs, and 4.5 -- 10 Hz QPOs. They are closely related to different temporal and spectral states. The first type of QPOs (0.5 -- 2 Hz) have positive hard lags at both the QPO fundamental and first harmonic frequencies. These QPOs were observed in the quiescent soft state. The second  type of QPOs (2 -- 4.5 Hz), which were also detected in the quiescent soft state, have opposite signs of hard lags at the QPO fundamental and first harmonic frequencies. The third type (4.5 -- 10 Hz), which showed up in medium soft quiescent/out-burst states, do not have significant higher harmonic peaks. There is a smooth transition between these three types of QPO behaviors. We did not detect 0.5 -- 10 Hz QPOs in the very soft state. We discuss some of the implications of these results. ",
        "introduction": "The superluminal source GRS 1915+105 displays a rich diversity of lightcurve morphology, power density spectrum (PDS), quasi-periodic oscillations (QPOs), phase lags, and coherence. The lightcurve morphology varies from out-burst to intermittent out-burst, to quiescence (e.g. Belloni et al. 1997, 2000). The PDS shape can be either a broken power law with a flat top or a simple power law (Morgan, Remillard, \\& Greiner 1997). The QPO fundamental frequency ranges from mHz to 67 Hz (e.g. Morgan, Remillard, \\& Greiner 1997), and some QPOs were detected up to the third harmonic (Cui 1999).  Several studies have related the temporal behaviors with the spectral properties. For example, using the standard disk blackbody and power law model, Muno, Morgan, \\& Remillard (1999) found that the 0.5 -- 10 Hz QPO frequency increases as the inner-disk color temperature increases from 0.7 to 1.5 keV. Chen, Swank, \\& Taam (1997) showed that the hardness ratio between the energy bands 11 -- 30.5 keV and 2 -- 11 keV is a good indicator for the presence of two types of QPOs. When this hardness ratio is above 0.1, GRS 1915+105 exhibited narrow 0.5 - 6 Hz QPOs. Broad QPOs or no QPO were observed when this hardness ratio is below 0.1. According to Markwardt, Swank, \\& Taam (1999), the 1 -- 15 Hz QPOs were present when the power-law component became hard and intense, but the QPO frequency is correlated with the parameters of the thermal component. In these studies, the 0.5 -- 10 Hz QPOs were treated as a single type of QPOs. However, our studies of the phase lag behaviors of the 0.5 -- 10 Hz QPOs indicate that there are three types of 0.5 -- 10 Hz QPO behaviors which can be classified according to the QPO frequency: 0.5 -- 2.0 Hz, 2.0 -- 4.5 Hz, and 4.5 -- 10 Hz.  Complex phase lag (or hard lag) behaviors have been observed in the 3 -- 12 Hz QPOs of XTE J1550--564 (Wijnands, Homan, \\& van der Klis 1999) and in the 66.8 mHz QPO of GRS 1915+105 (Cui 1999). Wijnands, Homan, \\& van der Klis (1999) found two types of 3 -- 12 Hz QPOs in XTE J1550--564. The first type of QPOs have a broad QPO peak with a QPO frequency of 6 Hz. The hard lags of this type of QPOs are negative at both the fundamental and first harmonic frequencies. The second type of QPOs, which have narrow peaks and a fundamental peak frequency of 3 Hz, switch signs for the hard lags at the different harmonic frequencies. Similarly, Cui (1999) found that the hard lags of the 66.8 mHz QPO in GRS 1915+105 alternate from negative to positive values as the frequency increases from the fundamental to higher harmonic frequencies.  In this letter, we show the different phase lag behaviors of 0.5 -- 10 Hz QPOs in 20 {\\it Rossi X-ray Timing Explorer (RXTE)} observations of GRS 1915+105. We find that the phase lag behaviors are closely related to the QPO fundamental frequency and the spectral states. In section 2, we describe how the data were analyzed. In section 3, the analysis results are presented. In the last section, we discuss the implications of these results. ",
        "conclusions": "Through studying the hard lag behaviors of the 0.5 -- 10 Hz QPOs, we have highlighted them into three types that are closely related to different temporal and spectral properties. The first type of QPOs (0.5 -- 2 Hz) have positive hard lags at the fundamental and first harmonic frequencies and were observed in the quiescent soft state. The second type of QPOs (2 -- 4.5 Hz) were also detected in the quiescent soft state but have opposite signs for the hard lags at the fundamental and first harmonic frequencies.  The third type (4.5 - 10 Hz), which showed up in medium soft quiescent/out-burst states, do not have significant harmonic peaks. There is a general trend that the hard lags at the QPO fundamental frequency decrease from positive to negative values as the fundamental frequency moves from 0.66 Hz to higher values. We also found that the photon spectra become softer as the QPO fundamental frequency increases. After we presented this work to the 195th AAS meeting in Atlanta on January 14th, 2000, it was brought to our attention that a recent paper by Reig et al. (2000) also reported such trends in GRS 1915+105.  The opposite signs of the hard lags at the QPO fundamental and first harmonic frequencies do not necessarily mean that the photon arrival times are different at different QPO harmonic frequencies. For example, the Fourier transform of a decaying oscillating signal $f (t) = e^{-\\lambda t} (cos(\\omega_0 t) + cos(2 \\omega_0 t)) $ is: \\begin{eqnarray} F(\\omega) & = & \\frac{1}{\\lambda - i(\\omega-\\omega_0)} + \\frac{1} {\\lambda - i(\\omega+\\omega_0)} \\nonumber \\\\ &   & + \\frac{1}{\\lambda - i(\\omega-2\\omega_0)} + \\frac{1}{\\lambda - i(\\omega + 2\\omega_0)} \\end{eqnarray} The power density has two QPO peaks at $\\omega_0$ and $ 2\\omega_0$ respectively. For $\\omega > 0$, the two dominant terms are: \\begin{equation} F(\\omega) \\approx \\frac{1}{\\lambda - i(\\omega-\\omega_0)}  + \\frac{1}{\\lambda - i(\\omega-2\\omega_0)} \\end{equation} So we have $F(\\omega=\\omega_0) \\approx \\frac{1}{\\lambda}  + \\frac{1}{\\lambda + i(\\omega_0)}$ and  $F(\\omega=2 \\omega_0) \\approx \\frac{1}{\\lambda}  + \\frac{1}{\\lambda - i(\\omega_0)}$. Therefore, $F(\\omega=\\omega_0) \\approx F(\\omega=2 \\omega_0)^* $, i.e. they have opposite signs for the phase term even though the oscillations have no time delays from each other. The two additional minor terms make the conclusion less obvious, but we still can find appropriate $\\lambda_0$ and $\\omega_0$ to make the phase lags have opposite signs. Therefore, it may be misleading to interpret the phase lag at a particular QPO frequency as simply the time delay between two oscillating signals.  In the classic multi-color thermal disk models, the negative hard lags can be explained by assuming that perturbations propagate from the inner disk to the outer disk. The perturbations on the inner edge may come from the central object or from the magnetic fields that are being sucked into the central object. Zhang, Cui, \\& Chen (1997) and Cui, Zhang, \\& Chen (1998) suggested that GRS 1915+105 contains a black hole rapidly spinning in the same direction as the accretion disk. In such a prograde system, the last stable orbit is much smaller than the non-spinning system. Therefore, strong perturbations on the inner edge of the accretion disk are very likely. There may also be perturbations propagating inwards from the outer edge of the disk, which generate positive hard lags. When the disk inner edge is farther away from the central object and thus the QPO frequency is smaller, it is natural  to assume that the perturbations on the inner edge is weaker. Therefore, the inward perturbations would dominate over the outward perturbations, and the hard lags are expected to be positive. This could explain why the lowest 0.5 -- 2 Hz QPOs have positive hard lags while the faster ones have negative hard lags.  Taam, Chen, \\& Swank (1997) proposed that the out-bursts in GRS 1915+105 are due to the ejection of the inner disk. A direct prediction of this model is that the inner disk edge is farther away from the central object in the out-burst state than in the quiescent state.  Our results may pose problems to this prediction. The observation made on July 5th, 1997, for example, had long out-burst and quiescent segments. We therefore performed separate timing analyses on the out-burst and quiescent segments. We found that the quiescent state has a lower QPO fundamental frequency than the out-burst state, and thus the disk inner edge should be closer to the central object in the out-burst state than in the quiescent state if we assume the QPO fundamental frequency is somehow related to the sound crossing time across the disk inner edge."
    },
    "0010/astro-ph0010413_arXiv.txt": {
        "abstract": "We present data for four ultra-Li-deficient, warm, halo stars. The Li deficiency of two of these is a new discovery.  Three of the four stars have effective temperatures $T_{\\rm eff}~\\sim~6300$~K, in contrast to previously known Li-deficient halo stars which spanned the temperature range of the Spite Plateau.  In this paper we propose that these, and previously known ultra-Li-deficient halo stars, may have had their surface lithium abundances reduced by the same mechanism as produces halo field blue stragglers.  Even though these stars have yet to reveal themselves as blue stragglers, they might be regarded as ``blue-stragglers-to-be.'' In our proposed scenario, the surface abundance of Li in these stars could be destroyed (a) during the normal pre-main-sequence single star evolution of their low mass precursors, (b) during the post-main-sequence evolution of a evolved mass donor, and/or (c) via mixing during a mass-transfer event or stellar merger.  The warmest Li-deficient stars at the turnoff would be regarded as emerging ``canonical'' blue stragglers, whereas cooler ones represent sub-turnoff-mass ``blue-stragglers-to-be.''  The latter are presently hidden on the main sequence, Li depletion being possibly the clearest signature of their past history and future significance.  Eventually, the main sequence turnoff will reach down to their mass, exposing those Li-depleted stars as canonical blue stragglers when normal stars of that mass evolve away.  Arguing {\\it against} this unified view is the observation that the three Li-depleted stars at $T_{\\rm eff}~\\simeq~6300$~K are {\\it all} binaries, whereas very few of the cooler systems show evidence for binarity; it is thus possible that two separate mechanisms are responsible for the production of Li-deficient main-sequence halo stars. ",
        "introduction": "%  $^7$Li is destroyed in stellar interiors where temperatures exceed $2.5\\times 10^6$~K, and Li-depleted material can in principle reach the stellar surfaces where it can be observed.  Thus, if one is to infer pre-stellar $^7$Li abundances from current-epoch observations, it is important to understand the stellar processing of this species. It has widely, though not universally, been supposed that warm ($T_{\\rm eff}~>~5700$~K), metal-poor ([Fe/H]~$<~-1$) stars retain their pre-stellar abundances (Spite \\& Spite 1982; Bonifacio \\& Molaro et al. 1997; but see also Deliyannis 1995; Ryan et al.  1996).  Although claims had been made of an intrinsic spread in the Li abundances by 0.04 -- 0.1 dex (Deliyannis, Pinsonneault, \\& Duncan 1993; Thorburn 1994), Ryan, Norris \\& Beers (1999) attributed these to an embedded $A$(Li) vs [Fe/H] dependence , and underestimated errors, respectively.  Ryan et al. (1999) set tight limits on the intrinsic spread of $^7$Li in metal-poor field stars as essentially zero, stated conservatively as $\\sigma_{\\rm int} < 0.02$~dex.  However, the subset of ultra-Li-deficient stars identified by Spite, Maillard, \\& Spite (1984), Hobbs \\& Mathieu (1991), Hobbs, Welty \\& Thorburn (1991), Thorburn (1992), and Spite et al. (1993) stands out as a particular exceptional counter-example to the general result.  These stars have only upper limits on their $^7$Li abundances, typically 0.5~dex or more below otherwise similar stars of the same $T_{\\rm eff}$ and metallicity.  Detailed studies of other elements in these objects have revealed some chemical anomalies, but none common to all, or which might explain {\\it why} their Li abundances differ so clearly from those of otherwise similar stars (Norris et al. 1997a; Ryan, Norris \\& Beers 1998).  In contrast to the situation for Population II stars, a wider range of Li behaviors is seen in Population I. In addition to a stronger increase with metallicity, thought to be due to the greater production of Li in later phases of Galactic chemical evolution (Ryan et al. 2001), there is also substantial evidence of Li depletion in certain temperature ranges. Open cluster observations, for example, show steep dependences on temperature for $T_{\\rm eff} \\la 6000$~K (e.g., Hobbs \\& Pilachowski 1988) and in the region of the F-star Li gap (6400~K~$<~T_{\\rm eff}~<$~7000~K; Boesgaard \\& Tripicco 1986). More problematic, for the young cluster $\\alpha$~Per (age 50 Myr) and the Pleiades (age 100 Myr), is the presence of a large apparent Li spread even at a given mass. Various explanations have been proposed involving mixing in addition to that due to convection. Extra mixing processes include rotationally-induced mixing (e.g., Chaboyer, Demarque \\& Pinsonneault 1995), structural changes associated with rapid rotation (Mart\\'in \\& Claret 1996), and different degrees of suppression of mixing by dynamo-induced magnetic fields (Ventura et al. 1998). Gravity waves have been proposed as yet another different mixing mechanism (Schatzman 1993; Montalb\\'an \\& Schatzman 1996). Consensus has not yet emerged concerning the range of possible mechanisms, or the relative importance of each.  Jeffries (1999) even questions the reality of a Li abundance spread in low mass Pleiades stars, due to a similar spread being seen in the \\ion{K}{1} resonance line. Amongst older open clusters, the spread at a given effective temperature is generally much less, though M67 (Jones, Fischer, \\& Soderblom 1999) is an exception. A class of stars with higher lithium abundances than otherwise similar stars is short-period tidally-locked binaries (Deliyannis et al. 1994; Ryan \\& Deliyannis 1995) which give credence to the view that physics related to stellar rotation can and does influence the evolution of Li in approximately solar-mass stars.  The fraction of warm, metal-poor stars that fall in the ultra-Li-deficient category has previously been estimated at approximately 5\\% (Thorburn 1994). However, recent measurements of Li in a sample of 18 warm ($T_{\\rm eff}~\\ga~6000$~K), metal-poor ($-2~\\la~$[Fe/H]~$\\la~-1$) stars yielded four ultra-Li-deficient objects, i.e. more than 20\\% of the sample (Ryan et al. 2001).  The Poisson probability of a 5\\% population yielding 4 or more objects in a sample of this size is just 0.013.  Clearly, the selection criteria for this sample have opened up a regime rich in ultra-Li-poor stars.  We now examine those criteria, and discuss the implications for the origin of such systems and for our understanding of Li-poor and Li-normal stars.  We note some similarities between Li-deficient halo stars and blue stragglers. Although these two groups have traditionally been separated due to the different circumstances of their {\\it discovery}, we question whether there is a reliable {\\it astrophysical} basis for this separation.  One must ask whether the process(es) that gives rise to blue stragglers is capable only of producing stars whose mass is greater than that of the main sequence turnoff of a $\\sim$13 Gyr old population. If, as we think is reasonable, the answer is ``no'', then one may ask what the sub-turnoff mass products of this process(es) would be. Our proposal is that they would be Li-deficient, but otherwise difficult to distinguish from the general population, and in this regard very similar to the ultra-Li-deficient halo stars.  \\section {Observations of the Ultra-Li-Poor Halo Stars} %  The ultra-Li-poor halo stars we consider were identified serendipitously in a study of predominantly high proper-motion halo stars having $T_{\\rm eff}~^>_\\sim~6000$~K and $-2~^<_\\sim$~[Fe/H]~$^<_\\sim~-1$, and are listed in Table~1(a).  Details of the sample selection and abundance analysis are given by Ryan et al. (2001); the key points are that high resolving power ($\\lambda/\\Delta\\lambda~\\simeq~50000$) \\'echelle spectra were obtained, equivalent widths were measured, and abundances were computed using a model atmosphere spectrum-synthesis approach.  Two of the Li-poor stars were subsequently found to have previous Li measurements; Wolf~550 was identified as G66-30, and G202-65 had been observed by Hobbs \\& Mathieu (1991) in a study targeted at blue stragglers.  The new spectra of the four stars, plus one with normal Li for comparison, are shown in Figure~1. The comparison star, CD$-31^\\circ$305, has $T_{\\rm eff}~=~5970$~K, [Fe/H]~=~$-1.0$, and $A$(Li)~=~2.24 (Ryan et al. 2001).  For convenience, previously known Li-depleted halo stars are listed in Table~1(b).  The full sample of Ryan et al. (2001) is plotted in Figure~2, along with additional stars from the literature.  It is immediately apparent that three of the four ultra-Li-deficient stars are amongst the hottest in our sample, though not {\\it the} hottest in the figure. It seems likely that high temperature is one biasing characteristic of these objects.  The stars with $T_{\\rm eff}~>~6300$~K and {\\it normal} Li abundances are listed in Table~1(c).  These have had comparatively high values of de-reddening applied, and it is possible that they are in reality cooler than Table~1 shows. An indication that high temperature is not the {\\it only} biasing characteristic of ultra-Li-poor stars is that the Ryan et al. (1999) study of 23 very metal-poor ($-3.5~^<_\\sim$~[Fe/H]~$^<_\\sim~-2.5$) stars in the same temperature range included only one ultra-Li-deficient star, G186-26. This rate, 1 in 23, is consistent with previous estimates for Population II stars as a whole.  However, very few relatively metal-rich ($-2~^<_\\sim $~[Fe/H]~$^<_\\sim~-1$) halo stars in this temperature range had been studied previously, so earlier works may have been biased against discovering ultra-Li-poor objects.  It appears, then, that the fraction of ultra-Li-deficient stars is higher as metallicity increases.  This may explain why our study, which targeted stars in the higher metallicity range {\\it and} with $T_{\\rm eff}~>~6000$~K, was so successful at yielding ultra-Li-deficient stars.  Figure~3 shows the distribution of objects in the $T_{\\rm eff}$,~[Fe/H] plane.  Whereas previously no region of parameter space stood out as ``preferred'' by Li-deficient stars, the objects are now more conspicuous as a result of their high temperatures and relatively high metallicities.  Also shown in Figure~3 are the $T_{\\rm eff}$ of the main-sequence turnoff as a function of metallicity, for 14 and 18~Gyr isochrones. The isochrones shown are the oxygen-enhanced curves of Bergbusch \\& VandenBerg (1992; solid curves; Y=0.235), and, for comparison, the Revised Yale Isochrones of Green, Demarque \\& King (1987; dotted curves; Y=0.24). Clearly there is disagreement of $\\simeq$4~Gyr between the two sets as to the ages that would be assigned to these stars, and there are uncertainties in the color-$T_{\\rm eff}$ transformations that have been applied to the observed data. However, these difficulties are not the issue here. Rather, we use the isochrones to indicate the {\\it shape} of the turnoff locus in the $T_{\\rm eff}$ vs [Fe/H] plane, and on that point the four isochrones are in overall agreement. They emphasise that even though HD~97916 is cooler than five other Li-depleted stars in the study, it is nevertheless close to the turnoff. That is, a star with $T_{\\rm eff}$ = 6100~K would appear below the turnoff if [Fe/H] = $-3$, but will be close to the turnoff if [Fe/H] = $-1$. Even excluding the definite blue straggler BD+25$^\\circ$1981, there are four Li-depleted stars amongst the eight whose symbols lie above or touch the 14~Gyr Revised Yale Isochrone. Clearly, all of these are very close to the turnoff once their metallicities are taken into account.\\footnote{ We resist the temptation to speak of a {\\it single} locus for the turnoff because of the possibility that an age spread exists at a given metallicity.  That issue has not yet been settled for the globular clusters (see Piotto et al. 2000 and Chaboyer 2000), despite those systems being better constrained. For the same reason, and because of random errors in the effective temperature estimates, we refrain from debating whether a particular star lying close to the turnoff is definitely above or below the turnoff.} Besides these Li-depleted stars close to the turnoff, four are 100--200~K cooler than the turnoff. We discuss later in this paper whether the these two groupings might have different origins.  \\section {Traditional blue stragglers} %  Blue stragglers are recognised observationally as stars that are considerably bluer than the main-sequence turnoff of the population to which they belong, but having a luminosity consistent with main-sequence membership.  Such objects were originally identified in globular clusters (e.g., M3; Sandage 1953), but are also known in the field (e.g., Carney \\& Peterson 1981), and in Population I as well as Population II (e.g., Leonard 1989; Stryker 1993). Their origin is not known with certainty, and it is possible that more than one mechanism is responsible for their presence.  A range of explanations was examined by Leonard (1989), but the discovery of Li destruction in blue stragglers in the halo field and the open cluster M67 led Hobbs \\& Mathieu (1991) and Pritchet \\& Glaspey (1991) to conclude that ``virtually all mechanisms for the production of blue stragglers {\\it other than} mixing, binary mass transfer, or binary coalescence appear to be ruled out ... .'' As Hobbs \\& Mathieu emphasized, internal mixing alone is also ruled out; mixing out to the surface is required.  Recent advances in high-resolution imaging have verified that the blue straggler fractions in at least some globular clusters are higher in their cores, strongly supporting the view that some blue stragglers are formed through stellar collisions, probably involving the coalescence of binary stars formed and/or hardened through exchanges, in these dense stellar environments (e.g., Ferraro et al. 1999).  However, it is neither established nor required that a single mechanism will explain all blue stragglers, and it is unclear how the field examples and those in the tenuous dwarf galaxy Ursa Minor (Feltzing 2000, priv. comm.) relate to those in the dense cores of globular clusters.  Probably even the halo field and dwarf spheroidal stars formed in clusters of some description (since the formation of stars in isolation is unlikely), but one should not be too quick to link the properties of surviving globular clusters to diffuse populations. This view is supported by Preston \\& Sneden's (2000) conclusion that at more than half (62\\% -- 100\\%) of their field blue metal-poor binaries are blue stragglers formed by mass transfer rather than mergers, due to the long orbital periods and low eccentricities of the field systems they observed.  Their conclusion is entirely consistent with the views of Ferraro et al. (1995), who ascribed blue straggler formation to interactions {\\it between} systems in high-density environments, but {\\it within} systems (primordial binaries) in lower-density clusters.  In contrast to but not contradicting Preston \\& Sneden's result for field systems, Mateo et al.  (1990) argue that all of the blue stragglers in the globular cluster NGC~5466 are the result of close binary mergers.  The mechanism for Li destruction in field blue stragglers is not known.  It is unclear what degree of mixing will occur as a result of coalescence. Early work by Webbink (1976) suggested substantial mixing would occur, whereas more recent simulations of head-on collisions by Sills et al. (1997), and grazing collisions and binary mergers by Sandquist, Bolte, \\& Hernquist (1997), have suggested otherwise. Sills, Bailyn \\& Demarque (1995) argue, however, that to account for the blue stragglers observed in NGC~6397, mixing is nevertheless required (unless the collision products have more than twice the turnoff mass), and may occur after the initial coalescence. This is perhaps consistent with the result of Lombardi, Rasio, \\& Shapiro (1996) that some mixing could occur as a merger remnant re-contracts to the main sequence.  Due to the fragility of Li, if some mixing of surface material does occur during the coalescence it will at least dilute, and possibly also destroy, any lithium remaining in the stars' thin convective surface zones up to that time. One might suppose that mass transfer in a detached system also destroys Li, though one could also imagine gentle mass-transfer processes where the rate is slow enough that the original envelope is not subjected to additional mixing, and where the transferred matter itself does not undergo additional Li-destruction.  Of course, mass transfer via Roche lobe overflow in a detached system, or wind accretion from a more distant companion, involve mass from an evolved star which may have {\\it already} depleted its surface Li due to single-star evolutionary processes.  Consequently, the mass transferred may be already devoid of Li, as in the scenario quantified by Norris et al. (1997a).  We also note the possibility that the accretor in a mass-transfer system, or the progenitors of a coalescence, was (were) devoid of Li prior to that event. Li is (normally) preserved in halo stars only over the temperature range from the turnoff ($T_{\\rm eff}~\\simeq~6300$~K) to about $T_{\\rm eff}~\\simeq~5600$~K, corresponding to a mass range from 0.80 down to 0.70~$M_\\odot$. Therefore it is likely that any mass accretor, and certain that any merger remnant, now seen in this mass range began life as one (or two) stars with initial mass(es) less than 0.70~$M_\\odot$ and had already destroyed Li normally, as lower-mass stars are known to do, prior to mass exchange. In such a scenario, it is not {\\it necessary} for any Li to have been destroyed as a result of the blue-straggler formation process itself, though this could occur as well.    \\section {Discussion} %  In view of the distributions of the ultra-Li-deficient stars in the $T_{\\rm eff}$,~[Fe/H] plane, with four at the turnoff and four 100--200~K cooler, we consider whether all represent the same phenomenon, or the possibility that two distinct processes have been in operation.  It is not a trivial matter to answer this question, because we do not know with certainty what mechanism(s) has affected any of the stars. However, we explore a number of possibilities in the discussion that follows. Ignoring again the obvious blue straggler BD+25$^\\circ$1981, of the 111 stars shown in Figure~3, 8 are ultra-Li-deficient.  If all ultra-Li-poor stars have the same origin, then we should begin by restating the frequency of such Li-weak objects as $\\simeq$7\\% of plateau stars rather than $\\simeq$5\\% as estimated previously when the parameter space was incompletely sampled, and with strong metallicity and temperature dependences in that fraction.  \\subsection {Do Ultra-Li-Deficient Stars and Field Blue Stragglers Share a Common Origin?}  Historically, blue stragglers and ultra-Li-deficient stars have been regarded as separate phenomena. However, we have been driven to consider whether there is any astrophysical basis for this separation.  One must ask whether the process(es) that gives rise to blue stragglers is capable only of producing stars whose mass is greater than that of the main sequence turnoff of a $\\sim$13 Gyr old population. If, as we think is reasonable, the answer is ``no'', then one may ask what the sub-turnoff mass products of this process(es) would be. Our proposal is that they would be Li-deficient, but otherwise difficult to distinguish from the general population.\\footnote{ The likelihood of sub-turnoff mass objects being produced by the blue-straggler forming process is independently addressed in the model by Preston \\& Sneden (2000, \\S 5.3), which came to our attention during finalisation of this manuscript.}  For ultra-Li-poor stars redder than the main sequence turnoff, Hipparcos parallaxes have established that G186-26 is on the main sequence rather than on the subgiant branch. Of those {\\it at} the turnoff, Wolf~550, G202-65, and BD+51$^\\circ$1817 also have Hipparcos parallaxes; two are almost certainly dwarfs, while G202-65 is subject to larger uncertainties and may be more evolved (see Ryan et al. 2001, Table~2). The argument that the evolutionary rate of subgiants is too rapid to explain the high frequency of observed Li-deficient objects, which persuaded Norris et al. (1997a) to reject the proposition that they might be the {\\it redward}-evolving (post-turnoff) progeny of blue-stragglers, is therefore redundant. However, the detection of several Li-weak stars at the bluest edge of the colour distribution has prompted us to re-examine their possible association with blue stragglers.  We would describe G202-65 as ``at'' the turnoff rather than classify it as a blue straggler in the conventional sense, as it is only marginally hotter (bluer) than the main sequence turnoff for its metallicity (see Figure~3). Hobbs \\& Mathieu, on the other hand, classified it as a blue straggler, based presumably on the photometry of Laird, Carney \\& Latham (1988) which they referenced. (Indeed, Carney et al (1994) declare it as a ``blue straggler candidate'', and Carney et al. (2000) treat it as one, though acknowledging at the same time that some normal stars may be included in this classification.) Our purpose is {\\it not} to debate how this star should be classified, but rather to underline the main suggestion of our work, that the blue straggler and halo ultra-Li-deficient stars may have a common origin.  Although blue stragglers have historically been recognised because they are bluer than the main-sequence turnoff, it is essential to remember that stars that have accreted mass from a companion, or that result from a coalescence can have a mass less than the current turnoff.  Such stars would be expected to share many of the properties of blue stragglers, but would not {\\it yet} appear bluer than the turnoff. However, at some future time, once the main-sequence turnoff reaches lower masses, these non-standard objects would lag the evolution of normal stars and hence appear bluer, showing canonical straggling behaviour.  Therefore, such stars might, for the present, be regarded as ``blue-stragglers-to-be,''\\footnote{Independently, Carney et al. (2000) have noted this possibility, and models by Portegies Zwart (2000) predict the existence of such objects.} and our speculation is that the ultra-Li-deficient halo stars in are in fact members of such a population. Note that this proposition is distinct from that of {\\it redward}-evolving systems considered and rejected by Norris et al.  (1997a).   If ultra-Li-deficient stars and blue stragglers are manifestations of the same process, then Li deficiency may be the only way of distinguishing sub-turnoff-mass blue-stragglers-to-be from normal main-sequence stars, prior to their becoming classical blue stragglers.  Mass transfer during their formation may also help clarify some of the unusual element abundances found by Norris et al. (1997a; see also Ryan et al. 1998).  Whereas an appeal to extra mixing (in a single-star framework) to explain the Li depletion would not necessarily affect other elements, mass transfer in a binary with an AGB donor may be capable of altering s-process abundances as well.  In this regard, we recall that two of the ultra-Li-deficient stars studied by Norris et al. (1997a; also Ryan et al. 1998) had non-standard Sr and Ba abundances. Mass transfer from an RGB donor would presumably leave a different chemical signature.\\footnote{Amongst very metal-poor stars with [Fe/H] $<$ $-2.5$, as many as 25\\% have C overabundances (e.g. Norris, Ryan, \\& Beers 1997b). At least some but not all of these (Norris, Ryan, \\& Beers 1997c) have s-process anomalies. Detailed studies have yet to be completed, so it is unclear what fraction of stars are formed from anomalous material and what fraction became modified later in their life. Whilst we cannot presently rule out the possibility that the s-process anomalies seen in some ultra-Li-deficient stars were inherited at birth, our expectation is that mass transfer from a companion star will be a more common mechanism.}   Some constraints on the progenitors of the Li-deficient stars may be obtained from their rotation rates and radial velocity variations.  Webbink's (1976) calculations of a coalesced star ($M_{\\rm total}~=~1.85~M_\\odot$) show that a high rotation rate is maintained at least until it reaches the giant branch. In contrast, previously known blue stragglers appear not to have uncommonly high rotation rates (e.g., Carney \\& Peterson 1981; Pritchet \\& Glaspey 1991). This tends to argue against the blue stragglers as having originated from coalesced main-sequence contact binaries, and points towards one of the other binary mass-transfer scenarios, unless mass loss (e.g., via Webbink's excretion disk) and magnetic breaking can dissipate envelope angular momentum during the main sequence lifetime of a coalesced star. To spin down, stars must have a way of losing surface angular momentum. In single stars, most of this is believed to occur during the pre- and early-main-sequence phase when magnetic coupling of the stellar surface to surrounding dust creates a decelerating torque on the star. It is not clear that two mature stars which merge will still have this coupling, because of the much lower mass loss rates beyond the early stages of evolution (unless they produce an excretion disk) and lower magnetic field strengths. (See also discussion by Sills et al. 1997, \\S5.5.) Leonard \\& Livio (1995) have proposed that the merger product acquires the distended form of a pre-main-sequence-like star which then spins down as it again approaches the main sequence, losing angular momentum in much the same way as conventional pre-main-sequence stars. \\footnote{Although stellar collisions will be rare for stars in the field, we should recall that most stars are probably born in clusters, and prior to cluster dissolution, collisions would have greater probability.}  For the four stars observed in this work, three had previous radial velocity measurements accurate to $\\simeq$~1~km~s$^{-1}$ (Carney et al. 1994). The new measurements (Ryan et al. 2001; Table 2) showed residuals of +1.0 (BD+51$^\\circ$1817), $-$3.3 (G202-65), and $-$6.9~km~s$^{-1}$ (Wolf~550); compared with the expected radial velocity accuracy of $\\sigma_v$~=~0.3--0.7~km~s$^{-1}$, these are consistent with significant motion. Carney et al. (2000) indicate periods of 168 to 694~days for these systems, and low eccentricities, except for Wolf~550 ($e$~=~0.3).  Similarly, the metal-poor field blue straggler CS~22966-043 has an orbital period of 319~days (Preston \\& Landolt 1999). If the brighter component has a mass of 0.8~M$_\\odot$ and its companion has a mass between 0.4 and 1.4~M$_\\odot$ (appropriate to a white dwarf) then the {\\it current} semi-major axis of the system will be in the range $a$~=~200--260~R$_\\odot$ (from Kepler's Third Law).\\footnote{Carney et al. (2000) argue that all of their blue-straggler observations are consistent with 0.55~M$_\\odot$ companions having a canonical white-dwarf mass.} Their second system, CS~29499-057, may have an even longer period of 2750~days, implying $a$~=~900--1100~R$_\\odot$. The periods of these and Carney et al's systems, and hence their large current separations, are more compatible with mass loss from an evolved companion rather than being short-period systems in contact on the main sequence.  The evidence presented to date has argued against internal mixing alone as an adequate explanation for the ultra-Li-deficient stars whose neutron-capture elements show abundance anomalies. Note, though, that certainly not all ultra-Li-deficient stars and blue stragglers exhibit neutron-capture element anomalies (Carney \\& Peterson 1981; Norris et al. 1997a; Ryan et al. 1998).  If mass transfer has occurred, systems in which s-process elements are abnormal would presumably indicate material originating with an AGB companion, whereas s-process-normal remnants would indicate mass transfer during an earlier stage of evolution (RGB) or from a pre-thermal-pulsing AGB mass donor. (We have no data on the N abundance, and the CH band in these stars is too weak to hope to measure the $^{12}$C/$^{13}$C ratio.) Likewise, the rotation rates of both blue stragglers and ultra-Li-deficient stars are apparently normal, arguing against coalescences having already occurred on the main sequence. Of the three mechanisms found to be viable by Pritchet \\& Glaspey (1991) and Hobbs \\& Mathieu (1991), this leaves mass transfer from a companion as the only one remaining, {\\it if} we are correct in speculating that the ultra-Li-deficient and blue straggler phenomena are manifestations of the same process.  \\subsection {The Hot Stars in Isolation} %  In the absence of an adequate theory for why eight otherwise-normal halo stars (excluding the traditional blue straggler BD+21$^\\circ$1981) should have low (zero?) Li abundances, it may be useful to consider the hot subsample (6200~K~$^<_\\sim$~$T_{\\rm eff}$~$^<_\\sim$~6300~K) as a distinct group. Several possibilities then arise that might account for the observed Li deficiency, including diffusion (the sinking of Li to below the photosphere), the F-star Li dip, and an unknown process that may be responsible for depletion in some (but not all) disk stars.  We consider each of these in turn.  We note that the three Li-deficient stars with $T_{\\rm eff}~\\simeq~6300$~K are confirmed binaries, whereas most cooler ones show no evidence of binary motion.  The binary/single distinction between warmer/cooler Li-depleted stars is pronounced; see Table~1, where the binary status (Carney et al. 1994, 2000; Latham 2000, priv.comm.) is given in the final column. If such a dichotomy is maintained as more Li-poor systems are discovered, it may indicate a genuine difference in the origin of the turnoff and sub-turnoff systems.  \\subsubsection{Diffusion} %  Deliyannis, Demarque \\& Kawaler (1990) and Proffitt \\& Michaud (1991) have computed the predicted effects of diffusion on the surface Li abundances of warm halo stars.  Diffusion is more significant in hotter stars because their surface convective zone is thinner.  The degree of depletion expected at $T_{\\rm eff}~\\sim~6300$~K is a function of effective temperature, changing by $\\simeq$~0.2~dex per 100~K in the former (for $\\alpha$~=~1.1), and $\\simeq$~0.2 and $>~0.2$~dex per 100~K in the latter (for $\\alpha$~=~1.7 and 1.5 respectively). This does not match the behavior observed (see Figure 2). For comparison, our ultra-Li-poor stars are depleted by $^>_\\sim0.8$~dex. This alone appears to rule out diffusion as the explanation, except possibly for the lower-$\\alpha$ model of Proffitt \\& Michaud. However, Li diffusion appears to have been inhibited in all other metal-poor samples (e.g., Ryan et al. 1996), so it would be unusual to see it suddenly present and with such effect only in isolated stars in our new sample.  \\subsubsection {The F-Star Li Dip} %  Boesgaard \\& Tripicco (1986) and Hobbs \\& Pilachowski (1988) showed that Li is severely depleted in Population I open cluster stars over the interval 6400~K~$<~T_{\\rm eff}~<~7000$~K. Various explanations have been proposed, including mass loss (e.g., Schramm, Steigman, \\& Dearborn 1990), diffusion (e.g., Turcotte, Richer \\& Michaud 1998), and slow mixing of various forms (e.g., Deliyannis \\& Pinsonneault 1997), but none has been convincingly established as responsible, and several mechanisms may be acting in concert (e.g., Turcotte et al.).  Whatever the correct explanation(s), is it possible that the hottest ultra-Li-deficient stars are encroaching on this regime and are affected by this phenomenon?  Although this cannot be ruled out completely for the hot subset, especially since we have questioned the reliability of the $E(B-V)$ (and hence $T_{\\rm eff}$) values of the hottest Li-preserving stars in Figure~2, the onset of destruction in the F-star dip seems too gradual with $T_{\\rm eff}$ to explain the new data. The Hyades observations (Boesgaard \\& Tripicco 1986) show a decrease of only 0.3~dex from 6200 to 6400~K, substantially less than the $^>_\\sim 0.8$~dex deficit in the ultra-metal-poor objects around 6300~K.\\footnote{The critic could object that there are deficiencies in comparing metal-rich and metal-poor objects in this fashion. We would agree, but would also note that such a comparison is justifiable if only to show that the two behaviors are dissimilar.} As noted above, Hipparcos parallaxes are available for five of the eight known ultra-Li-deficient stars and, with the possible exception of G202-65, rule out the possibility that these stars are redward-evolving {\\it descendants} of the Li-dip.  \\subsection {Anomalously-Li-Depleted Disk Stars}  Lambert, Heath \\& Edvardsson (1991) found that, in almost all cases, the low Li abundances in their Population~I sample could be ascribed to their being evolved descendants of Li-dip stars, or else being dwarfs exhibiting the Li-depletion that increases towards {\\it lower} temperature, as is normally associated with pre-main-sequence and/or main-sequence burning.  Anomalously high Li depletions were found in only 1--3 cases out of some 26 old-disk stars, and for a similar fraction of young-disk stars.  Based on this fraction, Lambert et al. proposed that a new class of highly Li-depleted stars, comprising less than about 10\\% of the population, might exist.  It is interesting to note that this proposal pre-dated the discovery of ultra-Li-deficient halo dwarfs.  The uncertain number of cases stated above arises because Lambert et al. recognised that uncertainties in the stellar luminosities, and hence mass, could drive stars into or out of the region of importance.  We now have the benefit of accurate Hipparcos parallaxes.  These indicate that two of the seven stars highlighted by their study, HD~219476 and HR~4285,  are indeed considerably more massive than reported in Lambert et al.'s tables and hence are probably descendants of the Li gap, thus reducing the number of {\\it genuine} cases to 2 out of 26 old-disk stars, and 3 out of a similar number of young-disk stars.  That is, the fraction of anomalously Li-depleted stars appears to be around 8-10\\%, albeit sensitive to small-number statistics. \\footnote{Errors in temperature could reduce these cases further.} Ultra-Li-depleted Population I stars are also seen in young open clusters. They can be recognised, for example, in Fig.~1 of Ryan \\& Deliyannis (1995), where $\\simeq$6\\% of the Hyades stars cooler than the F-star dip  appear to be ultra-Li-deficient.   Is it possible that the Li-depleted halo stars are of the same type?  The lack of examples in the two Pop~I and Pop~II classes to compare with precludes a detailed analysis, but we note that we see Li deficiency in about 7\\%\\ of halo objects, which is comparable to the ratio for the Pop~I objects. That is, the Pop~I and Pop~II examples could arise due to the same process, even though it remains unclear what that process is.  We note, for completeness, that Ryan et al.  (2001) showed that the kinematics of the new ultra-Li-depleted stars are clearly those of halo objects, and thus they genuinely belong to the halo Population, despite their metallicities being close to those of the most metal-poor thick-disk stars.  The stars remaining on Lambert et al's list of unusually Li-deficient objects are: HR~3648, HR~4657, HR~5968, HR~6541, and HD~30649. Upon searching the literature for evidence of binarity or abundance anomalies in these systems, we found that not only was HR~4657 a 850~day period binary, but Fuhrmann \\& Bernkopf (1999) had also been driven to consider this star as a blue straggler. It has an unexpectedly high rotational velocity (in contrast to the blue stragglers studied by Carney \\& Peterson 1981).  There is no evidence of s-process anomalies, but other unusual characteristics of the system include an observable soft X-ray flux and the very likely association of this object with GRB 930131.  HR~3648 (= 16~UMa = HD~79028) is a 16.2~day period chromospherically-active single-lined spectroscopic binary (Basri, Laurent, \\& Walter 1985).  HD~30649 (= G81-38) and HR~6541 (=HD~159332), in contrast, show no significant evidence of binarity (Carney et al. 1994).  HR~5968 (= $\\rho$ CrB) does not appear to have a stellar companion, though it has a planetary companion (Noyes et al. 1997), but Ryan (2000) argues that Li in this star is {\\it not} anomalous.  HR~3648 and HR~4657 have Ba abundance measurements from the study by Chen et al. (2000).  The latter also has been observed by Fuhrmann \\& Bernkopf (1999), but neither star appears abnormal in this element.  \\section {Implications and Summary} %  Ryan et al. (1999) have argued that the ultra-Li-deficient halo stars are distinct from the majority of halo stars that occupy the Spite plateau, and, in particular, that they do {\\it not} merely represent the most extreme examples of a {\\it continuum} of Li depletion. If the association with blue stragglers (or, for that matter, any distinct evolutionary phenomenon) is correct, then the mechanism for their unusual abundances will at last be understood and they will be able to be neglected with certainty from future discussion of the Spite plateau.  In the present work, we have proposed and discussed the possibility that ultra-Li-depleted halo stars and blue stragglers are manifestations of the same phenomenon, and described the former as ``blue-stragglers-to-be.''  We proposed that their Li was destroyed either during the formation process of blue stragglers or during the {\\it normal} single-star evolutionary processes of their precursors, namely during pre-main-sequence and/or main-sequence phases of low-mass stars, or during post-main-sequence evolution of mass donors, as in the scenario quantified by Norris et al. (1997a).  We note that in a study carried out separately but over the same time period as ours, Carney et al. (2000) have examined the orbital characteristics of blue stragglers, and have been driven towards similar considerations as we have. There are clearly still details to be clarified, but our two groups appear to be converging on a view unifying blue stragglers and ultra-Li-deficient systems.  Because there are numerous observational and theoretical issues surrounding this unified view, we seek to clarify the main arguments and possibilities using an itemised summary.  Observations: \\newline $\\bullet$ % In a study of 18 halo stars with $-2~^<_\\sim$~[Fe/H]~$^<_\\sim~-1$ and 6000~K~$^<_\\sim~T_{\\rm eff}~^<_\\sim$~6400~K, we have found four ultra-Li-deficient objects, i.e. a 22\\% detection rate. \\newline $\\bullet$ % The fraction of ultra-Li-deficient stars is very much higher amongst the hottest and most metal-rich halo main-sequence stars ($\\simeq$20\\%) than amongst cooler and more metal-poor ones ($\\simeq$5\\%). \\newline $\\bullet$ % Ultra Li-deficient stars exist both at the turnoff, and cooler than the turnoff, and with well-determined main-sequence luminosities from Hipparcos. \\newline $\\bullet$ % All of the turnoff ultra-Li-deficient halo stars, but none of the sub-turnoff ultra-Li-deficient halo stars, appear to be binaries. This may indicate that two different mechanisms are causing the halo ultra-Li-deficient phenomenon.  Theoretical framework: \\newline $\\bullet$ % Blue stragglers may form from {\\it several} mechanisms, but seem to require at least one of either complete mixing, binary mass transfer, or coalescence\\footnote{Coalescence may be between the components of an existing binary, possibly having been hardened via interactions with a third star, or through direct collisions (which may also be moderated by binary interactions).} (Hobbs \\& Mathieu 1991; Pritchet \\& Glaspey 1991).  Origins: \\newline $\\bullet$ % We speculate that ultra-Li-deficient stars and blue stragglers are manifestations of the same process, and that sub-turnoff-mass ultra-Li-deficient stars may be regarded as ``blue-stragglers-to-be.'' \\newline $\\bullet$ % Li could be destroyed at several stages: (i) in a mass-transfer event which induces extensive mixing; (ii) by single-star evolutionary processes (convective mixing) in a post-main-sequence mass donor; (iii) by single-star evolutionary processes (mixing) in pre-main-sequence (or possibly main-sequence) low-mass stars prior to their gaining mass. \\newline $\\bullet$ % Mass-transfer scenarios from an AGB star seem better able to explain the unusual neutron-capture element ratios {\\it sometimes} seen in ultra-Li-depleted stars (Norris et al. 1997a) than internal mixing, since $\\simeq$~0.8~M$_\\odot$ core-hydrogen-burning stars are not expected to process neutron-capture elements. This argues against internal mixing as the sole explanation for the existence of ultra-Li-depleted stars with unusual neutron-capture abundances. (Mass transfer from pre-AGB (most likely RGB) donors would produce the stars with normal neutron-capture abundances.) \\newline $\\bullet$ % Coalesced binaries are expected to maintain high rotation rates until they reach the giant branch, but neither blue stragglers nor ultra-Li-depleted halo stars have high rotation rates. This argues against coalescence of a binary as the explanation for these objects unless they have spun down. \\newline $\\bullet$ % The orbital periods of metal-poor field blue stragglers (Preston \\& Landolt 1999; Carney et al. 2000) suggest current semi-major axes in the range 200--1100~R$_\\odot$, arguing against these being coalescing stars (unless they began their lives as triple systems). \\newline $\\bullet$ % The arguments against solely internal mixing, and against coalescence of main-sequence contact binaries, leaves mass transfer as the most viable mechanism for field binaries.  This is {\\it not} to say that Li was destroyed during the transfer; it may have been destroyed by single-star mechanisms already. \\newline $\\bullet$ % The observed d$A$(Li)/d$T_{\\rm eff}$ is too steep compared with models of diffusion to be due to that process. \\newline $\\bullet$ % The observed d$A$(Li)/d$T_{\\rm eff}$ is too steep compared with the Hyades data to be due to the F-star Li dip. \\newline $\\bullet$ % The halo ultra-Li-deficient stars could be related to the Pop I anomalously-Li-depleted stars identified in the field by Lambert et al. (1991) and also seen in open clusters. \\newline $\\bullet$ % Hipparcos parallaxes rule out the possibility that the ultra-Li-deficient stars are redward-evolving post-turnoff stars. They have not descended from the F-star Li dip.  Implications: \\newline $\\bullet$ % Severe Li depletion may be the (only?) signature of sub-turnoff-mass blue stragglers. The halo population fraction comprising ultra-Li-poor stars is 7\\%. \\newline $\\bullet$ % Understanding the ultra-Li-depleted stars as resulting from a distinct process (not normally affecting single stars) would eliminate the need to include them in discussions of processes affecting the evolution of normal Spite plateau stars, and would explain why they appear so radically different from the vast majority of halo stars (Ryan et al. 1999). ",
        "conclusions": ""
    },
    "0010/astro-ph0010625_arXiv.txt": {
        "abstract": "We discuss the various contributions to the jet luminosity in proton blazar models of active galactic nuclei and describe a method of estimating the jet luminosity from the observed spectral energy distribution (SED) and the fitted model parameters.  We apply this to a synchrotron proton blazar (SPB) model for Markarian 501. ",
        "introduction": "The mechanical luminosity of an AGN jet of cross-sectional area $A$ with Lorentz factor $\\Gamma=1/\\sqrt{1-\\beta^2}$, containing jet-frame (primed variables) matter density density $\\rho'$, energy density $u'$ and pressure $p'$ is given by \\begin{eqnarray*} L_{\\rm jet,\\, mech} &=& \\Gamma^2 \\beta c A[\\rho'c^2(\\Gamma-1)/\\Gamma + u'+p']. \\end{eqnarray*} (see e.g. Leahy 1991).  To calculate the total jet luminosity $L_{\\rm{jet}}$, measured in the rest frame of the galaxy, we adapt the formulae of Bicknell (1994) and Bicknell \\& Dopita (1997), given for the synchrotron self-Compton model, to apply for the case of proton blazar models.  In this paper we shall apply the formula adapted for proton blazar models to a synchrotron proton blazar (SPB) model for Markarian 501 described by M\\\"ucke \\& Protheroe (2000). ",
        "conclusions": ""
    },
    "0010/nucl-th0010075_arXiv.txt": {
        "abstract": "We study the role of Coulomb and surface effects on the phase transition from dense nuclear matter to a mixed phase of nuclear and kaon-condensed matter.  We calculate corrections to the bulk calculation of the equation of state (EOS) and the critical density for the transition by solving explicitly for spherical, cylindrical and planar structures.  The importance of Debye screening in the determination of the charged particle profiles is studied in some detail.  We find that the surface and Coulomb contributions to the energy density are small but that they play an important role in the determination of the critical pressure for the transition, as well as affecting the size and geometry of favored structures. This changes the EOS over a wide range of pressure and consequently increases the maximum mass by $\\sim 0.1$ M$_\\odot$. Implications for transport properties of the mixed phase are also discussed. ",
        "introduction": "The effect of Debye screening on the radial dependence of the particle densities was shown in Fig. [\\ref{profile}].  Charged particle profiles had to readjust quite significantly to lower the coulomb energy. Clearly, the effect of the deformation of particle profiles caused by screening is to increase the hadronic part of the energy density associated with the structure.  This must occur because the fields no longer take on their optimal values over a sizable region of the structure.  The net effect of screening, however, is to lower the total energy, since the reduction in the Coulomb energy due to screening is larger than the cost in hadronic energy associated with the deformation.  The qualitative effects of screening on the Gibbs energy are shown in Fig. [\\ref{screen}]. The results illustrate the importance of the Coulomb energy relative to other contributions.  For example, the differences in Gibbs energies among unscreened droplets of different sizes are small compared to the differences between unscreened droplet and screened droplets of any size. Screening also results in a qualitative change in the radius dependence of the Gibbs free energy. The Gibbs free energy of the screened droplets decreases with increasing radius for small to intermediate size droplets.  \\begin{figure}[tbh] \\centering { \\epsfig{figure=screen.eps, width=.85\\textwidth} } \\caption{ Gibbs energy per baryon as a function of droplet radius, for different amounts of screening.  The screening is controlled by shifting the electrochemical potential at each radius by a fraction of the true electric potential.  By turning on the screening slowly, one clearly sees the qualitative effect of the overall energy per baryon being lowered due to the reduction in Coulomb energy.} \\label{screen} \\end{figure}  How are the results of the previous section affected by screening ?  To answer this, we must first point out another qualitative effect of screening which again involves electrical charge neutrality.  As shown in Fig. [\\ref{profile}], Coulomb forces tend to repel the negatively charged electrons away from the kaon droplet.  Whenever the Wigner-Seitz radius is large (which can happen either when the kaon structure itself is large, or when the kaon filling fraction is low) this repulsion results in an over-abundance of electrons in the region outside of the kaon structure.  But as before, once the electric charge density of the nucleonic phase becomes negative, it is no longer possible to define a Wigner-Seitz cell, and the structure under consideration is forbidden.  We find that precisely this problem occurs for all droplets below a pressure of about $p=.35$ fm$^{-4}$.  Above this pressure, where it is possible to construct a fully screened globally electrically neutral droplet, the Gibbs energy per baryon is indeed lowered somewhat from the values plotted in Fig. [\\ref{gibbs}].  This results in the critical pressure decreasing slightly from the new value quoted earlier, but the change here is small.  The first fully screened structure we can produce with Gibbs energy per baryon below that of the pure nucleonic phase occurs at a pressure of $.375$ fm$^{-4}$ (compare this with the critical pressure $\\sim 0.385$ fm$^{-4}$ when Debye screening was ignored).  Thus, while the Debye screening can have large effects on the particle density profiles, it has a relatively small effect on the free energy density (less than one MeV per baryon), as can be seen in the right panel of Fig. \\ref{screen}.  Consequently the results of the previous section remain qualitatively unchanged: the inclusion of finite-size effects significantly increases the critical pressure for transition to a mixed phase.  Because the changes in Coulomb energy associated with screening are large compared to the energy differences between various geometries at a given pressure, screening will also play a dominant role in the determination of structures that are preferred at a given pressure.  The qualitative feature of droplets being replaced by rods (and subsequently slabs) as the pressure is increased is expected to remain.  But the particular pressures at which the transition from one geometry to the next occurs may vary from the picture of Fig. [\\ref{gibbs}]. Thus, if one is only interested in the long wavelength thermodynamic properties of the system, it is safe to ignore Debye screening. However, to describe the mixed phase in more detail, i.e, specify the particle profiles and the geometries that are favored at a given pressure Debye screening must be accounted for. ",
        "conclusions": ""
    },
    "0010/astro-ph0010631_arXiv.txt": {
        "abstract": "We discuss the implications of a stellar-dominated UV background at high redshifts for the star formation history of Lyman-break galaxies (LBGs) and the thermal and ionization state of the intergalactic medium (IGM). The composite spectrum of 29 LBGs evaluated by Steidel \\etal (2000) at $\\langle z\\rangle=3.4$ can be well fit by a stellar population with ongoing star formation, a Salpeter initial mass function, modest or negligible dust reddening, and no intrinsic \\hi photoelectric absorption. Fading starbursts in which star formation has ceased for $10^{7}\\,$ yr or more cannot reproduce the observed flux shortward of 1 Ryd. The small \\hi optical depth in LBGs suggests that the neutral gas from which stars form is most likely contained in compact clouds of neutral gas with small covering factor. The escape fraction of H-ionizing photons  must be close to 100 percent for the observed sample of LBGs. The spectrum of ionizing photons  produced by a stellar population with ongoing star formation is similar to that of QSOs between 1 and 3 Ryd, but becomes softer between 3 and 4 Ryd and drops sharply shortward of 4 Ryd. A galaxy-dominated UV background appears inconsistent with the observed \\heii/\\hi opacity ratio at $z=2.4$, but might be able to explain the \\SiIV/\\CIV\\ abundances measured at $z>3$ in quasar absorption spectra. A scenario may be emerging where star-forming galaxies reionized intergalactic hydrogen at $z>6$ and dominate the 1 Ryd metagalactic flux at $z>3$, with quasi-stellar sources taking over at lower redshifts. If the large amplitude of the H-ionizing flux estimated by Steidel \\etal is correct,  hydrodynamical simulations of structure formation in the IGM within the cold dark matter paradigm require a baryon density (to explain the observed opacity of the \\op forest in quasar absorption spectra) which is similar to or larger than that favoured by recent CMB experiments, and is inconsistent with standard nucleosynthesis values. ",
        "introduction": "The strength and spectrum of the UV ionizing background at high redshift and the nature of the sources which reionized the hydrogen component of the intergalactic medium (IGM) are two of the big outstanding questions in observational cosmology. QSOs and star-forming galaxies have long been considered the two prime candidates. It has been argued that QSOs may fall short of producing a metagalactic flux as high as that inferred from the proximity effect  at high redshift (Bechtold 1994; Giallongo \\etal 1996; Cooke \\etal 1997; Scott \\etal 2000) since their space density declines rapidly at early epochs (Shapiro \\& Giroux 1987; Miralda-Escud\\'e \\& Ostriker 1990; Meiksin \\& Madau 1993; Madau \\etal 1999). The detection of a numerous population of Lyman-break galaxies (LBGs) at $z\\approx 3$ (Steidel \\etal 1996) makes the idea of massive stars dominating the UV background at early epochs quite plausible. At $1500\\,$\\AA\\ their emissivity outweighs that of QSOs by about a factor of fifteen (Madau \\etal 1999). Until recently it seemed likely that only a small fraction of the ionizing photons emitted within starburst galaxies could escape the dense \\hi layers into the IGM, as is observed in their low-redshift counterparts (Leitherer \\etal 1995; Hurwitz \\etal 1997). The detection by Steidel \\etal (2000) of flux beyond the Lyman limit (with no significant indication of a break at $912\\,$\\AA) in a composite spectrum of 29 LBGs at $\\langle z\\rangle=3.4$ thus comes as a surprise.  In this {\\it Letter} we assess the significance of the Steidel \\etal (2000) findings by comparing their spectrum to model spectra calculated with the population synthesis code STARBURST99 (Leitherer \\etal 1999). We discuss the implications of this result for the star formation history and gas content of these galaxies, for the spectrum of the UV background, for the cosmological reionization, and for the baryon content of the Universe. ",
        "conclusions": "We have shown that the detection of flux beyond the Lyman limit from LBGs is most easily explained as being due to the O stars present in a star forming region with ongoing star formation, a Salpeter IMF, modest or negligible reddening, and negligible absorption due to \\hi\\  in the local ISM. Fading starbursts in which star formation has ceased for $10^{7} \\yr$  or more cannot reproduce the observed flux shortward of 1 Ryd. The small \\hi\\  optical depth cannot  be due to photoionization. The cold neutral gas from which they stars form must be contained in compact small regions with small covering factor.  The resulting ionizing background will be dominated by O stars shortward of the Lyman limit and is as hard as a QSO spectrum between 1 and 3 Ryd. Shortward of 3 Ryd the spectrum is softer than a QSO spectrum and there is strong drop shortward of 4 Ryd. A consistent picture seems to emerge  where the hydrogen of the Universe is reionized by stars at $z>6$ while the reionization of helium is due to QSOs and is delayed to $z\\sim$ 4 or 3. The contribution of stars to the ionizing UV background is equal or larger than that  QSOs at least up to $z\\sim 3$. At smaller redshift  the ratio of \\heii\\  and \\hi\\  opacity and the observed \\SiIV/\\CIV\\ ratios in the associated metal absorption systems of intermediate column density QSO absorption systems argue for a UV background dominated by QSOs. This requires a substantial decrease of the escape fraction of ionizing photons from star-forming galaxies with decreasing redshift. The amplitude of the UV background inferred from the composite spectrum of Steidel \\etal makes the  measurements of the opacity of the \\op forest in QSO absorption spectra inconsistent with D/H measurements if the spectrum is representative for all LBGs and nucleosynthesis calculations for D/H are correct.  The inferred value for $\\Omega_{\\rm b}$ is similar to that that indicated  by the CMB experiments Boomerang and Maxima if $J_{21}= 0.4-0.6$."
    },
    "0010/astro-ph0010407_arXiv.txt": {
        "abstract": "{We discuss the distribution and kinematics of neutral hydrogen in the halos of spiral galaxies. We focus in particular on new results obtained for the nearby Sc galaxy NGC~2403  which have revealed the presence of extended HI emission (here referred to as the `beard') at anomalous velocities with respect to the `cold' HI disk. Modeling shows that this component has a mass of about 1/10 of the total HI mass and is probably located in the halo region. Its kinematics differs from that of the thin HI disk: it rotates more slowly and shows radial inflow. The origin of this anomalous gas component is unknown. It could be the result of a galactic fountain or of accretion of extragalactic `primordial' gas.} ",
        "introduction": "The study of the vertical structure and kinematics of the HI disks of spiral galaxies has in recent years been pursued in two main directions. One has been the investigation of the large-scale dynamics, with special attention to the phenomena of warping and flaring in the outer parts of the HI disks. Attempts have been made (e.g. Van der Kruit 1981, Olling 1995) to determine the mass of the stellar disks and to investigate the properties of dark matter halos. The other has been the search, in the inner parts of the disks, for evidence of vertical extensions and motions of the HI gas, possibly related to star formation activity, the expansion of superbubbles and the circulation of gas between disk and halo. The results are helpful for understanding the observations of the Milky Way and indeed, as we will see below, some of the HI features discussed here do have properties similar to those of the galactic High Velocity Clouds.  In addition to these `internal' phenomena, there are other events of external origin. These are the gravitational interactions with other systems and the accretion of gas, which may affect the density distribution and kinematics of the gas in the halo region and may also play an important role in the formation and evolution of disks (see for a recent review Sancisi 1999). ",
        "conclusions": "What are the origin and nature of the anomalous gas in NGC~2403? The overall pattern and the large-scale regularity indicated by the position-velocity maps (Fig.~2) suggest that this gas (both the `beard' and the forbidden emission) forms one coherent structure. Its anomalous velocities, despite a big spread of more than 150 km~s$^{-1}$, are confined inside the range of the rotational velocities of the disk.  The overall pattern of this anomalous gas and its coherent spatial and velocity structure seem to point to one phenomenon and one common origin. Possibilities are:  1. A galactic fountain (Shapiro \\& Field 1976; Bregman 1980) powered by supernova explosions and stellar winds which blow up the ionized gas from the disk into the halo. The gas may reach large distances from the plane and then cool and fall back onto the plane. This explanation finds some support in the active star formation and the presence of bright HII complexes in the disk of NGC~2403. Also the presence of a large number of holes in the HI layer (see Fig.~1) and of expanding supershells (Mashchenko, Thilker, \\& Braun 1999) supports the picture of an effervescent disk and of a significant disk-halo-disk circulation. We are pursuing the study of these phenomena further with deep spectroscopy of the H$\\alpha$ emission from the bright inner parts of NGC~2403 and $Chandra$ observations of the bright optical disk. The main difficulty with a standard fountain interpretation lies in the presence of the apparently counterrotating, `forbidden' emission in NGC~2403. In order to explain that, a new approach and different assumptions for the fountain dynamics, for instance no conservation of angular momentum, may be necessary.  2. Infall of {\\it primordial} intergalactic gas. The hypothesis of cosmological gas infall has been proposed in the past and discussed (cf. Oort 1970) in connection with the problem of the origin of the HVCs in our galaxy and with the need to re-supply the disks of spiral galaxies with fresh gas. Such a possibility has to be considered here. In that connection it would be essential to understand the overall kinematical pattern of the anomalous HI and to understand its large-scale coherent structure which follows the rotation of the disk of NGC~2403.  Also, it would be important to determine its metal abundance and to find out whether it is as low as found for the intergalactic gas or closer to solar.  The results of the present HI observations of NGC~2403 have also interesting implications regarding the High Velocity Clouds observed in the Milky Way (Wakker \\& van Woerden 1997) which are still a mystery and whose distances, nature and origin are still a matter of debate. Is it possible that the anomalous HI complexes discovered around NGC~2403 are analogous to the HVCs seen in the Galaxy? With what we know of the HVCs it seems fair to suggest that we probably have detected in NGC~2403 a population of HI clouds of the same type as those intermediate and high velocity clouds which are thought to be closely associated with the Galaxy and to be probably a galactic fountain type of phenomenon.  Clearly, they would be different from the Magellanic Stream.  How common are the `beards' of galaxies? As far as we know there are some cases, like M~33 and UGC~7766, which indicate the presence of similar structures. HI observations of higher sensitivity would tell whether there is any correlation with luminosity or mass or high surface brightness or environment, and which of these may play the crucial role."
    },
    "0010/astro-ph0010294_arXiv.txt": {
        "abstract": "}[2]{{\\footnotesize\\begin{center}ABSTRACT\\end{center} \\vspace{1mm}\\par#1\\par \\noindent {~}{\\it #2}}}  \\newcommand{\\TabCap}[2]{\\begin{center}\\parbox[t]{#1}{\\begin{center} \\small {\\spaceskip 2pt plus 1pt minus 1pt T a b l e} \\refstepcounter{table}\\thetable \\\\[2mm] \\footnotesize #2 \\end{center}}\\end{center}}  \\newcommand{\\TableSep}[2]{\\begin{table}[p]\\vspace{#1} \\TabCap{#2}\\end{table}}  \\newcommand{\\FigCap}[1]{\\footnotesize\\par\\noindent Fig.\\  % \\refstepcounter{figure}\\thefigure. #1\\par}  \\newcommand{\\TableFont}{\\footnotesize} \\newcommand{\\TableFontIt}{\\ttit} \\newcommand{\\SetTableFont}[1]{\\renewcommand{\\TableFont}{#1}}  \\newcommand{\\MakeTable}[4]{\\begin{table}[htb]\\TabCap{#2}{#3} \\begin{center} \\TableFont \\begin{tabular}{#1} #4 \\end{tabular}\\end{center}\\end{table}}  \\newcommand{\\MakeTableSep}[4]{\\begin{table}[p]\\TabCap{#2}{#3} \\begin{center} \\TableFont \\begin{tabular}{#1} #4 \\end{tabular}\\end{center}\\end{table}}  \\newenvironment{references}% { \\footnotesize \\frenchspacing \\renewcommand{\\thesection}{} \\renewcommand{\\in}{{\\rm in }} \\renewcommand{\\AA}{Astron.\\ Astrophys.} \\newcommand{\\AAS}{Astron.~Astrophys.~Suppl.~Ser.} \\newcommand{\\ApJ}{Astrophys.\\ J.} \\newcommand{\\ApJS}{Astrophys.\\ J.~Suppl.~Ser.} \\newcommand{\\ApJL}{Astrophys.\\ J.~Letters} \\newcommand{\\AJ}{Astron.\\ J.} \\newcommand{\\IBVS}{IBVS} \\newcommand{\\PASP}{P.A.S.P.} \\newcommand{\\Acta}{Acta Astron.} \\newcommand{\\MNRAS}{MNRAS} \\renewcommand{\\and}{{\\rm and }} {We present 100 multiple cluster candidates selected from the OGLE catalog of star clusters in the Large Magellanic Cloud. Statistical analysis shows that the significant fraction of these objects may constitute physical systems. Coeval ages of 102 components of multiple objects suggest their common origin. 53 components have very different ages. The comparison of the population of multiple clusters candidates from the SMC and the LMC shows that: a)~distributions of sizes and ages of multiple and single clusters from the Magellanic Clouds are very similar, b)~the difference of sizes of components of a given system is small, c)~in both distributions of separation of multiple clusters from the LMC and SMC two peaks are seen at about 9~pc and 15~pc, d)~both age distributions reveal peaks around 100~Myr, which may be connected with the last encounter of the LMC and the SMC.}  \\vskip32pt ",
        "introduction": "\\vskip9pt Previous searches revealed that the Magellanic Clouds possess numerous multiple cluster candidates. Bhatia and Hatzidimitriou (1988) and Bhatia \\etal (1991) constructed the catalog of cluster pairs from the LMC. Binary clusters from the SMC were described by Hatzidimitriou and Bhatia (1990). Recently, based on the OGLE catalog of star clusters from the SMC (Pietrzy{\\'n}ski \\etal 1998), the list of multiple star cluster candidates from the central parts of this galaxy was presented by Pietrzy{\\'n}ski and Udalski (1999a). Simple statistical considerations show that the number of multiple cluster candidates is significantly larger than the number expected from the chance line-up due to projection. In the case of selected cluster systems from the LMC their physical connections were confirmed based on detailed studies (Kontizas \\etal 1993, Vallenari \\etal 1998, Leon \\etal 1998, Dieball and Grebel 2000 and references therein).  Having important implications on the processes of formation and evolution of clusters, the multiple star clusters were also subject of several theoretical investigations. Fujimoto and Kumai (1997) pointed out that the binary clusters could be formed by the oblique collisions between massive gas clouds. This scenario led to systems of clusters with very similar ages. Another scenario of tidal capture in groups of clusters was proposed by Leon \\etal (1999). Such a mechanism explains formation of systems with members having large age difference.  The Optical Gravitational Lensing Experiment (OGLE) (Udalski, Kubiak and Szyma{\\'n}ski 1997) provided in precise {\\it BVI} observations of millions of stars from the Magellanic Clouds (Udalski \\etal 1998). The OGLE data, covering relatively large areas in the Clouds are very well suited for searching and analyzing the properties of star clusters from these galaxies. In a series of papers we present results of these investigations. Catalogs of clusters from the LMC and the SMC were presented by Pietrzy{\\'n}ski \\etal (1999) and Pietrzy{\\'n}ski \\etal (1998), respectively. Ages for about 700 of these clusters were determined using the standard procedure of isochrone fitting (Pietrzy{\\'n}ski and Udalski 1999b, 2000). Possible multiple clusters from the SMC were listed by Pietrzy{\\'n}ski and Udalski (1999c). The lists of eclipsing systems and Cepheids in coincidence with star clusters in Magellanic Clouds are given in Pietrzy{\\'n}ski and Udalski (1999 ad).  In this contribution we present multiple cluster candidates selected from the catalog of star clusters in central parts of the Large Magellanic Cloud and compare them with the objects from the OGLE catalog of multiple clusters from the SMC (Pietrzy\\'nski and Udalski 1999c).  \\vskip30pt ",
        "conclusions": "We presented the multiple cluster candidates from the LMC selected from the OGLE catalog of star clusters (Pietrzy{\\'n}ski and Udalski 1999c). Simple statistical considerations show that significant fraction of them may constitute physical systems. We compare the populations of potential multiple clusters from the LMC and SMC. Taking into account that the completeness of these catalogs should be comparable, we found that the fractions of potential multiple systems in these galaxies are almost the same, around 12\\%. Based on inspection of size and age distributions we can conclude that their shapes are very similar to the shapes of distributions of ages and sizes of single clusters. The difference of sizes between components of a given system is small. The distributions of separations reveal two peaks at 9 and 15~pc. Most clusters are young. Their distributions of ages show peaks at about 100~Myr. Further photometric and spectroscopic studies are necessary to confirm physical nature of the presented systems.  \\Acknow{The paper was partly supported by the Polish KBN grants 2P03D00814 to A.~Udalski. Partial support for the OGLE project was provided with the NSF grant AST-9530478 to B.~Paczy\\'nski.}"
    },
    "0010/astro-ph0010588_arXiv.txt": {
        "abstract": "The cosmic $\\gamma$-ray burst (GRB) formation rate, as derived from the variability-luminosity relation for long-duration GRBs, is compared with the cosmic star formation rate. If GRBs are related to the collapse of massive stars, one expects the GRB rate to be approximately proportional to the star formation rate. We found that these two rates have similar slopes at low redshift. This suggests that GRBs do indeed track the star formation rate of the Universe, which in turn implies  that the formation rate of massive stars that produce GRBs is proportional to the total star formation rate. It also implies that we might be able to use GRBs as a probe of the cosmic star formation rate at high redshift. We find that the cosmic star formation rate inferred from the variability-luminosity relation increases steeply with redshift at $z>3.0$.  This is in apparent contrast to what is derived from measurements of the cosmic star formation rate at high redshift from optical observations of field galaxies, suggesting that much high-$z$ star formation is being missed in the optical surveys, even after corrections for dust extinction have been made. ",
        "introduction": " ",
        "conclusions": ""
    },
    "0010/astro-ph0010541_arXiv.txt": {
        "abstract": "We present the results of a high angular resolution, multi--transition analysis of the molecular gas in M\\,82. The analysis is based on the two lowest transitions of $^{12}{\\rm CO}$ and the ground transition of the rare isotopes $^{13}{\\rm CO}$ and ${\\rm C}^{18}{\\rm O}$ measured with the PdBI, the BIMA array and the IRAM 30m telescope.\\\\ In order to address the question of how the intrinsic molecular cloud properties are influenced by massive star formation we have carried out radiative transfer calculations based on the observed CO line ratios. The calculations suggest that the kinetic temperature of the molecular gas is high in regions with strong star formation and drops towards the outer molecular lobes with less ongoing star formation. The location of the highest kinetic temperature is coincident with that of the mid infrared (MIR) peaks which trace emission from hot dust. The hot gas is associated with low H$_2$ densities while the cold gas in the outer molecular lobes has high H$_2$ densities. We find that CO intensities do not trace \\hh\\, column densities well. Most of the molecular gas is distributed in a double--lobed distribution which surrounds the starburst. A detailed analysis of the conversion factor from CO intensity to \\hh\\, column density shows that \\xco\\, depends on the excitation conditions. We find $\\xco\\, \\sim {\\rm T}_{\\rm kin}^{-1}\\,\\,{\\rm n(\\hh)}^{1/2}$, as expected for virialized clouds. ",
        "introduction": "M\\,82 is regarded as the archetypical starburst galaxy (Rieke \\etal\\ \\cite{rieke}). Its distance of only 3.9\\,Mpc (Sakai \\& Madore \\cite{sakai}) makes M\\,82 an excellent laboratory for studying the relevant physical processes connected with starburst activity in detail. The central few hundred parsecs of this galaxy are heavily obscured by dust and gas which hides the central starburst region against direct observations at optical wavelengths. Evidence for the strong star--forming activity in the central region comes from radio (e.g. Kronberg \\etal\\ \\cite{kronberg}; Wills \\etal\\ \\cite{wills99}) and infrared observations (e.g. Telesco \\& Gezari \\cite{telesco92}) and also from the prominent bipolar outflow visible in \\ha\\,(e.g. Bland \\& Tully \\cite{bland}; McKeith \\etal\\ \\cite{mckeith}; Shopbell \\& Bland--Hawthorn \\cite{shopbell}) and X--rays (e.g. Bregman \\etal\\ \\cite{bregman}). The massive star formation (SF) is believed to be fueled by the large amount of molecular gas which is present in the centre of M\\,82.\\\\ On the other hand, SF affects the distribution, kinematics and physical conditions of the surrounding interstellar medium (ISM). Early studies of the distribution of the molecular gas in M\\,82 unveiled a double--lobed circumnuclear distribution of CO which was interpreted as a molecular torus with a depletion of molecular gas in the central region (Nakai \\etal\\ \\cite{nakai}). More recent high--resolution studies by Shen \\& Lo (\\cite{shen}) using BIMA and Neininger \\etal\\ (\\cite{nico}) using the IRAM interferometer at Plateau de Bure (PdBI) showed a third molecular peak 65\\,pc west of the 2.2\\m\\, nucleus (Dietz \\etal\\ \\cite{dietz}). Using these high--resolution CO maps Wei\\ss\\, \\etal\\ (\\cite{aweiss}) identified an expanding superbubble in the molecular gas of M\\,82 which linked the triple peak CO distribution and its disturbed velocity field to the prominent outflow visible in \\ha\\, and X--rays.\\\\ Multi--transition analyses of molecular emission lines (CO, CS, HCN) showed that the starburst also affects the physical conditions of the molecular gas (Wild \\etal\\ \\cite{wild}; Henkel \\& Bally \\cite{henkel85}; Brouillet \\& Schilke \\cite{brouillet}). A large fraction of the molecular gas is concentrated in warm (\\tkin\\,=\\,50\\,K) and dense (\\dhh\\,=\\,10$^4$\\,cm$^{-3}$) clouds (Wild \\etal\\ \\cite{wild}; G\\\"usten \\etal\\ \\cite{guesten}). In a recent study Mao \\etal\\ (\\cite{mao}) analyzed CO mm and sub--mm emission lines in M\\,82 to investigate the physical properties of the molecular clouds. They conclude that the bulk of CO emission arises from photon--dominated regions (PDRs) while tracers of high--density gas like CS and HCN are less affected by the strong UV radiation from massive stars.\\\\ Even though these studies already provided a good global picture of the physical conditions of the molecular gas in M\\,82, no detailed high--resolution study exists so far that allows to resolve variations of the excitation conditions of the molecular gas over the central part of M\\,82. In this paper we present the results of a high angular resolution, multi--transition CO analysis and compare the intrinsic gas properties with observations of high--level star formation. In Sect.\\,2 we briefly summarize our observations and the data reduction. In Sect.\\,3 we describe the main results including a description of the CO morphology and kinematics, the CO line ratios, results from the LVG calculation and on \\xco. In Sect.\\,4 we compare our results to previous studies. Sect.\\,5 summarizes our conclusions. ",
        "conclusions": "We have observed the \\bco\\, and \\eco\\, emission lines in the starburst galaxy M\\,82 with high spatial resolution using the Plateau de Bure interferometer. Our main conclusions are:\\\\ \\\\ 1) The overall morphology and kinematics for both transitions are similar to those of \\aco\\, and \\cco\\, published by Shen \\& Lo (\\cite{shen}) and Neininger \\etal\\ (\\cite{nico}). The dynamical centre of the molecular gas coincides with the 2.2 \\m\\, nucleus while the centroid of the molecular mass is located 100\\,pc west of M\\,82`s centre. South of the expanding molecular superbubble (Wei{\\ss} \\etal\\ \\cite{aweiss}) an outflow of molecular gas with a total mass of $7.2 \\times 10^5\\,\\msol$ is detected.\\\\ 2) The \\bco/\\aco\\, line intensity ratios are lower ($\\leq\\,1.4$) than previously reported. Thus, CO line ratios in M\\,82 are not outstanding, but comparable . to values found in other starburst galaxies like NGC\\,253. Line ratios vary across the disk of M\\,82. Near the MIR peaks, the \\bco/\\aco\\, ratios are high; in the outer parts, that are less affected by the starburst, this ratio drops to unity.\\\\ 3) An LVG excitation analysis of the CO lines suggests that the excitation conditions of the molecular gas are strongly influenced by environmental effects. In the outer parts of the CO distribution we find \\hh\\, densities of $n(\\hh) = 10^{3.5-4.2} {\\rm cm}^{-3}$ and kinetic temperatures of ${\\rm T}_{kin}=45-60{\\rm K}$. Towards the star--forming regions, indicated by strong MIR emission, the kinetic temperatures raise above 150\\,K. The hot gas is associated with low \\hh\\, densities of only $n(\\hh) = 10^{2.8-3.1} {\\rm cm}^{-3}$. Area filling factors of $f_a = 0.2 - 0.4$ and volume filling factors of $f_v = 0.001 - 0.02$ indicate that the gas is organized in small clumps with a typical size of $r_{cloud} \\approx 1\\,{\\rm pc}$. [\\aba]/[\\abb] abundance ratios are about 70 without significant spatial variations across the galaxy. In contrast, [\\aba]/[\\abc] abundance ratios in the outer parts of M\\,82 are comparable to those found at the Galactic Centre ([\\aba]/[\\abc] = 270) but decrease to only [\\aba]/[\\abc] = 160 at the star--forming regions. Beam--averaged \\hh\\, column densities range from ${\\rm N(\\hh)_{4''}}=2.4 \\times10^{22}\\,{\\rm cm^{-2}}$ near the MIR peaks to ${\\rm N(\\hh)_{4''}}= 2.3\\times10^{23}\\,{\\rm cm^{-2}}$ at the western CO lobe. The \\hh\\, distribution has a double--peak morphology which surrounds the central starburst region. The central 300\\,pc are depleted in \\hh. Thus the \\hh\\, distribution differs from the CO distribution. This result even holds when the \\hh\\, column densities are calculated under the assumption of LTE conditions. The total molecular mass is $2.3 \\times 10^8\\,\\msol$.\\\\ 4) The conversion factor from $I$(CO) to \\nhh\\, ($X_{\\rm CO}$) depends on the excitation conditions of the CO--emitting volume. Even in regions which are less affected by the starburst \\xco\\, is about 3 times less than the standard Galactic value. From the LVG analysis we find that  $X_{\\rm CO} \\sim {\\rm T}_{\\rm kin}^{-1}\\,\\,{\\rm n(\\hh)}^{1/2}$. Therefore  \\xco\\, is lower in the central star--forming regions than in the outer molecular lobes."
    },
    "0010/astro-ph0010227_arXiv.txt": {
        "abstract": "We study relationships between the neutron-rich skin of a heavy nucleus and the properties of neutron-star crusts. Relativistic effective field theories with a thicker neutron skin in $^{208}$Pb have a larger electron fraction and a lower liquid-to-solid transition density for neutron-rich matter. These properties are determined by the density dependence of the symmetry energy which we vary by adding nonlinear couplings between isoscalar and isovector mesons. An accurate measurement of the neutron radius in $^{208}$Pb---via parity violating electron scattering---may have important implications for the structure of the crust of neutron stars. ",
        "introduction": " ",
        "conclusions": ""
    },
    "0010/astro-ph0010011_arXiv.txt": {
        "abstract": "Time-resolved, multi-epoch, {\\it Hubble Space Telescope\\/} ultraviolet eclipse spectrophotometry is presented for the magnetic cataclysmic variables (mCVs) \\vori, \\mnhya, \\voph, and \\vaql.  Separation of the eclipse light curves into specific wavebands allows the multiple emission components to be distinguished.  Accretion-heated photospheric spots are detected in \\vori\\ and \\voph, indicating covering factors $f<0.015$ and $\\sim$0.008, and blackbody temperatures $T_{\\rm spot}>150,000$~K and $\\sim$90,000~K, respectively. The cyclotron-emitting shock on \\mnhya\\ is detected in the optical/near-UV and found to occupy no more than $f=0.004$ of the stellar surface. Emission from the accretion stream is a prominent component in \\vori\\ and \\voph, and the protracted eclipse ingress of UV emission lines in the latter object indicates that the stream penetrates the white dwarf magnetosphere to a height of $\\sim$9~$R_{\\rm wd}$ before it is funneled onto the magnetic poles.  The UV emission-line spectrum of \\vori\\ (and to a lesser extent, \\mnhya) is unusual among mCVs, with the relative strength of \\ion{N}{5} $\\lambda$1240 in \\vori\\ matched only by the asynchronous system BY~Cam. The prominence of \\ion{N}{4} $\\lambda$1718 suggests that an overabundance of nitrogen may be the most likely explanation.  Three epochs of observation of the asynchronous \\vaql\\ cover about one-third of a 50-day lap cycle between the white dwarf spin and binary orbit.  The eclipse light curves vary enormously from epoch to epoch and as a function of waveband.  The dereddened UV spectrum is extremely bright compared with the optical, and the overall spectral energy distribution coupled with the duration of eclipse ingress indicate that the dominant source of energy is a hot ($T_{\\rm wd}\\sim35,000$~K) white dwarf.  This explanation also accounts for the modest brightness variation out of eclipse and the weakness of optical circular polarization due to cyclotron emission.  Undiminished line emission through eclipse indicates that the eclipse itself is caused by a dense portion of the accretion stream, not the secondary star. The temperature of the white dwarf in \\vaql\\ greatly exceeds that in any other AM Her system except the recent nova V1500 Cyg.  This, combined with its current asynchronous nature and rapid timescale for relocking, suggests that \\vaql\\ underwent a nova eruption in the past $\\sim$$75-150$~yr.  The reversed sense of asynchronism, with the primary star currently {\\it spinning up\\/} toward synchronism, is not necessarily at odds with this scenario, if the rotation of the magnetic white dwarf can couple effectively to the ejecta during the dense wind phase of the eruption. ",
        "introduction": "At least 11 of the nearly 5 dozen known AM Her variables are true eclipsing binaries where light curves through eclipse can be used to analyze the structure and relative importance of emission regions on the magnetic white dwarf and in the accretion column that connects it to the mass-losing companion.  The eclipsing systems are found with magnetic field strengths $7<B<60$~MG (MG $=10^6$~G), span the orbital period range $1.5<P_{\\rm orb}<8.0$~hr, and include one asynchronous system.  Thus, there is reasonable assurance that an understanding gained from the eclipsing systems can be applied to the class as a whole (see Cropper 1990 for a general review).  Because the magnetic field funnels accreting gas onto impact regions as small as $f=A_{\\rm spot}/4\\pi R_{\\rm wd}^2\\sim10^{-3}$, high time resolution, $\\Delta t\\lesssim1$~s, plus an adequate photon rate are required to resolve the structures.  The wavelength region of choice is the ultraviolet, where the white dwarf and heated impact region are prominent ($T_{\\rm wd}\\sim15,000$~K and $T_{\\rm spot}\\gtrsim30,000$~K, respectively), line and continuum emission from the funnel gas is available, and the contribution from the late-type secondary star is minimized. The first two papers in this series presented {\\it HST\\/} time-resolved spectroscopy of DP Leo (Stockman et al. 1994, hereafter Paper I) and UZ For (Stockman \\& Schmidt 1996, hereafter Paper II); the latter study also applied the method to ST LMi, where the accretion footpoint undergoes ``self-eclipse'' when it rotates behind the white dwarf limb.  DP Leo was observed in a state of very low accretion, nevertheless the eclipse light curves revealed the contribution of an accretion-heated spot ($T_{\\rm spot}\\sim50,000$~K) against the surrounding white dwarf photosphere ($T_{\\rm wd}\\sim16,000$~K). The fractional surface area of the near-UV spot, $f\\sim0.006$, was found to be very similar to that of the optical cyclotron source in a high state.  Observations of UZ For showed that the accretion stream itself can be a dominant UV source in an active accretion state, owing to both strong resonance line emission and free-bound continua. The temperature of the hot spot was measured at $T\\gtrsim30,000$~K and found to occupy at least $f=0.01$ of the white dwarf surface area.  Attempts to model the line emission led to a two-phase composition for the gas stream, consisting of low filling-factor, dense clumps threading a more rarefied flow.  Many of the observed properties of UZ For in the UV are shared by the well-studied HU Aqr (e.g., Schwope 1998), and similar conclusions have been reached.  In this paper we present Faint Object Spectrograph (FOS) spectroscopy of the final 4 eclipsing magnetic cataclysmic variables (mCVs) observed in this program.  All four are relatively bright when actively accreting, $V<17$, and together display an interesting variety of properties: \\vori\\ is the longest-period AM Her system known, with an eclipse persisting for more than 40~min.  Given the rather modest magnetic field of $B\\sim60$~MG, spin-orbit locking in this binary challenges current theories of the magnetic interaction mechanism.  In the high state, \\mnhya\\ is a classic 2-pole accretion system whose strong optical polarization enables detailed models. The weakest magnetic field measured on any AM Her system is found on \\voph, a characteristic which has frustrated polarimetrists but motivated a number of photometric and spectroscopic analyses.  \\vaql\\ is the only one of the 4 for which an eclipse by the secondary star is in doubt; it is also the sole asynchronous eclipsing system, made all the more unusual because the spin period appears to {\\it exceed\\/} the orbital period by 0.3\\%.  These and other properties of the systems are summarized together with basic references to previous work in Table~1.  Recently, several sophisticated techniques used to study the emission of disk CVs have found their way to magnetic systems.  These include Doppler tomography (e.g., Schwope et al. 1999) and maximum entropy methods (MEM) for modeling eclipse light curves in broad bands (Harrop-Allin, Hakala, \\& Cropper 1999b; Harrop-Allin et al. 1999a) as well as of full-orbit phase-resolved emission-line spectroscopy (Sohl \\& Wynn 1999; Vrielmann \\& Schwope 1999; Kube, G\\\"ansicke, \\& Beuermann 2000). Several of these approaches utilize Genetic Algorithms (e.g., Hakala 1995) to efficiently approach global optimization of the many model variables.  Each of the methods has its strengths and limitations.  Doppler tomography maps the velocity distribution of emitting regions, not their locations, and the technique is insensitive to motions out of the orbital plane (see, e.g., Schwope et al. 1999).  MEM and genetic optimization codes were developed in an attempt to deal with the tremendous ambiguity of generalized model-fitting approaches, where the number of emitting sources may exceed the number of independent data points. The various tools are in many ways complementary and together represent the state-of-the-art in analytical techniques for time-resolved studies of accretion binaries.  We have chosen to apply a more qualitative approach to interpret the eclipse light curves of these mCVs for several reasons.  First, the photon rate in the near-UV from {\\it HST\\/}+FOS cannot compete with what is now commonplace from modern instrumentation and large ground-based telescopes working in the visible ($\\sim$$0.35-1.0~\\mu$m).  This, plus the low spectral resolution of FOS effectively rule out quantitative Doppler studies of the emission lines. Time allocation considerations on {\\it HST\\/} limited the observing program on each source to windows just encompassing eclipse, and for some targets scheduling constraints precluded obtaining both ingress and egress at the same epoch. With such limitations, the data presented here are amenable to identifying the location and relative importance of principle sources of line and continuum emission, but not for unraveling their detailed structure or kinematics. ",
        "conclusions": "Time-resolved {\\it HST\\/} UV spectroscopy of four eclipsing mCVs has enabled us to identify:  \\begin{itemize}  \\item{Accretion-heated photospheric spots in \\vori\\ and \\voph\\ with covering factors $f<0.015$ and $\\sim$0.008, and blackbody temperatures $T_{\\rm spot}>150,000$~K and $\\sim$90,000~K, respectively.}  \\item{A cyclotron shock on \\mnhya\\ with $f\\lesssim0.004$.}  \\item{Prominent emission from the accretion stream in \\vori\\ and \\voph.  A protracted emission-line ingress in the latter object indicates the stream penetrates the magnetosphere to a height of $\\sim$9~$R_{\\rm wd}$ before it is funneled onto the magnetic poles.}  \\item{Unusually strong \\ion{N}{5} $\\lambda$1240 in \\vori\\ and, to a lesser extent, \\mnhya, that is suggestive of an overabundance of nitrogen in the gas stream.}  \\item{A spectral energy distribution for \\vaql\\ that strongly suggests the dominant source of energy is a hot ($T_{\\rm wd}\\sim35,000$~K) white dwarf.  The undiminished line emission through eclipse indicates that the eclipse is caused by a dense portion of the accretion stream, not the secondary star. The very hot white dwarf, combined with the current asynchronous nature and rapid timescale for relocking, suggests that \\vaql\\ underwent a nova eruption in the past $\\sim$$75-150$~yr.  A reversed sense of asynchronism, with the primary star currently spinning up toward synchronism, is not at odds with this history, if the white dwarf rotation can magnetically couple to the ejecta during the dense wind phase of the eruption.}  \\end{itemize}  Time-tagged UV spectrophotometry is an important tool for understanding the sizes and relative contributions of the accretion stream, hot spot, and white dwarf photospheres in mCVs.  Both the Space Telescope Imaging Spectrograph (STIS) and the Cosmic Origins Spectrograph (COS) offer sensitivities an order of magnitude greater than the FOS.  Unfortunately, with the limited aperture of {\\it HST\\/}, it will be difficult to resolve objects smaller than 10\\% of the white dwarf radius.  Future UV telescopes, with apertures exceeding 4~m and high efficiency, energy-resolving detectors will be ideal facilities for discerning details of structures as small as the accretion shock itself."
    },
    "0010/astro-ph0010361_arXiv.txt": {
        "abstract": "We present measurements of absorption by the 21cm hyperfine transition of neutral hydrogen toward radio sources at substantial look-back times. These data are used in combination with observations of rotational transitions of common interstellar molecules, to set limits on the evolution of the fine structure constant: ${{\\dot\\alpha}\\over{\\alpha}} < 3.5 \\times 10^{-15}$~year$^{-1}$, to a look-back time of 4.8 Gyr. The neutral hydrogen observations employed Very Long Baseline Interferometry in order to mitigate the substantial uncertainty arising from the fact that observations at very different wavelengths may probe different lines-of-site due to frequency dependent structure of the background source. We discuss the implication of these results on theories unifying natural forces based on  compact quantum dimensions. In the context of string theory, the limit on the secular evolution of the scale factor of these compact dimensions, $R$, is ${{\\dot R}\\over{R}} < 10^{-15}$~year$^{-1}$. Including terrestrial and other astronomical measurements places limits (2$\\sigma$) on  slow oscillations of   $R$ from the present to the epoch of cosmic nucleo-synthesis, just seconds after the big bang, of ${{\\Delta R}\\over{R}} < 10^{-5}$. ",
        "introduction": " ",
        "conclusions": ""
    },
    "0010/astro-ph0010157_arXiv.txt": {
        "abstract": "We report \\bs observations of spatially resolved hard X-ray emission from IC~443, a supernova remnant interacting with a molecular cloud. The emission is shown to come from two localized features spatially correlated with bright molecular emission regions. Both hard X-ray features have soft X-ray counterparts, in one case shifted by $\\sim 2^\\prime$ toward the remnant interior.  The spectra of X-ray emission from both isolated features have photon index $\\lsim 2$ in the MECS regime.  The emission detected from the remnant with PDS detector extends up to 100 keV. We discuss the observed properties of the hard X-ray features in relation to non-thermal emission from shocked molecular clumps and pulsar wind nebula. ",
        "introduction": "Massive stars that are the likely progenitors of core collapsed supernovae are expected to be spatially correlated with molecular clouds. IC~443 (G189.1+3.0)  is one of the best space laboratories to study rich phenomena accompanying supernova interaction with a molecular cloud. This was established from observed shock-excited molecular line emission from OH, CO and H$_2$ (e.g. Burton \\etal 1988; van Dishoeck \\etal 1993; Cesarsky \\etal 1999). Observations of OH masers and CO clumps towards IC~443, W28 and W44 also provide convincing evidence of molecular cloud interaction in this case (Claussen et al 1997).  IC 443 was a target of  X-ray observations with { \\it HEAO 1} (Petre \\etal 1988), {\\it Ginga} (Wang \\etal 1992), {\\it ROSAT} (Asaoka \\& Aschenbach 1994) and {\\it ASCA} (Keohane \\etal 1997, K97 hereafter). The soft X-ray 0.2--3.1 keV surface brightness map of IC 443 from the {\\it Einstein} Observatory (Petre \\etal  1988) shows bright features in the northeastern part of the remnant as well as bright soft emission from the source central part.  The presence of nearly uniform X-ray emission from the central part of the remnant was clearly seen also by {\\it ROSAT} (Asaoka \\& Aschenbach, 1994), and corresponds to the emission from hot (T$\\sim 10^7$ K), low density gas interior to the shock.  {\\it ASCA GIS} observations by K97 discovered the localized character of the hard  X-ray emission. They concluded that most of the 2-10 keV {\\it GIS} photons came from an isolated emitting feature and from the southeast elongated ridge of hard emission. Preite-Martinez et al. (1999) have reported a hard ($>14$ keV) component with BeppoSAX/PDS and two hot spots corresponding to the ASCA sources of K97 using the MECS.  The supernova remnants (SNRs) that are likely candidates to be $\\gamma$-ray sources in {\\it CGRO EGRET} observations (Esposito \\etal 1996) also show evidence for interaction with molecular gas.  In the case of IC 443, the molecular line emission region is partially inside the {\\it EGRET} \\gr detection circle.   Theoretical models of SNRs interacting with clouds were discussed recently by Sturner \\etal (1997), Baring \\etal (1999), Chevalier (1999). The non-thermal multi-wavelength spectrum of a SNR interacting with a molecular cloud was studied by Bykov \\etal (2000). They showed that the propagation of a radiative shock wave within a molecular cloud lead to a substantial non-thermal emission both in hard X-rays and in $\\gamma$-ray.  The complex structure of molecular cloud consisting of dense massive clumps embedded in the inter-clump medium could result in localized sources of hard X-ray emission correlated with both bright molecular emission and extended source of non-thermal radio and \\gr emission. In this Letter, we present the archive BeppoSAX observations of IC443 and, in particularly, we discuss the features detected in the MECS hard X-ray mosaiced map of the remnant. ",
        "conclusions": "The compact hard X-ray features may be interpreted in terms of the interaction between the remnant shock and a molecular cloud.  Bykov \\etal (2000) have presented a model of non-thermal emission from an evolved SNR interacting with a molecular cloud, but the predicted diffuse hard X-ray emission in the MECS bandwidth is below the sensitivity threshold. On the other hand, several clumps have been observed in the molecular cloud near IC443.  The spatial scales of the molecular clumps from IR line observations are in the range from 150$^{\\prime \\prime}$ down to 1$^{\\prime \\prime}$.  The non-thermal X-rays from clumps were modelled by Bykov \\etal (2000), who predicted a hard spectrum with a photon index below 2.0. The emission is originated from electrons of energy below 1 GeV. The predicted flux density is $\\sim 6\\times 10^{-5}$ keV cm$^{-2}$ s$^{-1}$ keV$^{-1}$ at 10 keV for a 30 $\\kms$ shock in a clump of half parsec radius and density of 10$^4~ \\cmc$.  Using a power law with $\\gamma=1.5$, this corresponds to $\\sim 3\\times 10^{-2}$ MECS cnt s$^{-1}$ in the 4.0--10.5 keV energy range and $\\sim 10^{-2}$ PDS cnt s$^{-1}$ in the 15--30 keV. The predicted count-rates are consistent with the observed rate of Src B (Table \\ref{rates}), having in mind that $10^4$ cm$^{-3}$ density is a representative number and other parameters could easily account for the difference.  Because of hard spectrum and heavy absorption, localized spots of few arcmin size would be seen only in hard X-rays. Thus, the compact Src B correlated with bright spot of molecular hydrogen emission (Fig. \\ref{mecs}) can be shocked molecular clump. The observed soft X-ray brightening having apparent shift from hard X-ray features towards the SNR interior (Fig. \\ref{pspc}), can be attributed to the effects of the shocked clump edges. The morphology and the spectrum of the Src A are somewhat different from that of Src B.  As already discussed by K97, a low-luminosity pulsar nebula is a plausible explanation for isolated hard X-ray sources (especially for the slightly extended Src A), but we do not expect to see two of them in one SNR\\footnote{However, Asaoka \\& Aschenbach (1994) pointed out that the soft X-ray emission is consistent with the presence of two large SNRs, $10^3$ and $10^5$ yr old.}. High resolution Chandra or XMM data are needed to distinguish between pulsar nebula and shocked clump interpretation.  We have also performed a timing analysis of MECS events collected inside a circle of $3^\\prime$ radius in the 4.0--10.5 energy range for both Src A and B, and we found no pulsations at 99\\% confidence level, with an upper limit of the pulsed fraction of a sinusoidal signal in the $10^{-3}-64$ Hz frequency range of 5\\% and 6\\%, respectively, more stringent than the one derived by K97 using ROSAT data.   We have also estimated the Src A and B EGRET \\gr $E>100$ MeV flux contribution, using the best-fit models reported in Table \\ref{bestfit}, and they are $3.0^{+0.5}_{-0.4}\\times 10^{-8}$ and $100^{+380}_{-92}\\times 10^{-8}$ photons cm$^{-2}$ s$^{-1}$. The expected \\gr flux of Src B therefore is compatible with the value of $50.0\\pm 3.0\\times 10^{-8}$ photons cm$^{-2}$ s$^{-1}$ observed by Esposito et al.  (1996). Gamma-ray observations with forthcoming {\\it INTEGRAL} mission with expecting angular resolution about 12$^\\prime$ could better constrain the spectrum of the sources above 30 keV.  \\begin{figure} \\centerline{\\hbox{ \\psfig{figure=srcab_pc_bl30_0.2_2.0_imh_trim_smo.ps,width=8cm} }} \\caption{0.2-2.0 keV PSPC image of the region cotaining the two MECS hard X-ray sources. The hard X-ray contours of Fig. \\protect\\ref{mecs} are overlaid. The arrow shows the soft X-ray counterpart of Src B.}  \\label{pspc} \\end{figure}"
    },
    "0010/astro-ph0010194_arXiv.txt": {
        "abstract": "The influence of a circumbinary (CB) disk on the evolution of cataclysmic variable (CV) binary systems is investigated. We show that CB mass surface densities sufficient to influence the evolution rate are plausibly provided by the outflows observed in CVs, if the net effect of these winds is to deliver $10^{-4}$--$10^{-3}$ of the mass transfer rate to the CB disk. The torque exerted by the CB disk provides a positive feedback between mass transfer rate and CB disk mass which can lead to mass transfer rates of $\\sim 10^{-8} -10^{-7} \\mpy$.  This mechanism may be responsible for causing the range of variation of mass transfer rates in CV's. In particular, it may explain rates inferred for the novalike variables and the supersoft X-ray binary systems observed near the upper edge of the period gap ($P \\sim 3 - 4$ hr), as well as the spread in mass transfer rates above and below the period gap. Consquences and the possible observability of such disks are discussed. ",
        "introduction": "Angular momentum losses play an essential role in the origin and evolution of close binary systems containing compact objects. An excellent example is the class of systems known as cataclysmic variable (CV) binaries in which a low mass star transfers mass to its more massive white dwarf companion. Significant orbital angular momentum must have been lost with ejected mass during the formation process to transform a long period progenitor system into a CV via the common envelope evolutionary phase (see reviews by Iben \\& Livio 1993 and Taam \\& Sandquist 2000). On the other hand, angular momentum loss without significant mass loss is thought to occur during the secular evolution of CV's as a result of gravitational radiation (Paczy\\'nski 1967; Faulkner 1971) or magnetic braking (Verbunt \\& Zwaan 1981) processes.  The need for a process such as magnetic braking for the long term evolution of CV's stems from the fact that the inferred mass transfer rates can exceed that driven by gravitational radiation by more than an order of magnitude. Although the inclusion of such processes in evolutionary models has been modestly successful, detailed comparisons of observations with theory lead to some difficulties. Specifically, it is found that the mass transfer rates primarily depend on the time scale of the angular momentum loss and the mass of the donor with little dependence on the white dwarf companion.  This leads to a strong correlation between mass transfer rate and orbital period. The inferred mass transfer rates do not follow such a tight correlation (Patterson 1984; Warner 1987); there is about an order of magnitude spread at a given orbital period.  This discrepancy has been further aggravated with the discovery of supersoft X-ray sources among the short period cataclysmic variable population. In particular, three sources have been detected with orbital periods near the upper edge of the period gap, i.e., J0439.8-6809 at 3.37 hr (Schmidtke \\& Cowley 1996), J0537.7-7304 at 3 hr (Orio et al. 1997), and 1E 0035.4-7230 at 4.13 hr (Crampton et al. 1997).  Such sources are believed to be systems for which the white dwarf component accretes at rates $\\sim 1 - 4 \\times 10^{-7} \\mpy$, sufficient for steady hydrogen burning to take place on its surface (van den Heuvel et al. 1992).  In order to explain the large spread in mass transfer rates for a given orbital period, suggestions involving intermittent cycles produced by nova explosions (Shara et al. 1986) or by irradiation or mass loss effects (King et al. 1996) have been invoked.  However, the recent lack of detection of the mass losing component in novalike variables in the infrared spectral wavelength region by Dhillon et al. (2000) suggests that the high mass transfer rates in these systems ($\\mdot \\sim 10^{-8} \\mpy$) are secular rather than cyclic. This would cause them to be expanded and cool compared with main sequence stars in the same range of orbital periods. A range in evolutionary states for the mass losing star has also been suggested for the spread in mass transfer rates by Pylyser \\& Savonije (1988), but the dearth of systems inside the period gap is hard to reconcile with this idea.  As a possible resolution to these problems in our understanding of the evolution of CVs, we suggest that a circumbinary (CB) disk can effectively drain orbital angular momentum from the system, promoting mass transfer at rates above those calculated using the magnetic braking or gravitational radiation process. This is in contrast to the self-excited winds picture proposed by King \\& van Teeseling 1998).  In such a situation the orbital angular momentum can be removed from the binary by tidal torques (see Lin \\& Papaloizou 1979; Eggleton \\& Pringle 1985; Pringle 1991). We present here a model incorporating such a disk into the secular evolution of CV systems. In the next section, we present the assumptions of the model and describe the properties of such disks and their critical role for the secular evolution.  The implications of these results are discussed in section 3. There we also address a number of observational puzzles that may be related to the presence of circumbinary disks. In addition to the large spread in mass transfer rates, this includes the low luminosity of the secondaries in novalike systems. We also touch upon the origin of the period gap in CVs, a problem that arguably is aggravated rather than solved by a CB disk picture. ",
        "conclusions": ""
    },
    "0010/astro-ph0010477_arXiv.txt": {
        "abstract": " ",
        "introduction": "NGC~5018, a relatively near, giant E3 galaxy, deserved quite a large attention in the past. Evidence of past interaction with other systems comes mainly from the extended pattern of dust lanes/patches clearly recognizable from ground-based observations. What is more, this object captures immediately the attention of the observer for (i) its anomalous location on the Mg$_2$-luminosity relation of Schweizer {\\em et al.} (1990), having one of the weakest absorption-line strengths (Mg$_2$=0.218) relative to its absolute magnitude and central velocity dispersion among giant ellipticals and (ii) the extremely weak far-UV emission Bertola {\\em et al.} (1993) find in its IUE spectrum (strictly resembling the UV energy distribution of the dwarf elliptical M32). Interesting enough, though one expects ongoing star formation plays a major role in a recent merger such as NGC~5018, both the observed low value of its optical Mg$_2$ line strength index {\\em and} its weak far-UV flux can fit into scenarios {\\em not} invoking recent star formation activity. According to this latter view, NGC~5018 has been alternatively described either as a sort of ``Celacanthus galaxy'', {\\em i.e.} a rare, chemically unevolved giant elliptical hosting a dominating M32-like population (Bertola {\\em et al.} 1993), or a system fully dominated by an intermediate-age (2.8~Gyr) population with a roughly solar composition as put forward by Leonardi \\& Worthey (2000). On the other side, since other groups ({\\em e.g.} Carollo \\& Danziger 1994) are prone to conclude that a large amount of truly young population {\\em and strong dust absorption} might conspire to cause the above UV/optical properties, strict constraints on the amount of internal extinction are urgently needed in order to properly address the issue of the dominant population age and metallicity. ",
        "conclusions": "The above goal can now be achieved thanks to the recent set of Schweizer's WFPC2 F555W and F814W deep images (program ID~7468) which allows us to finally get rid of the persistent suspicion of a {\\em heavy} dust absorption affecting the UV emission of this galaxy. In particular, starting from the high S/N E(V$-$I) map shown in Fig.~1, one derives---after removing the underlying V$-$I color gradient---a luminosity weighted average value E(B$-$V)$\\sim$0.02 due to structured dust within the IUE large aperture (10''$\\times$20''). This, in turn, assures that the collected UV flux is only marginally affected by the internal extinction due to dust (namely $\\sim$0.06~dex at $\\lambda$~1500~\\AA).  \\begin{figure} \\plotfiddle{busonl1.eps}{10.6cm}{0}{86}{86}{-262}{-342} \\caption{Calibrated E(V$-$I) image of the inner portion (18''$\\times$18''; $\\sim$ 2$\\times$ the IUE aperture) of the PC field of NGC~5018 showing its intricate net of dust lanes/patches. The brightest areas correspond to a maximum reddening value E(V$-$I)$\\sim$0.35. The (V$-$I) color of the barely resolved nucleus is 0.3 mag bluer than the underlying stellar population. North is up and East to the left. A set of the F555W image isophotes is superimposed to help locating dust features.} \\end{figure}"
    },
    "0010/astro-ph0010531_arXiv.txt": {
        "abstract": "We report the first results of our multicolour broad-band BVRI photometry for a statistically well defined sample of interacting pairs, drawn from the Catalogue of Isolated Pairs of Galaxies in the Northern Hemisphere Karachentsev (1972). At present, the details of photometric structure of disk-disk (S+S) galaxies are just beginning to be studied because of their intrinsic difficulties. We take advantage of the statistical properties of this well defined sample of interacting galaxies to try to isolate the main structural components of galaxies. Due to their environmental simplicity (compared to groups and clusters), the sample provides an unique opportunity to peform a photometric study of galaxies and their structural components in a non-equilibrium configuration. In this contribution, the first results of a deep multicolour (BVRI) photometric study of a sample of 45 (S+S) isolated pairs of galaxies is presented. Our attention is focused on the morphology and the global photometric properties (integrated magnitudes, colours and mean surface brightness profiles). ",
        "introduction": " ",
        "conclusions": ""
    },
    "0010/hep-ph0010004_arXiv.txt": {
        "abstract": "Including contributions of scale-dependent vacuum expectation values, we derive new analytic formulas and obtain substantially different numerical predictions for the running  masses of quarks and charged-leptons at higher scales in the SM, 2HDM and MSSM. These formulas exhibit significantly different behaviours with respect to their dependence on gauge and Yukawa couplings than those derived earlier. At one-loop level the masses of the first two generations are found to be independent of Yukawa couplings of the third generation in all the three effective theories in the small mixing limit. Analytic formulas are also obtained for running $\\tan\\beta(\\mu)$ in 2HDM and MSSM. Other numerical analyses include study of the third generation masses at high scales  as functions of low-energy values of $\\tan\\beta$ and SUSY scale $M_S=M_Z-10^4$ GeV. ",
        "introduction": "\\label{sec1} \\par One of the most attractive features of current investigations in gauge theories is the remarkable unification of the gauge couplings of the standard model (SM) at the SUSY GUT scale, $M_U=2\\times 10^{16}$ GeV, when extrapolated through the minimal supersymmetric standard model (MSSM) \\cite{1}. Although the nonsupersymmetric standard model (SM), or the two-Higgs doublet model (2HDM) do not answer the question of gauge hierarchy, unification of the gauge couplings is also possible at the corresponding GUT scales when they are embedded in nonSUSY theories like $SO(10)$ and the symmetry breaking takes place in two steps with left-right models as intermediate gauge symmetries \\cite{2}. Grand unification of gauge couplings of the SM in single-step breakings of GUTs has also been observed when the grand desert contains additional scalar degrees of freedom \\cite{3} and the minimal example is a $\\xi(3,0,8)$ of SM contained in $\\underline{75}\\subset SU(5)$ or $\\underline{210}\\subset SO(10)$ with mass $M_{\\xi}=10^{11}-10^{13}$ GeV \\cite{4}. Unification of gauge couplings in nonSUSY $SO(10)$ has been demonstrated with relatively large GUT-threshold effects \\cite{5}. Yukawa coupling unification at the intermediate scale has also been observed in nonSUSY $SO(10)$ with 2HDM as the weak scale effective gauge theory \\cite{6}. Apart from unity of forces at high scales, SM, 2HDM and MSSM have tremendous current importance as effective theories as they emerge from a large class of fundamental theories. \\par Recent experimental evidences in favour of neutrino masses and mixings have triggered an outburst of models many of which require running masses and mixings of quarks and charged-leptons at high scales as inputs for obtaining predictions in the neutrino sector \\cite{7}. The running masses are not only essential at the weak scale, but they are also required at the intermediate and the GUT scales in order to testify theories based upon quark-lepton unification with different Yukawa textures and for providing unified explanation of all fermion masses \\cite{8,9,10,11,12}. Quite recently extrapolation of running masses and couplings to high scales have been emphasized as an essential requirement for testing more fundamental theories \\cite{12}. \\par In a recent paper one of us (M.K.P) and Purkayastha \\cite{13} have obtained new analytic formulas and numerical estimations for the fermion masses at higher scales in MSSM including contributions of scale-dependent vacuum expectation values (VEVs) where the SUSY scale $(M_S)$ was assumed to be close to the weak scale $(M_S\\approx M_Z)$. In this paper we extend such investigations to SM, 2HDM and MSSM with the SUSY scale $M_S\\ge O$ (TeV). \\par It is also possible that in a different renormalisation scheme, similar to that formulated by Sirlin et.~al.~\\cite{14}, the VEVs themselves do not run when they are expressed in terms of physical parameters defined on the mass shell. This makes it possible to avoid separate running of the VEVs and Yukawa couplings, but have just the fermion masses directly as running quantities. While it would be quite interesting to examine the consequences of such a scheme, the purpose of the present and the recent works \\cite{13} is to address the outcome of the most frequently exploited renormalisation scheme where the Yukawa couplings and the VEVs run separately \\cite{15,16,17,18,19,20,21,22,23}. \\par This paper is organised in the following manner. In Sec.~\\ref{sec2} we cite examples where running VEVs have been exploited by a number of authors and state relevant renormalisation group equations (RGEs). In Sec.~\\ref{sec3} we derive analytic formulas. In Sec.~\\ref{sec4} we show how the formulas derived earlier for MSSM are modified when $M_S \\gg M_Z$. Numerical predictions at higher scales are reported in Sec.~\\ref{sec5}. Summary and conclusions are stated in Sec.~\\ref{sec6}. \\par ",
        "conclusions": "\\label{sec6} \\par In the frequently exploited renormalisation scheme in gauge theories, the Yukawa couplings and VEVs in the SM, 2HDM and MSSM run separately \\cite{15,16,17,18,19,20,21,22,23,24,25,26,27,28,29}. The effect of scale dependence of the VEVs has been ignored while deriving analytic formulas \\cite{28,30,31} and obtaining numerical predictions at higher scales for the running masses of fermions \\cite{32}, but appropriately taken into account more recently \\cite{13}. In this paper we have derived new analytic formulas in the SM and 2HDM and generalised the formulas of \\cite{13} for any supersymmetry scale $(M_S>M_Z)$. The new formulas exhibit substantially different functional dependence on gauge and Yukawa couplings in all the three effective theories. In particular, the running masses of the first two generations are found to be independent of Yukawa couplings of the third generations in the small mixing limit. Numerical predictions at two-loop level shows that all the running masses in the SM and only the up-quark masses in the MSSM and 2HDM decrease at high scales when compared with the predictions taking scale-independent VEVs. But in the case of MSSM and 2HDM the down-quark and the charged-lepton masses increase over the corresponding predictions obtained with scale-independent assumptions on the VEVs. Compared to the MSSM the perturbatively allowed region of $\\tan\\beta$ is larger in 2HDM. In MSSM the allowed region shows a slow increase with the SUSY scale. We have also made predictions of the running masses at the GUT scale as a function of supersymmetry scale exhibiting new behaviours. We suggest that these new analytic formulas and improved estimations on the running masses and $\\tan\\beta$ at high scales be used as inputs to test models proposing unified explanations of quark-lepton masses. \\par"
    },
    "0010/astro-ph0010582_arXiv.txt": {
        "abstract": "Observations and theories of the stellar initial mass function are reviewed. The universality and large total mass range of the power law portion of the IMF suggest a connection between stellar mass and cloud structure.  The fractal patterns and formation times of young clusters suggest a further connection to turbulence.  There is also a similarity between the characteristic mass at the lower end of the power law and the minimum self-gravitating mass in a typical star-forming cloud. All of this suggests that the power law part of the IMF comes from star formation in self-gravitating cloud pieces that are formed by compressible turbulence.  Timing constraints involving cloud destruction and competition for gas might limit the upper stellar mass to several hundred suns. This upper limit has to be less than the mass of a clump that has a dynamical time equal to several times the dynamical time of the characteristic mass.  The smallest stars and brown dwarfs may not come directly from cloud clumps, which are not self-gravitating at this mass in the cloudy phase, but from collapsed fragments or other pieces connected with the formation of more massive stars. ",
        "introduction": "The stellar initial mass function (IMF) is an important ingredient for studies of stellar evolution in clusters and galaxies, and it offers a clue to the physical processes involved with star formation. But since the time when the first IMF was extracted from field stars corrected for age (Salpeter 1955), there has been no general understanding of its origin, and few direct observations of stellar mass functions in their initial form, i.e., in very young clusters.  Fortunately, the observational situation is improving with infrared surveys down to sub-stellar masses in nearby young clusters (Comer\\'on, et al. 1993; Strom, Kepner, \\& Strom 1995; Lada \\& Lada 1995; Luhman, et al. 1998; Luhman \\& Reike 1998, 1999; Najita, Tiede \\& Carr 2000) and with ground- and space-based observations of distant massive clusters (e.g., Massey \\& Hunter 1998; Grillmair et al. 1998; Selman et al. 1999; Sirianni et al. 2000; see reviews in Massey 1998 and Brown 1998).  The theory of the IMF is making progress too. There are numerical simulations that reproduce the IMF by following the motion and accretion of protostars in a model cluster (Bonnell et al. 1997) or the evolution of protostar interactions (Price \\& Podsiadlowski 1995), and there are simulations based on sampling from fractal clouds (Elmegreen 1997, 1999a, 2000a,c; S\\'anchez \\& Parravano 1999). There are also analytical models of the IMF that include many physical processes involved with star formation (e.g., Silk 1995; Nakano, Hasegawa, \\& Norman 1995; Adams \\& Fatuzzo 1996; Myers 2000a).  Several reviews of the IMF are in the conference series {\\it The Stellar Initial Mass Function}, edited by G. Gilmore and D. Howell (Cambridge University Press, 1998). Cayrel (1990) has an earlier review the theory of the IMF, and Leitherer (1999) has a recent summary of IMF observations in starburst galaxies. Other reviews are in Elmegreen (1998; 1999c; 2000d).  This paper discusses the basic classes of IMF models and various observational constraints for them. These constraints suggest that the IMF is approximately uniform in diverse environments, the star formation process is relatively rapid and possibly independent for each star, even in clusters, and that stars sample the IMF randomly in each cloud. The recent observation of IMF-like mass functions in pre-stellar condensations suggests further that the IMF may be decided in the gaseous phase, and not be part of the collapse process during which this gas gets converted into stars. These observations limit the physical processes that can be involved in determining the IMF. ",
        "conclusions": "The IMF has many properties suggestive of a connection to turbulence. It is universal, stochastic, and power-law. Star formation also has dynamic and structural properties similar to turbulence: the duration of star formation in a cloudy structure is usually only a few crossing times, regardless of scale, and starbirth positions are often fractal.  We have found that virtually any mechanism for star formation that operates on a dynamical time scale in an initially fractal cloud will produce the Salpeter IMF at intermediate to high mass. This IMF is a consequence of random sampling from the various mass structures in the cloud, and not a consequence of the star formation processes. Only the mass at the lower limit of the power law part of the IMF, and the masses at the upper and lower limits, should depend on the detailed physics of star formation. The power law part could be determined in the gas phase.  The low-mass break in the power law part of the IMF might result from a lack of self-gravity in tiny clumps. This break point should be higher in regions with higher temperatures and/or lower pressures. The flat-to-falling part of the IMF below the break may come from fragmentation or other secondary processes inside the dense clumps or disks where stars form, or from pressure fluctuations affecting the Jeans mass. That is, low-mass stars and brown dwarfs should form along with other more massive stars in the smallest cloud clumps. The lack of supermassive stars in galaxies was considered to be a constraint on the IMF. The IMF should fall off more rapidly than the Salpeter slope at high mass because of timing limitations and the Eddington limit. For the timing limit, clumps whose internal dynamical timescales are longer than several times the dynamical time of a clump with the characteristic mass, $M_J$, should not proceed directly to form a single star. Smaller stars should form in these clumps first, making the gas unavailable for the formation of super-massive stars.  Observations of variations in the IMF in different environments would be a good check on the theories. Extreme environments such as high-temperature and high-pressure starburst regions, remote or low-pressure field regions, highly compressed or triggered regions, ultracold gas, and high-pressure cluster cooling flows, might shift the characteristic mass for star formation up or down, depending on the thermal Jeans mass. If this is the only change that is observed in these regions, then models where the power-law part of the IMF comes from universal cloud geometry would be supported. If the power law slope of the observed IMF changes with environment, however, and not just as a result of mass segregation or selective cloud destruction discussed above, then the IMF would not have the universal and scale-free aspect that is expected from turbulence. Other IMF models that are more dependent on the detailed processes of star formation would be required."
    },
    "0010/astro-ph0010061_arXiv.txt": {
        "abstract": "The Galactic star forming region in the southern sky, associated with IRAS 09002-4732 has been mapped simultaneously in two far infrared bands ($\\lambda_{eff}$ = 148 \\& 209 $\\mu$m), with $\\sim$ 1\\arcmin ~angular resolution. Fifteen sources, including IRAS 08583-4719, 08589-4714, 09002-4732 and 09014-4736 have been detected, some of which are well resolved. Taking advantage of similar beams in the two bands, a reliable dust temperature [T(148/209)] map has been obtained, which detects colder dust ($<$ 30 K) in this region. The HIRES processed IRAS maps at 12, 25, 60 and 100 $\\mu$m, have also been used for comparison. The optical depth maps, $\\tau_{200}$ and $\\tau_{100}$, generated from these far-IR data quantify the spatial distribution of the dust. The diffuse emission from this entire region has been found to be 35 \\% of the total FIR luminosity. The slope of the IMF in the mass range 4--16 $M_{\\odot}$ has been estimated to be $-1.25^{+0.75}_{-0.65}$ for this star forming complex.  Radiative transfer models in spherical geometry have been explored to fit available observations of the 4 IRAS sources and extract various physical parameters for corresponding dust-gas clouds. Whereas a constant ($r^{0}$) radial density distribution is favoured in IRAS 08583-4719, 08589-4714 and 09002-4732, the $r^{-1}$ law is inferred for IRAS 09014-4736. The dust composition is found to be similar (Silicate dominated) in all the four IRAS sources modelled. The luminosity per unit mass is found to be in the narrow range of  44 -- 81 $L_{\\odot}/M_{\\odot}$  for these star forming regions. ",
        "introduction": "The study of far infrared (FIR) emission from the relatively denser regions of the Galactic interstellar medium (ISM) is useful for understanding the star formation process. High spatial resolution mapping of Galactic molecular clouds in the FIR continuum (in $\\ge$ 2 bands) provides direct information about the physical conditions (e.g. total luminosity, dust distribution, dust temperature etc.) of the region. For a constant gas-to-dust ratio, the dust optical depth traces matter along the line of sight. A comparison of the dust temperature distribution with that of the gas temperature (obtained from molecular line measurements) can give important information on the dust-gas coupling in star forming regions. For medium to high mass embedded YSOs, dust is the dominant source of heating of the gas (outside the \\ion{H}{ii} region). A FIR study of an individual star formation complex (at a known distance) can measure the luminosity function above a threshold which can usually be translated to the present day mass function and even initial mass function.  Systematic differences can be expected between star forming regions at different galactocentric distances, since the star formation process depends on the physical conditions of the local ISM (e.g. density, composition, radiation field etc.). This important issue has been addressed by many authors using various techniques. Systematic studies in radio continuum and molecular line emission based on colour selected samples from the IRAS Point Source Catalog (hereafter IRAS PSC) have led to very useful databases which can be the starting point for further studies (Wood \\& Churchwell \\cite{WC89a}; Wouterloot \\& Brand \\cite{Wo89}).  Although the IRAS mission has proved to be a corner stone in improving our understanding of star formation in the Galaxy, its sub-optimal angular resolution in the 60 and 100 $\\mu$m bands has been a major limitation. In addition, recent ISO measurements of nearby spiral galaxies have shown the importance of emission at wavelengths $>$ 100 $\\mu$m and colder dust (T $<$ 25 K) seems to be a dominant constituent of spiral galaxies in general (Alton et al. \\cite{Al98}). This suggests a similar role for colder dust in our Galaxy also.  In view of the above, the TIFR 1-metre balloon borne telescope is flown regularly to map Galactic star forming regions simultaneously in two FIR bands with near diffraction limited angular resolution. These FIR bands have been chosen to be trans-IRAS ($\\lambda_{eff}$ = 148~ \\& ~209 $\\mu$m), so that the distribution of interstellar dust cooler than $\\approx$ 30 K which might have been missed by IRAS, can be traced (Ghosh \\cite{Gh00}, Mookerjea et al. \\cite{Mo99}, \\cite{Mo00a}, \\cite{Mo00b}). Accessibility of more than 90 \\% of the Galactic plane ($l$ = 315${}^{0}$ -- 0${}^{0}$ -- 290${}^{0}$) from the Indian balloon launching station has been exploited by targeting the southern sky, which is observationally less explored in the FIR.  The catalogue of CO emission associated with colour selected IRAS PSC sources by Wouterloot \\& Brand (\\cite{Wo89}; hereafter WB89) covers the longitude range of the Galactic plane of interest to us. Based on information from this catalogue, the star forming complex around IRAS 09002-4732 was selected for the present study because : (i) all the four bright IRAS PSC sources in this region have similar kinematic distances (WB89; Sugitani \\& Ogura \\cite{Su94}); three of them (IRAS 08589-4714, 09002-4732 and 09014-4736) are known to be physically associated, hence this complex presents an opportunity to derive its PDMF / IMF if a reasonable number of sources are detected; (ii) it lies marginally outside the solar circle, hence a comparison of the physical conditions in this region with that in the 4--5 kpc molecular region and regions inside the solar circle would be interesting; (iii) the complex is very close to the  Galactic mid plane with $z \\approx 20$ pc (WB89) as compared to the scale height of 90 pc for the embedded OB stars (Wood \\& Churchwell \\cite{WC89b}). This region around IRAS 09002-4732 is associated with the very extended Vela Molecular Ridge detected in the wide angle survey of $^{12}$CO emission by May et al. (\\cite{May88}) and Murphy and May (\\cite{Mur91}), and more recently mapped by Yamaguchi et al. (\\cite{Ya99}) in both $^{12}$CO and $^{13}$CO lines.  Most of the available observational data for this region pertain to IRAS 09002-4732 itself and clearly indicate the presence of an embedded young stellar object in a dense molecular cloud. These include measurements of radio continuum (Caswell \\& Haynes \\cite{Ca87}; Manchester \\& Goss \\cite{Ma69}; Whiteoak \\cite{Wh92}; Walsh et al. \\cite{Wa98}), infrared spectra from the IRAS Low Resolution Spectrometer Catalog (hereafter LRS; Olnon \\& Raimond, \\cite{Ol86}), ISO Long Wavelength Spectrometer data (ISO-LWS; Clegg et al. \\cite{Cl96}) and photometric and imaging measurements in the near and mid infrared (Lenzen \\cite{Le91}; Walsh et al. \\cite{Wa99}). In addition, millimeter wave continuum emission (Reipurth et al. \\cite{Re96}) and molecular line detection of CS have been reported (Zinchenko et al. \\cite{Zi95}). A recent multi transitional study of IRAS 09002-4732 has led to the detection of bipolar outflow emission in the CS line with distinct red and blue shifted peaks displaced symmetrically relative to the central IR object (Lapinov et al. \\cite{La98}). However, no bipolar emission in the CO line has been seen by them.   The observations of a large region (area $>$ 500 sq. arc min) centered at IRAS 09002-4732 are presented here. This Galactic \\ion{H}{ii} region / molecular cloud complex has been studied with the aim of understanding the energetics, physical sizes, spatial distribution of interstellar dust and its temperature. The next section describes the observations and the results alongwith the discussions are presented in section 3. ",
        "conclusions": "\\subsection{148 $\\mu$m and 209 $\\mu$m bands}  The MEM deconvolved TIFR maps at 148 and 209 $\\mu$m of the Galactic star forming region associated with IRAS 09002-4732 are presented in Fig. 2 (a \\& b). The insets show the deconvolved intensity contours for the point-like source (Jupiter), to quantify the achieved angular resolution. The dotted lines in these figures mark the boundary of the region covered by the telescope bore-sight. However the actual observations extend beyond this boundary due to the sky chopping and the effective size of the PSF array. As a result, the intensity maps generated by the MEM deconvolution extend further by 3\\farcm 4 along in-scan and 6\\arcmin ~along cross-scan directions. The limits for reliable extraction of discrete sources and structural information in the form of isophot contours have been established by using the data from two independent sets of raster scans. The dynamic range has been found to be $\\approx$ 200 for both bands. Several regions of extended emission are seen in both the 148 and the 209 $\\mu$m maps. A total of 15 sources have been detected, details of which are listed in Table 2 and positions marked on the maps. The four most prominent peaks are associated with IRAS PSC sources, viz., 08583-4719, 08589-4714, 09002-4732 and 09014-4736. The listed flux densities have been obtained by integrating over a circle of 3\\arcmin ~diameter.  The positions of the global peaks in both the 148 and 209 $\\mu$m maps are identical to that of IRAS 09002-4732. The direction of the major extended emission seen from IRAS 09002-4732 in both the TIFR maps [size at 10 \\% (50 \\%) of peak : 3\\farcm 8 $\\times$ 2\\farcm 9 (1\\farcm 8 $\\times$ 1\\farcm 6) at both these bands] matches quite well with mm-wave molecular line maps. The CS (2--1) map of Zinchenko et al. (\\cite{Zi95}) shows a FWHM size of $\\approx$ 2\\arcmin $\\times$ 1\\farcm 5 with the major axis at almost identical position angle as the FIR extension. Higher angular resolution map by Lapinov et al. (\\cite{La98}) in the CS (7--6) line resolves the central region into two peaks with different LSR velocities nearly symmetrically displaced relative to the FIR peak. The line joining these two peaks also has a similar position angle.  \\subsection{IRAS bands}  The HIRES processed maps in all the four IRAS bands for the region of present study are shown in Fig. 3 (a,b,c \\& d). The dynamic range of these HIRES maps is larger than that of the TIFR maps, and the isophots have been shown down to 0.1\\% of respective peak intensities. Due to this larger dynamic range, the diffuse emission is more prominently visible in these HIRES maps.  The angular resolutions achieved in these maps (in-scan $\\times$ cross-scan) are 0\\farcm 52 $\\times$0\\farcm 63, 0\\farcm 58$\\times$0\\farcm 70, 1\\farcm 15$\\times$1\\farcm 35  and 1\\farcm 87$\\times$ 1\\farcm 97 at 12, 25, 60 and 100 $\\mu$m respectively.  \\begin{figure*} \\resizebox{\\hsize}{!}{\\includegraphics{9891.f3.eps}} \\caption{ The HIRES processed IRAS maps for a similar region as shown in Fig. 2, in the four bands -- {\\bf (a)} at 12 $\\mu$m with peak = 374 Jy/sq. arcmin, {\\bf (b)} at 25 $\\mu$m  with peak = 5470 Jy/sq. arcmin. {\\bf (c)} at 60 $\\mu$m  with peak = 12300 Jy/sq. arcmin. {\\bf (d)} at 100 $\\mu$m  with peak = 5830 Jy/sq. arcmin. The isophot contour levels in (a) \\& (b) are 90, 30, 10, 3, 1, .3, \\& .1 \\% and in (c) \\& (d) are 80, 40, 20, 10, 5, 2.5, 1, .5, .25 \\& .1 \\% of the respective peaks. } \\end{figure*}   A new algorithm has been used to extract sources in the HIRES processed IRAS survey maps. The scheme works on the local maxima in the intensity maps as candidates and models the local background (encompassing the source of an assumed size, which is a variable : 1--4\\arcmin ~) and subtracts the same. The flux densities (3\\arcmin ~diameter) for these HIRES sources are also presented in Table 2.  \\subsection{Discrete sources }  Out of the total 15 FIR sources that have been detected from the TIFR maps, 12 are common to both the 209 $\\mu$m and 148 $\\mu$m bands (Table 2). All these 15 sources are found to be associated with sources extracted from the HIRES maps in at least one band (13 in at least 2 HIRES bands). The positional match demanded is within 1\\farcm 5 ~for the purpose of association. Four of these are associated with IRAS PSC sources whose SEDs are studied in greater detail (Sect. 3.6). The IRAS PSC flux densities for these four sources are also listed in Table 2. The dust temperatures in the FIR, T${}_{FIR}$, have been computed from the flux densities in the TIFR bands, assuming an emissivity law of $\\epsilon_{\\lambda} \\propto {\\lambda}^{-2}$. These are also listed in Table 2.  An attempt has been made to determine $\\gamma$, the slope of the IMF ($dN/d(log M) \\propto M^{\\gamma}$), for the region around IRAS 09002-4732, under reasonable assumptions. We assume all the 15 sources to be nearly at the same distance based on the finding of WB89 (3 out of the 4 IRAS PSC sources in this region for which CO emission has been found, are at distances within $\\pm$ 10 \\%). If each source represents one embedded ZAMS star, and their luminosities are translated to stellar masses using Tout et al. (\\cite{To96}), then the local slope of the IMF, $\\gamma$ has been found to be $-1.25^{+0.75}_{-0.65}$ for the mass range 4--16 $M_{\\odot}$. The effect of incompleteness near the sensitivity limit has been avoided by appropriate choice of the mass bins based on the source number distribution. This slope is similar to those found for Galactic OB associations (Massey et al. \\cite{Ma95}) and clusters (Phelps \\& Janes \\cite{Ph93}).   \\begin{table*} \\caption{Position and flux density details of the detected sources.} \\vskip 0.5cm \\begin{tabular}{ccccccccccc} \\hline \\#& RA & Dec & IRAS PSC&\\multicolumn{6}{c}{Flux Density (Jy)} & T${}_{FIR}^{b}$ \\\\ \\cline{5-10} & (1950)    &(1950)  &  associations & \\multicolumn{2}{c}{TIFR} & \\multicolumn{4}{c}{IRAS$^a$} &\\\\ & h ~~m ~~s & o ~~\\arcmin ~~\\arcsec &&    209 $\\mu$m & 148 $\\mu$m& 100 $\\mu$m& 60 $\\mu$m& 25 $\\mu$m& 12 $\\mu$m & (K) \\\\ \\hline S1& 8 58 15.0 & -47 18 44& 08583-4719 & 54 &209 & 1242 & 1309 & 233 & 44 &$>$45\\\\ &\"&\"&\"&&&          $<$1692$^c$ & $<$469$^c$ &    34.1$^c$ & 6.8$^c$ &\\\\ S2& 8 58 36.9 & -47 20 27&                 & 70 &--  &  767 &   -- &  36 & 25.3 &\\\\ S3& 8 58 59.0 & -47 13 35& 08589-4714 & 142& 195&  680 & 365 & 30.1&  15.1 &21\\\\ &\"&\"&\"&&&          551$^c$ & 200$^c$ &    13.5$^c$ & 2.2$^c$ &\\\\ S4& 8 59 13.7 & -47 14 57&                 &  --& 179&  459 &   -- &  -- & 10.7 &\\\\ S5& 8 59 15.3 & -47 27 50&                 &  90& 118&   -- & 44 &  -- &  3.7 &20 \\\\ S6& 8 59 16.6 & -47 17 05&                 &  83& 155&   -- & 110 &  -- & 10.9 & 27\\\\ S7& 8 59 22.8 & -47 28 51&                 & 102& 144&  197 & 55 &  -- &  5.3 & 21\\\\ S8& 8 59 26.1 & -47 26 08&                 &  72& 121&   -- & 43 &  -- &  -- & 24\\\\ S9& 8 59 30.5 & -47 28 07&                 & 102& 111&  187 & 54 &  -- & 3.4 & 18\\\\ S10& 8 59 32.4 & -47 23 08&                 &  65& 102&   53 & 43 &  -- &  -- & 23\\\\ S11& 8 59 38.3 & -47 24 18&                 &  --& 134&   53 & 43 &  -- &  -- & \\\\ S12& 8 59 45.0 & -47 35 00&                 & 200& 384&  111 & 46 &  17 &  6.3 & 28\\\\ S13& 9  0 12.1 & -47 32 07& 09002-4732 & 4753& 9127& 14057 & 13611& 2577& 203 & 28\\\\ &\"&\"&\"&&&        14707$^{c}$ & 11877$^{c}$ &   1962$^{c}$ & 121$^{c}$ &\\\\ S14& 9  0 40.7 & -47 29 06&                 &  66& 111  & --  & 27 &  --  & -- & 25\\\\ S15& 9  1 39.1 & -47 37 00& 09014-4736 & 282& 607  & 902 & 795 &  100 &  52 & 33 \\\\ &\"&\"&\"&&&          957$^{c}$ & 621$^{c}$ &    79.6$^{c}$ & 30.0$^{c}$ &\\\\ \\hline \\end{tabular} \\vskip 0.3cm ${}^a$ From HIRES processed maps unless specified otherwise. The flux densities are integrated over a circle of 3\\farcm 0 diameter.  ${}^b$ Determined using the flux densities in TIFR bands and assuming a grey body spectrum with emissivity $\\epsilon_{\\lambda} \\propto {\\lambda}^{-2}$.  ${}^c$ From IRAS Point Source Catalog.  \\end{table*}  \\subsection{Diffuse emission}  Using the T(148/209) temperature distribution for the entire region mapped (Sect. 3.5), and the intensity maps in the same region, the total far-infrared luminosity has been estimated, assuming the dust emissivity law to be $\\epsilon_{\\lambda} \\propto \\lambda^{-2}$. For the regions of the intensity maps with S/N $<$ 3 in either band, a fixed value of 10 K has been assumed for the dust temperature. The contribution of diffuse emission in this complex is found to be $\\sim$2.8$\\times$10$^{4}$ $L_{\\odot}$, which is 35 \\% of the total emission. This does not include contribution from any diffuse (structureless at $<$ 10\\arcmin) component which may be associated with the Vela Molecular Ridge clouds along the line of sight. This fraction is in agreement with the values found for young Galactic star forming regions, viz., 25 \\% for Orion A region (Mookerjea et al. \\cite{Mo00b}); 35 \\% for W 31 complex (Ghosh et al. \\cite{Gh89}).   \\subsection{Distribution of dust temperature and optical depth}   Although the HIRES processed IRAS maps have a much higher dynamic range, the angular resolution of the TIFR maps is superior to the IRAS maps at 60 and 100 $\\mu$m, due to the smaller and circular beams employed. Taking advantage of the nearly identical beams of the TIFR bands at 148 and 209 $\\mu$m and the simultaneity of observations, we have generated reliable maps of dust temperature and optical depth (at 200 $\\mu$m, $\\tau_{200}$; translated from 209 $\\mu$m), assuming a dust emissivity of $\\epsilon_{\\lambda}\\propto\\lambda^{-2}$ following the same procedure as described in Mookerjea et al. (\\cite{Mo99}, \\cite{Mo00b}). The dust temperatures are calculated based on the assumption that the dust is optically thin which is reasonable at these wavelengths. Considering the uncertainties in the flux densities measured by us and the wavelength regime probed by us, the temperature computed has an uncertainty of $\\sim$ $\\pm$2K, between 15 K and 35 K which gradually increases to $\\pm$5K at 50 K.  \\begin{figure} \\resizebox{\\hsize}{!}{\\includegraphics{9891.f4.eps}} \\caption{ The distribution of dust temperature T(148/209), and optical depth at 200 $\\mu$m, $\\tau_{200}$, for the region around IRAS 09002-4732 (A) assuming a dust emissivity law of $\\epsilon_{\\lambda} \\propto \\lambda^{-2}$. The isotherms correspond to 15 K to 50 K in steps of 5 K. Temperature values for a few contours are displayed to avoid ambiguity. The $\\tau_{200}$ contours represent 90, 70, 50, 30, 20, 10, 5, 2 \\& 1 \\% of the peak value of 0.051. The cross marks the position of IRAS 09002-4732. } \\end{figure}   Two regions with relatively stronger FIR emission have been selected out of the entire mapped area for presenting the T(148/209) and $\\tau_{200}$ distribution. These two regions cover the target source IRAS 09002-4732 (hereafter, A) and the neighbouring source IRAS 09014-4736 (B). The spatial distribution of T(148/209) as isotherm contours and contours of constant $\\tau_{200}$, for the regions A and B are presented in Figs. 4 and 5 respectively.  The HIRES maps at 60 and 100 $\\mu$m trace the distribution of somewhat warmer dust (30--50 K). We have used these to generate maps of the colour temperature, $T(60/100)$, and optical depth ($\\tau_{100}$) for the region A (around IRAS 09002-4732). The intensity maps at 60 and 100 $\\mu$m were spatially averaged before computing $T(60/100)$ and $\\tau_{100}$ in a manner similar to that described by Ghosh et al. (\\cite{Gh93}) for an emissivity law of ${\\epsilon}_{\\lambda} \\propto \\lambda^{-1}$. This emissivity law is more appropriate for the 60--100 $\\mu$m wavelength range. The isotherms of T(60/100) and contours of constant $\\tau_{100}$ are presented in Fig. 6 (a \\& b).  The position of the intensity peaks in both the TIFR as well as all four IRAS bands for the region A are identical to that of IRAS PSC coordinates of IRAS 09002-4732, which has been marked on Figs. 4 and 6. It may be noted here that the positions of the associated radio continuum source at 843 MHz (Whiteoak \\cite{Wh92}) and H$_2$O maser source are within 15\\arcsec ~of IRAS 09002-4732 (Lapinov et al. \\cite{La98}).  The T(148/209) temperature distribution is very asymmetrical around IRAS 09002-4732. Hotter dust and a steeper temperature gradient are seen towards the north-west (NW) of IRAS 09002-4732 and a relatively smoother temperature distribution in the range of 20 -- 30 K is observed towards the south-east. The band of hotter dust is situated $\\sim$ 2\\farcm 2 to the NW of the peak position of the FIR intensity. The dust temperature drops from a value of $\\sim$ 50 K to 20 K within a distance of $\\approx$ 0.8 pc. Similar large variations in dust temperature has been seen for the regions associated with IRAS 00494+5617 and IRAS 05327-0457 (Mookerjea et al. \\cite{Mo00a}).  The T(60/100) map shows one major peak, coinciding with the position of IRAS 09002-4732 and the temperature drops (to 25 K) on all sides almost symmetrically except for the region to the north. The values of T(60/100) would come closer to T(148/209), if the dust emissivity exponent is chosen to be -2 (see Fig. 6 caption). The lack of complex structure in this T(60/100) map vis-a-vis T(148/209) map may be due to the poorer angular resolution of the IRAS maps.  The distribution of optical depth ($\\tau_{200}$) around IRAS 09002-4732, somewhat resembles the intensity distribution : the peak position matches with that of $I_{148}$ or $I_{209}$. The peak $\\tau_{200}$ value is 0.051. There is a secondary peak in this map shifted towards the NW (beyond the band of higher temperature), positionally matching with relatively cooler dust [T(148/209) $\\sim$ 20 -- 25 K].  \\begin{figure} \\resizebox{\\hsize}{!}{\\includegraphics{9891.f5.eps}} \\caption{ Same as in Fig. 4, but for the region around IRAS 09014-4736 (B). The isotherms refer to the same temperatures as in Fig. 4. The $\\tau_{200}$ contours represent 10, 5, 2 \\& 1 \\% of the global peak value of 0.051. The cross marks the position of the TIFR source (S15) associated with IRAS 09014-4736. } \\end{figure}   The $\\tau_{100}$  distribution is similar to that of $\\tau_{200}$ only near the peak of the intensity maps. This may not be surprising, since the IRAS and TIFR maps are probing dust column densities at different temperatures. The highest value of $\\tau_{100}$ occurs $\\sim$ 2\\arcmin ~south-west of IRAS 09002-4732. The value of $\\tau_{100}$ near the peak of the intensity maps, is 0.023.  The T(148/209) temperature distribution in IRAS 09014-4736 also shows asymmetric structures (Fig. 5a). A ridge of hot dust runs approximately east-west $\\sim$ 1\\farcm 5 away (south) from the peak of FIR intensity distribution. In addition, a local hotspot is seen due north-west (1\\farcm 2) from the intensity peak. The $\\tau_{200}$ distribution shows one major peak positionally corresponding to $\\sim$ 20 K dust, and an isolated minor peak at the  position of the hotspot.  \\begin{figure} \\resizebox{\\hsize}{!}{\\includegraphics{9891.f6.eps}} \\caption{ The {\\bf (a)} dust temperature T(60/100) and {\\bf (b)} optical depth at 100 $\\mu$m ($\\tau_{100}$) distribution from the HIRES maps for the region around IRAS 09002-4732 (A) assuming a dust emissivity law of $\\epsilon_{\\lambda} \\propto \\lambda^{-1}$. The isotherms displayed in (a) correspond to the temperatures 70, 60, 50, 40, 30 \\& 25 K. The innermost contour is for the hottest dust (70 K). (The isotherms would represent 50, 45, 39, 33, 26 \\& 22 K if the emissivity index was -2). The contours in the optical depth map (b) represent 50, 40, 30, 20 \\&  10 \\%  of the global peak value of 0.11. The cross marks the position of IRAS 09002-4732. } \\end{figure}  \\subsection{Radiative transfer modelling}  With the aim of extracting important physical parameters, an attempt has been made to construct radiative transfer models for the four IRAS PSC sources IRAS 09002-4732, 09014-4736, 08583-4719 and 08589-4714, which have been covered in our FIR maps. The self consistent scheme developed by Mookerjea \\& Ghosh (\\cite{MG99}) has been used for this purpose which is briefly described below. This scheme models spherically symmetric cloud with centrally embedded young stellar object/(s), immersed in an average interstellar radiation field. Although the gas-dust coupling is not considered explicitly, the role of the gas in the energetics of the cloud has been included in a self consistent manner. Two commonly used types of interstellar dust have been explored (Draine \\& Lee \\cite{Dr84}, hereafter DL; Mezger et al. \\cite{Me82}, hereafter MMP), with a power law size distribution given by Mathis et al. (\\cite{Ma77}). For each type of dust, the relative proportion of graphite and astronomical silicate grains is selectable. The observational constraints include the SED due to the dust, angular sizes at different wavelengths and the radio continuum emission.  The following parameters are explored in order to get an acceptable fit to all the data : (i) the nature of the embedded source, which could either be a single ZAMS star or a cluster of ZAMS stars consistent with the Salpeter Initial Mass Function; (ii) radial density distribution law (only three power laws have been explored, viz., $n(r) \\propto r^{\\alpha}$, with $\\alpha$ = 0, $-1$ or $-2$); (iii) the relative abundances of the two constituent grain types; (iv) total radial optical depth due to the dust (inclusive of all constituents) at a selected wavelength ($\\tau_{100}$ at 100 $\\mu$m); (v) the gas to dust ratio by mass (the predicted radio continuum emission is sensitive to this); (vi) geometric details of the dust cloud (e.g. cavity size, outer size of the cloud).  The distances to all the four IRAS PSC sources in the present study, have been taken to be 1.4 kpc.  \\subsubsection{IRAS 09002-4732}  The observed SED of IRAS 09002-4732 has been constructed using TIFR, IRAS-HIRES, IRAS-LRS and ISO-LWS data (Fig. 7a). In addition, 1.3 mm continuum measurement of Reipurth et al. (\\cite{Re96}) and near infrared data from Lenzen (\\cite{Le91}) have been used. The total radio continuum emission has been found to be 3.8 Jy at 5 GHz (Caswell \\& Haynes \\cite{Ca87}) from this source in a wide beam (4\\farcm 4). Whiteoak (\\cite{Wh92}) found the flux density at 843 MHz to be 2.1 Jy. Recent high angular resolution (synthesized beam $\\approx$ 2\\arcsec ~) measurement for the central region by Walsh et al. (\\cite{Wa98}) at 8.64 GHz, gives a peak intensity of 0.5 Jy/beam. The total luminosity and the rate of Lyman continuum photon generation has been estimated to be 6.9 $\\times 10^{4} L_{\\odot}$ and 7.1 $\\times$ $10^{47}$ s${}^{-1}$ respectively by Simpson \\& Rubin (\\cite{Si90}). By integrating the observed SED, we estimate the total luminosity to be 1 $\\times 10^{5} L_{\\odot}$. In a study of a sample of ultracompact \\ion{H}{ii} regions which includes IRAS 09002-4732, Walsh et al. (\\cite{Wa99}) have fitted the infrared part of the SED by representing it as a two-component black body spectrum and also by a radiative transfer model. Our self consistent modelling scheme includes angular sizes at different wavelengths, radio continuum data as well as dust composition and grain size distribution.  The best radiative transfer model for IRAS 09002-4732 has been identified by varying the parameters and comparing the following predictions with observations : (i) the SED originating from the dust, (ii) the angular sizes at different wavebands, and (iii) the radio continuum emission from the gas component. The predicted angular sizes have been convolved with the relevant telescope beams, prior to comparison with observed FWHM sizes. The embedded source is found to be a single ZAMS O7 star with $N_{Lyc}$ = 3.6 $\\times 10^{48}$ s$^{-1}$ and $T_{eff}$ = 38,500 K. The preferred type of dust is DL and the density distribution is uniform. The gas to dust ratio is 100 : 1 by mass. The other parameters of the best fit model for IRAS 09002-4732 are presented in Table 3.   \\begin{table*} \\caption{Parameters for the best fit radiative transfer models.} \\vskip 0.5cm \\begin{tabular}{ccccccccc} \\hline Source & $\\alpha$& $R_{max}$& $R_{min}$& $r_{H II}$& $\\tau_{100}$&Luminosity & Dust composition$^{a}$ &$M_{dust}$\\\\ &&(pc)&(pc)&(pc)&&($10^{3} L_{\\odot}$)& Silicate:Graphite & $(M_{\\odot}$)\\\\ \\hline IRAS 08583-4719& 0.0&0.3& $5.0\\times 10^{-5}$& $2.9\\times 10^{-3}$ & 0.02& 4.0 & 81 : 19 &0.50\\\\ IRAS 08589-4714& 0.0&0.2& $8.0\\times 10^{-5}$& $1.2\\times 10^{-3}$& 0.05& 2.4 & 81 : 19 &0.55\\\\ IRAS 09002-4732&  0.0&1.0&$3.0\\times 10^{-2}$& $5.5\\times 10^{-2}$& 0.05& 100 & 72 : 28 &14\\\\ IRAS 09014-4736& -1.0&0.8& $8.0\\times 10^{-5}$& $2.9\\times 10^{-5}$ & 0.03& 6.5 & 81 : 19 &0.80\\\\ \\hline \\end{tabular} \\vskip 0.3cm $^{a}$Dust is of DL type for all the four sources. \\end{table*}  The predicted SED from this model is compared with the observations in Fig. 7 (a). The model fits the data quite well. The predicted total radio continuum emissions (1.5 Jy at 843 MHz; 3.7 Jy at 5 GHz \\& 3.5 Jy at 8.6 GHz) are also in reasonable agreement with observations. For our model, the \\ion{H}{ii} region extends into the dust cloud beyond the dust-free central cavity ($r_{H II} > R_{min}$), and its size is of similar order as inferred from the 8.6 GHz map by Walsh et al. (\\cite{Wa98}). A comparison of the predicted angular sizes at mid to far infrared wavelengths with the measurements is presented in Table 4. The predicted angular sizes increase as a function of the wavelength and the values are in general agreement within errors.  The IRAS LRS catalog has classified IRAS 09002-4732 to be of type 81 based on the strengths of emission of the unidentified 11.3 $\\mu$m feature and the [Ne II] line at 12.8 $\\mu$m. Our inference of an embedded O7 star from the above modelling is consistent with the detection of the [Ne II] line. However, the detection of 11.3 $\\mu$m feature, generally attributed to be arising from single photon absorption processes of PAH molecules, indicates importance of non-equilibrium heating processes, which is beyond the scope of the present study.  The total mass of our model cloud is 1400 $M_{\\odot}$ (gas to dust ratio = 100) which can be compared with inferences from molecular line observations. The virial mass estimate from the CO data of WB89 and using the relation of Evans (\\cite{Ev99}; see equation 8), we get the cloud mass to be 1710 $M_{\\odot}$ which is reasonably close to the mass from our model considering the uncertainties. On the other hand, the mass estimate from CS line measurements is 607 $M_{\\odot}$ (Zinchenko et al. \\cite{Zi95}). The $(L/M)$ ratio of the total IR luminosity to cloud mass from the model, the virial mass based on CO data, and the CS core mass, are 71, 58 and 165  $L_{\\odot}/M_{\\odot}$ respectively, for IRAS 09002-4732.  \\begin{table} \\caption{Comparison of predicted sizes from the best fit model for IRAS 09002-4732 with observed  FWHMs.} \\vskip 0.5cm \\begin{tabular}{|c|c|c|} \\hline Wavelength & Model & Observations \\\\ ($\\mu$m) & (~\\arcmin ~) & (~\\arcmin ~) \\\\ \\hline 12  & 0.59 & 0.54 \\\\ 25  & 0.61 & 0.62 \\\\ 60  & 1.07 & 0.89 \\\\ 100 & 1.96 & 1.33 \\\\ 148 & 2.4 & 1.8 \\\\ 209 & 2.5 & 1.8 \\\\ \\hline \\end{tabular} \\end{table}   \\subsubsection{IRAS 09014-4736}  The FIR emission from IRAS 09014-4736 at 148 and 209 $\\mu$m is quite extended as evident from Fig. 2  and it is listed in the IRAS Small Scale Structure Catalog (SSSC; X0901-476) due to its extended emission in 25 and 60 $\\mu$m bands. It is gratifying to note that the flux densities listed in the SSSC (109 Jy at 25 $\\mu$m and 794 Jy at 60 $\\mu$m) match extremely well with flux densities (100 and 795 Jy respectively; see S15 in Table 2) determined using our source extraction scheme from the HIRES processed maps (Sect. 3.2). The angular sizes estimated in SSSC are : 1\\arcmin ~at 25 $\\mu$m and 1\\farcm 6 at 60 $\\mu$m. The FWHM sizes (major $\\times$ minor) in the 148 and 209 $\\mu$m bands are 3\\farcm 5 $\\times$ 2\\farcm 7 and 3\\farcm 8 $\\times$ 2\\farcm 5 respectively. The size increases with wavelength as expected.  The observed SED for IRAS 09014-4736 has been compiled using the TIFR, IRAS-HIRES and IRAS-LRS data (Fig. 7b). No radio continuum data for this source are available in the literature. However, this region has been covered by the 843 MHz survey with 43\\arcsec ~angular resolution (Whiteoak \\cite{Wh92}). The quoted completeness limit of 50 mJy by Whiteoak (\\cite{Wh92}) has been used here as an upper limit. Modelling the SED of IRAS 09014-4736 leads to a better fit for an $r^{-1}$ density distribution law than for an $r^0$ law. The embedded source is found to be a ZAMS star of B1 type ($T_{eff}$ = 22,600 K). The preferred dust is of DL type with the radial optical depth, $\\tau_{100}$ = 0.03. The other parameters corresponding to the best fit model for IRAS 09014-4736 are listed in Table 3. The spectral fit of this model to the observed SED is displayed in Fig. 7 (b) which is reasonably good. There is hardly any detectable radio continuum emission as expected for a B1 type star.  Near and mid-IR spectrophotometry of IRAS 09014-4736 shows recombination line emission of hydrogen (Persson \\& Campbell \\cite{Pe87}; Beck et al. \\cite{Be91}; Porter et al. \\cite{Po98}). The LRS spectrum of this source has been classified as type 80 based on the detection of 11.3 $\\mu$m feature and absence of any fine structure line from ionized heavier elements, typical of low excitation \\ion{H}{ii} region. Detection of several additional PAH features (Cohen et al. \\cite{Co89}; de Muizon et al. \\cite{Mu90}; Zavagno et al. \\cite{Za92}) further supports that IRAS 09014-4736 has an embedded YSO. Although IRAS 09014-4736 is classified as an unidentified source in the IRAS PSC, it is situated within $\\approx$ 20\\arcsec ~of the small optical nebula, BBW 225 (size $<$ 2\\arcmin ~), identified by Brand et al. (\\cite{Br86}). The embedded YSO (B1) inferred from our modelling is consistent with all of the above. However, the actual geometry of the source has to be far from spherically symmetric, in order to reconcile with the size of the optical nebula and the near infrared data. Similar conclusions have been arrived at by Beck et al. (\\cite{Be91}) for IRAS 09014-4736.  The only available molecular line observation from this source is the CO emission (WB89). We estimate the molecular gas density and the virial mass for the cloud associated with IRAS 09014-4736 to be 5.2 $\\times$ 10$^3$ cm$^{-3}$ and 395 $M_{\\odot}$ respectively. The mass of the model cloud is 80 $M_{\\odot}$ for a gas to dust mass ratio of 100. The latter is not constrained well due to the lack of radio continuum emission. The $(L/M)$ ratio corresponding to the model and the virial mass of this cloud are 81 and 16 $L_{\\odot}/M_{\\odot}$ respectively.  \\begin{figure*} \\resizebox{\\hsize}{!}{\\includegraphics{9891.f7.eps}} \\caption{ Comparison of the predicted spectral energy distributions (SEDs) from the best fit models of {\\bf (a)} IRAS 09002-4732, {\\bf (b)} IRAS 09014-4736 {\\bf (c)} IRAS 08583-4719 \\& {\\bf (d)} IRAS 08589-4714 with observations. The filled circles represent the TIFR and IRAS-HIRES data. The dashed lines are IRAS-LRS spectra (for IRAS 09002-4732 and IRAS 09014-4736) and the dash dot line is the ISO-LWS spectrum of IRAS 09002-4732. The triangles denote 1.3 mm data from Reipurth et al. (\\cite{Re96}) for IRAS 09002-4732 and near-IR data from Lenzen (\\cite{Le91}) for IRAS 09002-4732, from Persson \\& Campbell (\\cite{Pe87}) for IRAS 09014-4736, from Lorenzetti et al. (\\cite{Lo93}) for IRAS 08589-4714. The solid line denotes the prediction of the best fit model. See text and Table 3 for details of the model parameters. } \\end{figure*}  \\subsubsection{IRAS 08589-4714}  In addition to the measurements presented here, the near-IR data of Lorenzetti et al. (\\cite{Lo93}) have been used to construct the SED of IRAS 08589-4714 (Fig. 7c). The parameters of the best fit radiative transfer model are presented in Table 3. A constant density cloud with DL type dust grains is implied. The total radial $\\tau_{100}$ = 0.05 and the model cloud mass is 55 $M_{\\odot}$. Since the energizing stellar source is detectable in the near-IR, this is an example of very evolved stage of star formation, where only a small fraction of the original interstellar matter is left around the star at present.  The source IRAS 08589-4714 has been detected in CO and CS surveys by Wouterloot \\& Brand (\\cite{Wo89}) and Bronfman et al. (\\cite{Br96}) respectively. Whereas our model gives a value for the $L/M$ ratio to be $\\approx$ 44 $L_{\\odot}/M_{\\odot}$, the virial mass of the cloud estimated from the CO data of WB89 indicates a value of 3.7 $L_{\\odot}/M_{\\odot}$.  \\subsubsection{IRAS 08583-4719}  IRAS 08583-4719 is contained in the catalog of bright rimmed clouds which are candidates for star formation by radiation driven implosion (Sugitani \\& Ogura \\cite{Su94}). Unfortunately no additional observational data pertaining to a SED are available other than those presented here. The physical size estimate has been taken from Sugitani \\& Ogura (\\cite{Su94}). Once again a constant density dust (DL) cloud gives the best fit to the SED (Fig. 7d). However, the fit to the observations even for this best model is not very satisfactory. The parameters for this model are listed in Table 3. The mass of the model cloud is 50 $M_{\\odot}$ leading to a $(L/M)$ ratio of 80 $L_{\\odot}/M_{\\odot}$.  \\subsubsection{Inferences from the modelling}  The DL type of dust is favoured for all the four sources modelled here. It may be noted that the emissivity index in the 100 -- 200 $\\mu$m range for DL type dust (for both silicate and graphite), is $\\approx$ -2, consistent with our choice for generation of T(148/209) maps (Sect. 3.5). Three of these four sources suggest a constant density dust envelopes around the YSOs. This conclusion seems to be very common for several Galactic star forming regions : e.g. for IRAS 18314-0720, 18355-0532 and 18316-0602 (Mookerjea \\& Ghosh, \\cite{MG99}); IRAS 00338+6312, 04004-5114 and RAFGL 5111 (Mookerjea et al. \\cite{Mo99}); IRAS 00494+5617 and 05327-0457 (Mookerjea et al. \\cite{Mo00a}). The following comments can be made by comparing the parameters for the four sources modelled here : The dust cloud corresponding to IRAS 09002-4732 has the largest size of the central cavity, which contains the most luminous embedded source. Except IRAS 09014-4736, for all other three sources, the \\ion{H}{ii} regions extend well into the respective dust clouds. This emphasizes the role of dust grains on the emission of recombination lines and radio continuum. The radial optical depth due to the dust ranges between 0.02 to 0.05 which is slightly on the lower side compared to IRAS 00494+5617 (Mookerjea et al. \\cite{Mo00a}), 00338+6312 (Mookerjea et al. \\cite{Mo99}). The interstellar dust composition is found to be nearly identical in all the four sources modelled, viz., Silicate dominated. The ratio between the graphite and the silicate grains, in general vary drastically among the Galactic star forming regions (Mookerjea et al. \\cite{Mo99}, \\cite{Mo00a}). Hence, nearly identical dust composition found here further supports the hypothesis that the entire complex around IRAS 09002-4732 mapped by us is physically associated.  The $L/M$ ratio for the four sources studied here, as found from our radiative transfer modelling, falls in a narrow range of 44 -- 81 $L_{\\odot}/M_{\\odot}$. For three of these for which CO measurements are available, this ratio based on virial masses, fall in somewhat broader range (3.7 -- 58 $L_{\\odot}/M_{\\odot}$). These can be compared with the results from other studies : (i) A study similar to the present one, for the W 31 complex has lead to a value of $(L/M)$ in the range of 1.6 -- 3.6 $L_{\\odot}/M_{\\odot}$ (Ghosh et al. \\cite{Gh89}); (ii) the average value for a large sample of clouds using virial mass based on CO measurements, is $\\sim$ 4 $L_{\\odot}/M_{\\odot}$ with a spread exceeding a factor of 100 (Mooney \\& Solomon \\cite{Mo88}, Evans \\cite{Ev91}); (iii) for denser cores studied in CS emission (Plume et al. \\cite{Pl97}) give $\\sim$ 190 $L_{\\odot}/M_{\\odot}$ with a spread of a factor of 15. Considering the uncertainties arising due to the fact that the different tracers are sampling different components of the ISM (at possibly different spatial scales), the star forming complex associated with IRAS 09002-4732 can be termed as typical of the Galaxy."
    },
    "0010/astro-ph0010257_arXiv.txt": {
        "abstract": "We model the radio, optical, and $X$-ray emission for the afterglows of GRB 980703, 990123, 990510, and 991216, within the framework of relativistic jets, to determine their physical parameters. The models that yield acceptable fits to the data have jet energies mostly between $10^{50}$ to $10^{51}$ erg and initial opening angles between $1\\deg$ and $4\\deg$. The external medium density is uncertain by at least one order of magnitude in each case, being around $10^{-3}\\cm3$ for GRB 980703 and 990123,  $\\sim 10^{-1}\\cm3$ for GRB 990510, and $\\sim 3 \\cm3$ for GRB 991216. If the jets are uniform (\\ie there are no angular gradients of the energy per solid angle) then the 20 keV -- 1 MeV radiative efficiency during the GRB phase must have been at least 2-3\\% for GRB 990510, 20\\% for GRB 990123, and 30\\% for GRB 991216. ",
        "introduction": "There are two basic quantities one needs to try understanding the nature of any astronomical source -- the distance and the energy of the source. For the long duration GRBs, lasting more than 10 seconds, the former is well established to be cosmological. However, the energy associated with the GRBs remains uncertain.  The efficiency of producing $\\gamma$-ray emission in the generally accepted internal shock model (\\Meszaros \\& Rees 1994), which can explain the observed temporal variability, is less than a few percent (Kumar 1999, Lazzati, Ghisellini \\& Celotti 1999, Panaitescu, Spada \\& \\Meszaros 1999, see also Beloborodov 2000). This suggests that the total energy in the explosion is larger than the energy observed in $\\gamma$-rays by a factor of 20--50.  The goal of this work is to infer the physical parameters of the ejecta, including their energy and the external medium density, from modeling of radio, optical and $X$-ray data for GRB afterglows with known redshifts. The modeling is carried out in the framework of collimated ejecta interacting with an isotropic external medium. The model is described in \\S\\ref{model} and the results of the numerical calculations for individual afterglows are given in \\S\\ref{aglows}. ",
        "conclusions": "We model the emission of GRB afterglows in the framework of collimated, uniform, lateral spreading jets interacting with an external medium. The model was used to determine the initial jet energy $E_0$, opening angle $\\theta_0$, external medium density $n$, and parameters $\\epsel$ and $\\epsmag$ quantizing the minimum random Lorentz factor of shock-accelerated electrons and the strength of the magnetic field, respectively, for the afterglows of GRB 980703, 990123, 990510, and 991216.  As illustrated in Figures 5 and 6, the jet aperture $\\theta_0$ and the index $p$ of the power-law distribution of electrons injected in the down-stream region are well constrained by observations, as the effects of collimation are seen at a time which depends strongly on $\\theta_0$ (\\eq [\\ref{tjet}]), while $p$ determines the afterglow decay rate (\\eqs [\\ref{Fs0}], [\\ref{Fs2}], [\\ref{Fjet}]). Other model parameters -- $E_0$, $n$, and $\\epsmag$ -- are less well determined as observations in three frequency domains (radio, optical, and $X$-ray) provide at most three constraints. Observations in a fourth domain, sub-millimeter, millimeter, or far-infrared, could help determine all the afterglow parameters, provided that the spectral breaks are located between adjacent observational frequencies.  Analysis of the available data for the 990123 and 990510 afterglows and the jet interpretation of the break exhibited by their optical emissions rule out a wind-like profile for the medium which decelerates the jet. However, the emission of the 980703 and 991216 afterglows can be accommodated by both types of external media (homogeneous or a pre-ejected wind). For 980703, 990123, and 990510, the allowed range of external densities is below $1 \\cm3$, indicating that these bursts did not occur in hydrogen clouds.  For those afterglows with optical light-curve breaks we find jet energies lying mostly in the $10^{50}$--$10^{51}$ erg range (Figure 7), and jet initial half-angles below $4\\deg$. For the afterglow of GRB 970508, the only one observed in radio, optical and $X$-ray that is not included in this work due to its non-standard optical and $X$-ray brightening, Frail, Waxman \\& Kulkarni (2000) found a similar jet energy, $E_0 = 5\\times 10^{50}$ erg, a much larger initial half-angle, $\\theta_0 \\sim 30\\deg$, and an external density $n \\sim 1 \\cm3$.  The minimum BATSE range (20 keV -- 1 MeV) efficiency of the afterglows whose optical behavior indicates that the ejecta was well collimated ranges from 3\\% to 30\\% (Figure 7). The former limit is within the reach of current calculations of internal shock efficiency (Kumar 1999, Lazzati \\etal 1999, Panaitescu, Spada \\& \\Meszaros 1999), but the latter exceeds it. However, if the energy distribution within the jet aperture is far from isotropy, such that the GRBs we see have an energy-per-solid angle peaking in the direction toward the observer, then the minimum required efficiency can be significantly smaller."
    },
    "0010/astro-ph0010127_arXiv.txt": {
        "abstract": "We have obtained Near-Infrared Camera and Multi-Object Spectrometer images of 16 radio quiet quasars observed as part of a project to investigate the ``luminosity/host-mass limit.''  The limit results were presented in McLeod, Rieke, \\& Storrie-Lombardi (1999). In this paper, we present the images themselves, along with 1- and 2-dimensional analyses of the host galaxy properties. We find that our model-independent 1D technique is reliable for use on ground-based data at low redshifts; that many radio-quiet quasars live in deVaucouleurs-law hosts, although some of the techniques used to determine host type are questionable; that complex structure is found in many of the hosts, but that there are some hosts that are very smooth and symmetric; and that the nuclei radiate at $\\sim2-20\\%$ of the Eddington rate based on the assumption that all galaxies have central black holes with a constant mass fraction of 0.6\\%.  Despite targeting hard-to-resolve hosts, we have failed to find any that imply super-Eddington accretion rates. ",
        "introduction": "\\newcommand\\cola {\\null} \\newcommand\\colb {&} \\newcommand\\colc {&} \\newcommand\\cold {&} \\newcommand\\cole {&} \\newcommand\\colf {&} \\newcommand\\colg {&} \\newcommand\\colh {&} \\newcommand\\coli {&} \\newcommand\\colj {&} \\newcommand\\colk {&} \\newcommand\\coll {&} \\newcommand\\colm {&} \\newcommand\\coln {&} \\newcommand\\colo {&} \\newcommand\\colp {&} \\newcommand\\colq {&} \\newcommand\\colr {&} \\newcommand\\cols {&} \\newcommand\\colt {&} \\newcommand\\colu {&} \\newcommand\\colv {&} \\newcommand\\colw {&} \\newcommand\\colx {&} \\newcommand\\coly {&} \\newcommand\\colz {&} \\newcommand\\colA {&} \\newcommand\\eol{\\\\}    Host galaxy studies got the opportunity for a real boost in February 1997 when the Near-Infrared Camera and Multi-Object Spectrometer (NICMOS) was installed on the Hubble Space Telescope (HST). NICMOS combines the superb spatial resolution of HST with the benefits that long wavelengths provide for imaging the redder hosts against the overwhelming glare of the bluer quasar nuclei. We have used NICMOS to image 16 radio quiet quasars as part of a project to investigate the ``luminosity/host-mass limit,'' the results of which were presented in McLeod, Rieke, \\& Storrie-Lombardi (1999; hereafter MRS).  In this paper, we present the images themselves, along with 1- and 2-dimensional analyses of the host galaxy properties.  The sample, listed in Table \\ref{tab-1d}, is composed of all 10 quasars from our ``high-luminosity sample'' (the 26 highest-luminosity PG quasars with $z<0.3$; \\cite{mr94b}) that had not been previously observed with HST.  To this we added 6 luminous quasars out to $z=0.4$ for which ground-based attempts to resolve a host galaxy had failed.  All 16 objects are in the redshift range $0.13 < z < 0.40$ with an average $z=0.25$.  \\begin{deluxetable}{llllllllll} \\scriptsize \\tablecaption{Results of 1D analysis\\tablenotemark{a}\\label{tab-1d}} \\tablehead{ \\colhead{Name} & \\colhead{$z$} & \\colhead{$m_H$} & \\colhead{$M_H$} & \\colhead{$m_H$} & \\colhead{$M_H$}  & \\colhead{$M_B$} & \\colhead{$F_{host}/F_{nuc}$}&\\colhead{$L/L_{Edd}\\tablenotemark{c}$} \\\\ & & \\colhead{(nuc)} & \\colhead{(nuc)} & \\colhead{(host)} & \\colhead{(host)\\tablenotemark{b}} & \\colhead{(nuc)\\tablenotemark{b}} } \\startdata  PG0026+129 & 0.142 &  13.3 & -25.5 &  15.2 & -23.6 & -23.7 &  0.17 & 0.16 \\\\ PG0947+396\\tablenotemark{d} & 0.206 &  14.3 & -25.3 &  15.4 & -24.2 &  -23.1 &  0.37 & 0.05 \\\\ PG1048+342 & 0.167 &  15.5 & -23.7 &  15.1 & -24.1 & -23.2 &  1.48 & 0.07 \\\\ PG1121+422\\tablenotemark{d} & 0.224 &  14.6 & -25.3 & 17.0\\tablenotemark{e} & -22.8\\tablenotemark{e} &  -23.7 &  0.11 & 0.31\\tablenotemark{e} \\\\ PG1151+117 & 0.176 &  14.6 & -24.7 &  15.9 & -23.3 & -23.6 &  0.29 & 0.19 \\\\ PG1322+659 & 0.168 &  14.3 & -24.9 &  15.6 & -23.5 & -23.2 &  0.29 & 0.10 \\\\ PG1352+183 & 0.158 &  15.0 & -24.0 &  15.1 & -23.9 & -23.2 &  0.92 & 0.07 \\\\ PG1354+213 & 0.300 &  15.7 & -24.9 &  16.1 & -24.4 & -24.6 &  0.73 & 0.16 \\\\ PG1427+480\\tablenotemark{d} & 0.221 &  15.0 & -24.9 &  16.1 & -23.7 &  -23.3 &  0.35 & 0.11 \\\\ PG2233+134 & 0.325 &  15.2 & -25.6 &  16.7 & -24.0 & -24.3 &  0.26 & 0.20 \\\\ MARK876    & 0.129 &  13.6 & -25.0 &  13.3 & -25.3 & -23.0 &  1.36 & 0.02 \\\\ UM357      & 0.334 &  15.1 & -25.8 &  16.7 & -24.0 & -24.2 &  0.22 & 0.17 \\\\ Q0530-379  & 0.334 &  15.5 & -25.4 &  16.9 & -23.9 & -23.4 &  0.28 & 0.09 \\\\ NAB1612+26 & 0.395 &  15.6 & -25.6 &  17.3 & -23.9 & -23.4 &  0.22 & 0.10 \\\\ 1628.6+3806\\tablenotemark{d} & 0.394 &  15.2 & -26.0 &  16.7 & -24.4 & -23.7 &  0.26 & 0.08 \\\\ KUV18217+6419\\tablenotemark{d} & 0.297 &  12.3 & -28.2 &  14.7 &-25.7& -26.0 &  0.11 & 0.20 \\\\ \\enddata \\tablenotetext{a}{$H_0=\\rm 80~km~s^{-1}~Mpc^{-1},~q_0=0$} \\tablenotetext{b}{Includes k-correction.} \\tablenotetext{c}{Using $M_B=M_H-2.1-2.5[log_{10}(L/L_{edd})+log_{10}({\\Upsilon_{\\rm V}\\over{7.2M_\\odot/L_\\odot}})+log_{10}({f\\over{0.006}})-log_{10}({BC\\over{12}})]$ with $\\Upsilon_{\\rm V}, f,~\\rm and~BC$ given by their default values (MRS)} \\tablenotetext{d}{Used PSF0026 instead of quasar's own PSF star} \\tablenotetext{e}{Detection is uncertain} \\end{deluxetable}  We have used a zero point of 21.80 magnitudes in the Vega system for the F160W filter (\\cite{leh00}), and refer to the resulting magnitudes simply as ``H.''  We note that this is fainter by 0.31~mag than the value we used in MRS. For computing rest-frame properties of the hosts, we assume $H_0=\\rm 80~km~s^{-1}~Mpc^{-1},~q_0=0,~\\Lambda_0=0$ throughout, but our results are not strongly sensitive to cosmology.  For example, at $z=0.4$, the difference in proper distance between $q_0=0$ and 1 is 10\\%. We also apply galaxy H-band k-corrections appropriate for star-forming galaxies, but note these amount to less than 0.1 magnitude for the redshifts in our sample.  For the nuclei, we have used colors and k-corrections from Cristiani \\& Vio (1990).  To compute $M_B(nuc)$, we have assumed that in the B-band all of the light belongs to the nucleus. This is a reasonable assumption for these high-luminosity quasars because (i) we know {\\it ex post facto} that the galaxies are generally less luminous than the nuclei in the H-band, and (ii) the galaxy contribution relative to the nucleus falls dramatically at shorter wavelengths (see e.g. McLeod \\& Rieke 1995, Fig. 1). ",
        "conclusions": "We have imaged 16 low-redshift, high-luminosity quasars with NICMOS. Using stellar images as PSFs has worked extremely well, and NICMOS has done a superb job of showing the host galaxies with high contrast against the bright nucleus.  We find the following results.  \\begin{itemize}  \\item [(i)] For redshifts $z\\lesssim0.3$, the host-galaxy magnitudes derived from ground-based data using our 1D analysis technique are very reliable (within $\\sim 0.30~\\rm mag$) unless there are companions that cannot be resolved.  Therefore, general conclusions about host luminosities from our previous papers should be robust.  \\item [(ii)] Distinguishing between deVaucouleurs and exponential hosts based on various fitting techniques is questionable, especially given that neither law is likely perfect.  For 2D fits, we recommend a very conservative $\\chi^2$ criterion for discrimination.  While some previously used methods appear to distinguish between types in almost every case, they are often at odds with the more conservative criterion, which can distinguish between types in only $\\sim3$ cases out of 16.  The 1D radial profiles are also very useful.  Fortunately, HST allows us to look directly for spiral arms and other morphological features in the hosts.  We concur with previous studies that have found radio-quiet quasars often live in deVaucouleurs hosts.  \\item [(iii)] Approximately 4 of the 16 hosts are undergoing strong interactions with companions.  However, nearly half of the hosts are smooth and symmetric, a reminder that current-epoch interactions are not a necessary condition of quasar activity.  \\item [(iv)] Assuming that galaxies contain central black holes with 0.6\\% of the galaxy's mass (see MRS for details), 15 of the 16 quasars in our sample radiate at $\\sim2-20\\%$ of the Eddington rate.  Our host detection for PG1121+422 is uncertain, so its rate of 31\\% Eddington may be a lower limit.  Despite intentionally choosing high-luminosity quasars whose hosts were hard to detect from the ground, we have failed to find any object that violates a luminosity/host-mass limit corresponding to $L=L_{Edd}.$  \\end{itemize}"
    },
    "0010/astro-ph0010305_arXiv.txt": {
        "abstract": "s{ We use the two-point correlation function in redshift space, $\\xi(s)$, to study the clustering of the galaxies and groups of the Nearby Optical Galaxy (NOG) Sample, which is a nearly all-sky, complete, magnitude-limited sample of $\\sim$7000 bright and nearby optical galaxies. The correlation function of galaxies is well-described by a power-law, $\\xi(s)= (s/s_0)^{-\\gamma}$, with $\\gamma\\sim1.5$ and $s_0\\sim 6.4\\; h^{-1}$ Mpc. We find evidence of morphological segregation between early- and late-type galaxies, with a gradual decreasing of the strength of clustering from the S0 to the late-type spirals, on intermediate scales. Furthermore, luminous galaxies (with $M_B\\leq -19.5 + 5 \\log h$) are more clustered than dim galaxies. The groups show an excess of clustering with respect to galaxies. Groups with greater velocity dispersions, sizes, and masses are more clustered than those with lower values of these quantities.} ",
        "introduction": "Following  a series of studies (Marinoni {\\it et al.}~\\cite {ma98}, Marinoni {\\it et al.}~\\cite{ma99}, Giuricin {\\it et al.}~\\cite{gi00a}; Marinoni {\\it et al.}~\\cite{ma00}) in which we investigate on the large-scale galaxy distribution in the nearby universe by using a nearly all-sky sample of optical galaxies, in this paper we analyze the clustering of the galaxies and groups of the Nearby Optical Galaxy (NOG) sample (Giuricin {\\it et al.}~\\cite{gi00a}).    The NOG is a complete, distance-limited ($cz\\leq 6000$ km/s) and magnitude-limited ($B\\leq 14$ mag) sample of $\\sim 7000$ nearby and bright optical galaxies, which covers $\\sim2/3$ ($|b|>20^{\\circ}$) of the sky (8.27 sr). The degree of redshift completeness of the NOG is estimated to be 97\\%.  The $B$-magnitudes are homogenized total blue magnitudes transformed to the standard system of the RC3 catalog (de Vaucouleurs {\\it et al.}~\\cite{de91}) and fully corrected for Galactic extinction, internal extinction, and K-dimming. Groups have been identified within the NOG by means of the hierarchical (H) and percolation (P) algorithms. About 45\\% of the NOG galaxies were found to be group members.  In this paper we use  the redshift-space two-point correlation function to analyze the clustering of the NOG galaxies (7028 galaxies with $cz\\geq$ 50 km/s), the NOG H groups (474 groups with at least three members), and the NOG P groups (506 groups with at least three members). Here we analyze the P groups obtained with the variant of the P algorithm in which the distance link parameter, corresponding to a minimum number density contrast of 80, is scaled with distance (it is $0.67\\;h^{-1}$ Mpc at 4000 km/s), whilst the velocity link parameter is kept constant at the value of 350 km/s (here we adopt $H_0 = 100\\; h$ km s$^{-1}$ Mpc$^{-1}$). The variant of the P algorithm in which both link parameters are scaled with distance leads to very similar groups (see Giuricin {\\it et al.}~\\cite{gi00a} for details on group selection).  Almost all NOG galaxies (98.7\\%) have a morphological classification. We divide NOG galaxies into two broad morphological bins, early-type galaxies (E-S0, $T<-1.5$) and late-type galaxies (Sp, $T\\geq -1.5$) and into six fine morphological bins, E ($T<-2.5$), S0 ($-2.5\\leq T < -1.5$), S0/a ($-1.5\\leq T <0.5$), Sa ($0.5 \\leq T < 2.5$), Sb ($2.5 \\leq T < 3.5$), and later types (hereafter denoted as Scd). The earliest bin, hereafter denoted simply as E, comprise also lenticulars, since it contains also objects broadly classified as E-S0. ",
        "conclusions": ""
    },
    "0010/astro-ph0010419_arXiv.txt": {
        "abstract": "\\noindent We measure the radio luminosity function (RLF) of steep-spectrum radio sources using three redshift surveys of flux-limited samples selected at low (151 \\& 178 MHz) radio frequency, low--frequency source counts and the local RLF. The redshift surveys used are the new 7C Redshift Survey (7CRS) and the brighter 3CRR and 6CE surveys totalling 356 sources with virtually complete redshift $z$ information. This yields unprecedented coverage of the radio luminosity versus $z$ plane for steep-spectrum sources, and hence the most accurate measurements of the steep-spectrum RLF yet made.  We find that a simple dual-population model for the RLF fits the data well, requiring differential density evolution (with $z$) for the two populations. The low--luminosity population can be associated with radio galaxies with weak emission lines, and includes sources with both FRI and FRII radio structures; its comoving space density $\\rho$ rises by about one dex between $z \\sim 0$ and $z \\sim 1$ but cannot yet be meaningfully constrained at higher redshifts.  The high--luminosity population can be associated with radio galaxies and quasars with strong emission lines, and consists almost exclusively of sources with FRII radio structure; its $\\rho$ rises by nearly three dex between $z \\sim 0$ and $z \\sim 2$.  These results mirror the situation seen in X-ray and optically-selected samples of AGN where: (i) low luminosity objects exhibit a gradual rise in $\\rho$ with $z$ which crudely matches the rises seen in the rates of global star formation and galaxy mergers; and (ii) the density of high luminosity objects rises much more dramatically. The integrated radio luminosity density of the combination of the two populations is controlled by the value of $\\rho$ at the low--luminosity end of the RLF of the high--luminosity population, a quantity which has been directly measured at $z \\sim 1$ by the 7CRS.  We argue that robust determination of this quantity at higher redshifts requires a new redshift survey based on a large ($\\sim 1000$ source) sample about five times fainter than the 7CRS. ",
        "introduction": "\\label{sec:intro}  The radio luminosity function (RLF) seeks to derive from observed samples and surveys of radio sources, their space density per unit co-moving volume and how this changes with source luminosity. It is an essential input to both gravitational lensing studies which probe cosmic geometry (e.g. Kochanek 1996) and modelling of the clustering of radio sources (Magliocchetti et al. 1998). The RLF derived at low radio frequencies is important for understanding the content of high frequency surveys and hence deriving jet beaming parameters (e.g. Jackson \\& Wall 1999). The shape and evolution of the RLF provide important constraints on the nature of radio activity in massive galaxies and its cosmic evolution.  Longair (1966) was one of the first to attempt to determine the evolution of the radio source population. At that time there were very few radio source redshifts known and the main constraint upon his models came from the low--frequency source counts down to $S_{151}\\approx0.25$ Jy. Longair found that the data were best-fit by models where the most powerful radio sources undergo greater cosmic evolution (in comoving space density $\\rho$) than less-powerful sources. The subsequent acquisition of a substantial fraction of the redshifts for the 3CRR complete sample of Laing, Riley \\& Longair (1983) enabled the evolution to be better constrained. Using the $V/V_{\\rm max}$ test (see Section \\ref{vvmax}) they concluded that the most powerful radio sources undergo substantial evolution with similar evolution for both powerful radio galaxies and quasars. Unfortunately, the tight correlation between radio luminosity and redshift in this sample meant that they were unable to make any progress on the form of the evolving RLF. Wall, Pearson \\& Longair (1980) suggested that complete samples a factor of $\\approx 50$ times fainter than the 3CRR sample would be necessary to differentiate between possible models of the RLF.  The most comprehensive study of the radio luminosity function prior to this work was performed by Dunlop \\& Peacock (1990; hereafter DP90). Using several complete samples selected at 2.7 GHz with lower flux-limits ranging from 2.0 Jy to 0.1 Jy, they considered the flat- and steep-spectrum populations separately and derived the RLF for each population. Their constraints on the RLF of flat-spectrum sources were somewhat tighter than those derived by Peacock (1985), but the major breakthrough of their paper concerned the constraints they were able to place on the RLF of steep-spectrum radio sources. The strong positive evolution in $\\rho$ with $z$ out to $z \\sim 2$ inferred by Longair (1966), Wall et al. (1980) and Laing et al. (1983) was mapped out in some detail, and more controversially their results indicated a decline in the co-moving number density of both populations beyond a redshift of $\\approx 2$ -- the so-called `redshift cut--off'. DP90 concluded that for powerful sources a model of pure luminosity evolution (PLE -- evolution only of the break luminosity with redshift, analogous to that of Boyle, Shanks \\& Peterson 1988 for the quasar optical luminosity function) fitted the data well (but only for $\\Omega=1$). A similar model which incorporates negative density evolution at high--redshift (the LDE model) was found to work well for both $\\Omega=1$ and $\\Omega=0$. A recent update on the DP90 work can be found in Dunlop (1998), and a critical re-evaluation of the evidence for a redshift cut--off in the flat-spectrum population can be found in Jarvis \\& Rawlings (2000).  Derivations of the luminosity functions of luminous AGN selected at all frequencies (radio, optical and X-ray) have shown a broken power-law form with a steeper slope at the high--luminosity end than at the low luminosity end (e.g. DP90, Boyle et al. 1988, Page et al. 1996). These studies were also able to adequately describe the positive evolution of the luminosity function from $z=0$ to $z \\approx 2$ by assuming pure luminosity evolution. More recent studies with different datasets have shown some deviations from PLE (e.g. Goldschmidt \\& Miller 1998, Miyaji, Hasinger \\& Schmidt 2000). The physical rationale behind PLE is that the AGN have long lifetimes (comparable with the Hubble Time) and they decline in luminosity from $z \\approx 2$ to $z=0$. However, at least in the case of the high--luminosity radio population, we know such a model is not viable because the ages of FRII radio sources in the surveys are well-constrained to be $\\ltsimeq$ a few $10^8$ yr (e.g. Blundell, Rawlings \\& Willott 1999). Furthermore, for X-ray and optically-selected quasars, accretion at the rates required for their high luminosities (over a Hubble Time) would lead to much more massive black holes than are observed in the local Universe (e.g. Cavaliere \\& Padovani 1989). Thus, although PLE models do fit the gross features of the data, the physical basis of these models is not clear.  The radio source population is composed of several seemingly different types of object. One division between types is concerned with radio structure which falls into two distinct classes: FRI, or `twin-jet', sources; and FRII, or `classical double', sources (Fanaroff \\& Riley 1974). At low luminosities $ \\log _{10} (L_{151}) \\ltsimeq 25.5$, the vast majority of sources have an FRI radio structure (apart from the very low luminosity starbursts), whilst at higher luminosities the majority of sources have FRII structure. A second division is concerned with the optical properties of the population. At $\\log _{10} (L_{151}) \\sim 26.5$ the fraction of objects in low--frequency selected samples with observed broad-line nuclei changes rapidly from $\\approx 0.4$ at higher radio luminosities to $\\approx 0.1$ at lower luminosities (Willott et al. 2000).  It is not yet clear whether this represents a fundamental change in central engine properties because, for example, it could be due to the opening angle of an obscuring torus increasing with decreasing luminosity (see Willott et al. 2000). However, it is clear that at least some FRII sources lack even indirect evidence for an active quasar nucleus because, as pointed out by Hine \\& Longair (1979) and Laing et al. (1994), they have weak or absent emission lines. Indeed it is plausible that the majority of FRII radio galaxies (per unit comoving volume) show passive elliptical-galaxy spectra (Rixon, Wall \\& Benn 1991) and are part of the low--luminosity population considered here.  In this paper we have chosen to investigate the RLF in terms of a dual-population model where the less radio-luminous population is composed of FRIs and FRIIs with weak/absent emission lines, and the more radio-luminous population of strong-line FRII radio galaxies and quasars. Our motivation for doing this is that the presence/absence of a quasar nucleus, as indicated by emission line strength, is a distinction intimately connected to the properties of the central engine, whereas radio structure is strongly influenced by the larger-scale environment (e.g. Kaiser \\& Alexander 1997).  The critical radio luminosity [$\\log _{10} (L_{151}) \\sim 26.5$] which crudely divides the populations is also very close to the break in the RLF (e.g. from DP90), whereas the luminosity of the FRI/FRII divide is clearly well below the break. Tentative evidence for an increase in the fraction of quasars with redshift from $z=0$ to $z \\approx 1$ can then be explained if this low--luminosity population evolves less rapidly than the high--luminosity population (Willott et al. 2000). The $\\rho$ of FRI sources is already known to evolve less-strongly with $z$ than that of the radio-luminous FRIIs (e.g. Urry \\& Padovani 1995), but here we will assume that this is because they are a sub-set of a larger population of low--luminosity radio sources, including FRIIs, which evolve less strongly.  We will show that differential evolution between the two populations mimics PLE to a certain extent, but with a more plausible physical basis.  Padovani \\& Urry (1992), Urry \\& Padovani (1995) and Jackson \\& Wall (1999) have adopted a different approach to determining the RLF from high--frequency complete samples. Instead of separating the flat- and steep-spectrum sources and deriving the RLF for each population, they use models built around unified schemes and relativistic beaming to determine the relative fractions of various classes of objects in the samples. Their models assume that BL Lac objects are the beamed versions of FRI and (in the case of Jackson \\& Wall) weak-line FRII radio galaxies, and that flat-spectrum quasars are favourably oriented FRII radio galaxies.  However, both these analyses required the RLF of unbeamed objects to be determined independently, i.e from low--frequency samples. Because they used only the 3CRR sample to determine the luminosity distribution of radio sources, the tight $L_{178}-z$ correlation meant a degeneracy between luminosity and redshift effects.  Current models of this type are thus most useful for constraining the values of various beaming parameters and how these affect the population mix in high--frequency selected samples, and have not provided new information on the evolution of steep-spectrum sources.  In this paper we present a new investigation of the steep-spectrum RLF which combines the bright 3CRR sample with the new, fainter 6CE and 7CRS redshift surveys (based on the 6C and 7C radio surveys, respectively). Note that an investigation of the RLF of steep-spectrum quasars (based on the 3CRR and 7CRS samples) has been presented by Willott et al. 1998 (W98). As the basis for an investigation of the steep-spectrum RLF, the 3CRR/6CE/7CRS dataset improves on that used by DP90 in four different ways.  \\begin{itemize}  \\item{ {\\bf Spectroscopic completeness.}  Many of the sources in the DP90 samples do not have spectroscopic redshifts [including, at the time of DP90, $\\approx50\\%$ of the sources in their faintest sample -- the Parkes Selected Regions (PSR)], and redshifts had to be estimated from the infrared Hubble diagram (the $K-z$ relation). Dunlop (1998) reports that further spectroscopy of these sources has shown their redshifts tended to be overestimated (due to the positive correlation between radio luminosity and $K$-band luminosity, Eales et al. 1997), which he contends strengthens the DP90 evidence for a redshift cut--off in high--frequency selected samples. However, as we discuss later (Section 5.1), the assignment of systematically high redshifts to many sources may in fact have caused DP90 to overestimate the strength of any redshift cut--off. In contrast to the PSR, the 6CE and 7CRS have virtually complete spectroscopic redshifts.}   \\item{ {\\bf Coverage of the radio luminosity versus $z$ plane.} The flux-limit of the PSR is $S_{2.7}=0.1$ Jy. For steep-spectrum sources with $\\alpha = 0.8$, this corresponds to $S_{151} \\approx 1$ Jy, and for $\\alpha = 1.2$ to $S_{151} \\approx 3$ Jy, so the 7C Redshift Survey (7CRS) with a flux-limit of $S_{151}=0.5$ Jy is a factor $2-6$ times deeper at a given $z$ than the PSR depending on spectral index. Because of the correlations between $\\alpha$ and radio luminosity and $z$ (e.g. Blundell et al. 1999) this makes the 7CRS sensitive to lower radio luminosity objects at the higher redshifts. Note that because the PSR covers an area 3.5 times larger than the 7CRS, the two samples actually contain similar numbers of high--redshift objects.}  \\item{ {\\bf Radio completeness.} DP90 found the source counts in the PSR at 0.1 Jy to be a factor of two lower than expected from the source counts at other frequencies. They suggested that this is due to incompleteness in the original survey near its flux-limit, although this is by no means certain (J.V. Wall, priv. comm.). Although they attempted to correct for this, there remains some possibility of bias: methods based on the $V / V_{\\rm max}$ statistic are particularly sensitive to such problems, as objects close to the flux-limit have $V / V_{\\rm max} \\approx 1$, and losing such objects would bias the statistic towards a spuriously low value.  The 6CE sample and the 7CRS are based on surveys which are complete at the flux-densities of interest. As Riley (1989) discusses, the number of large angular size sources omitted from surveys such as these due to surface brightness effects is likely to be negligible.}  \\item{ {\\bf Benefits of low--frequency selection}.  The ultimate goal of our programme is to model the evolution in the RLF of steep-spectrum sources. As detailed in Blundell et al. (1999), the physics of FRII radio sources means that the mapping between radio luminosity and one of the key physical variables, the bulk power in the jets, is much closer if the radio luminosity is evaluated at low rest-frame frequencies provided, of course, these lie above the synchrotron self-absorption frequency of a given source. Modelling of high frequency luminosity is much more problematic, both because lobe emission at these frequencies is highly sensitive to source age and redshift (because of inverse-Compton scattering of lobe electrons by the microwave background), and because Doppler-boosted cores and hotspots become much more important at high rest-frame frequencies. }  \\end{itemize}  In Section 2 we discuss the data available to constrain the steep-spectrum RLF. Section 3 describes the model fitting procedures and results. In Section 4 the evolution of the radio source population is tested via the $V/V_{\\rm max}$ statistic. Section 5 compares our findings with those of previous RLF determinations. In Section 6 we summarise our results and indicate the future work necessary to resolve the outstanding uncertainties.  The convention for the radio spectral index, $\\alpha_{\\rm rad}$, is that $S_{\\nu} \\propto \\nu^{-\\alpha_{\\rm rad}}$, where $S_{\\nu}$ is the flux-density at frequency $\\nu$. We assume throughout that $H_{\\circ}=50~ {\\rm km~s^{-1}Mpc^{-1}}$ and $\\Omega_ {\\Lambda}=0$. All results are presented for both flat ($\\Omega_ {\\rm M}=1$) and open ($\\Omega_ {\\rm M}=0$) cosmologies. ",
        "conclusions": "\\begin{itemize} \\item We have used three low-frequency selected surveys with essentially complete spectroscopic redshift information, giving unprecedented coverage of the radio luminosity versus $z$ plane, to derive the most accurate measurement of the steep-spectrum RLF yet made.  \\item We find that a simple dual-population model for the RLF fits the data well, requiring differential density evolution (with $z$) for the two populations.  \\item The low--luminosity population can be associated with radio galaxies with weak emission lines, and includes sources with both FRI and FRII radio structures; its comoving space density $\\rho$ rises by about one dex between $z \\sim 0$ and $z \\sim 1$ but cannot yet be meaningfully constrained at higher redshifts.  The high--luminosity population can be associated with radio galaxies and quasars with strong emission lines, and consists almost exclusively of sources with FRII radio structure; its $\\rho$ rises by nearly three dex between $z \\sim 0$ and $z \\sim 2$.  \\end{itemize}"
    },
    "0010/astro-ph0010133_arXiv.txt": {
        "abstract": "}[2]{{\\footnotesize\\begin{center}ABSTRACT\\end{center} \\vspace{1mm}\\par#1\\par \\noindent {~}{\\it #2}}}  \\newcommand{\\TabCap}[2]{\\begin{center}\\parbox[t]{#1}{\\begin{center} \\small {\\spaceskip 2pt plus 1pt minus 1pt T a b l e} \\refstepcounter{table}\\thetable \\\\[2mm] \\footnotesize #2 \\end{center}}\\end{center}}  \\newcommand{\\TableSep}[2]{\\begin{table}[p]\\vspace{#1} \\TabCap{#2}\\end{table}}  \\newcommand{\\FigCap}[1]{\\footnotesize\\par\\noindent Fig.\\  % \\refstepcounter{figure}\\thefigure. #1\\par}  \\newcommand{\\TableFont}{\\footnotesize} \\newcommand{\\TableFontIt}{\\ttit} \\newcommand{\\SetTableFont}[1]{\\renewcommand{\\TableFont}{#1}}  \\newcommand{\\MakeTable}[4]{\\begin{table}[htb]\\TabCap{#2}{#3} \\begin{center} \\TableFont \\begin{tabular}{#1} #4 \\end{tabular}\\end{center}\\end{table}}  \\newcommand{\\MakeTableSep}[4]{\\begin{table}[p]\\TabCap{#2}{#3} \\begin{center} \\TableFont \\begin{tabular}{#1} #4 \\end{tabular}\\end{center}\\end{table}}  \\newenvironment{references}% { \\footnotesize \\frenchspacing \\renewcommand{\\thesection}{} \\renewcommand{\\in}{{\\rm in }} \\renewcommand{\\AA}{Astron.\\ Astrophys.} \\newcommand{\\AAS}{Astron.~Astrophys.~Suppl.~Ser.} \\newcommand{\\ApJ}{Astrophys.\\ J.} \\newcommand{\\ApJS}{Astrophys.\\ J.~Suppl.~Ser.} \\newcommand{\\ApJL}{Astrophys.\\ J.~Letters} \\newcommand{\\AJ}{Astron.\\ J.} \\newcommand{\\IBVS}{IBVS} \\newcommand{\\PASP}{P.A.S.P.} \\newcommand{\\Acta}{Acta Astron.} \\newcommand{\\MNRAS}{MNRAS} \\renewcommand{\\and}{{\\rm and }} ",
        "introduction": " ",
        "conclusions": ""
    },
    "0010/astro-ph0010243_arXiv.txt": {
        "abstract": "We report the detection of a new 3.06\\,ms binary pulsar in the globular cluster NGC6544 using a Fourier-domain ``acceleration'' search.  With an implied companion mass of $\\sim0.01\\,\\sm$ and an orbital period of only $P_b\\sim1.7$\\,hours, it displays very similar orbital properties to many pulsars which are eclipsed by their companion winds.  The orbital period is the second shortest of known binary pulsars after 47~Tuc~R.  The measured flux density of $1.3\\pm0.4$\\,mJy at 1332\\,MHz indicates that the pulsar is almost certainly the known steep-spectrum point source near the core of NGC6544. ",
        "introduction": "\\label{sec:intro} Globular clusters are rich sources of radio pulsars.  Since the discovery of the first pulsar in a cluster by \\citet{lbm+87}, a series of deep searches have accumulated almost 50 pulsars in globular clusters, with the majority of these being binary millisecond pulsars (MSPs).  See \\citet{lyn95} and \\citet{ka96} for reviews.  Fruchter \\& Goss (1990; 2000, hereafter FG00) \\nocite{fg90, fg00} used the Very Large Array (VLA) to detect steep-spectrum radio sources, which are likely pulsars, in numerous globular clusters.  They discovered a relatively bright point-like source within $5''$ of the optical center of cluster NGC6544 ($1.2\\pm0.1$\\,mJy at 20\\,cm and $6.5\\pm0.8$\\,mJy at 90\\,cm).  \\citet{bl96} searched NGC6544 at 50\\,cm as part of a survey of globular clusters and found no pulsars down to a limiting flux density of $\\sim4$\\,mJy.  It has been argued that Doppler smearing from orbital motion could explain the lack of detection of pulsations from the pulsar-like sources found in VLA surveys (FG00).  The sensitivities of observations whose durations exceed even a few percent of the orbital period may be drastically reduced for traditional analyses \\citep{jk91}.  Various schemes have been developed to correct, at least partially, for this loss of sensitivity when the Doppler effect of the orbital motion can be approximated as a constant frequency derivative \\citep[e.g.][]{mk84, agk+90, whn+91}.  These ``acceleration'' searches have met with significant success, including the discovery of nine new binary MSPs in the globular cluster 47~Tucanae \\citep{clf+00}.  We have discovered a binary MSP in NGC6544 using Fourier-domain techniques to correct for pulsar orbital motion.  This pulsar is almost certainly the radio source reported by FG00 near the center of the cluster.  Results from our single observation indicate that this object shares many of the properties displayed by the burgeoning class of eclipsing binary MSPs.  These systems are characterised by short orbital periods ($P_b \\sim 1-10$\\,hours) and very low-mass companions ($m_2 \\sin i \\simlt 0.05\\,\\sm$)\\ \\citep{rpr00, nzt00}.  \\citet{dlm+00} independently discovered this pulsar and report the results of follow-up observations. ",
        "conclusions": "\\label{sec:discuss}  There are currently eight binary pulsar systems that display similar orbital characteristics to those of \\psr: PSR~B1957$+$20 \\citep{fst88} and PSR~J2051$-$0827 \\citep{sbl+96} in the Galactic disk, PSR~J1910$+$0004 in NGC6760 \\citep{dma+93}, and pulsars 47~Tuc I, J, O, P, and R \\citep{clf+00}.  Each of these systems displays a pulse period of a few milliseconds, an orbital period of 1-10\\,hours, and a very low-mass companion ($m_2 \\sin i \\simlt 0.05\\,\\sm$).  Some of these systems display radio eclipses, particularly at longer wavelengths (i.e. 50-90\\,cm).  Our current 1332-MHz data show no evidence for an eclipse (see Figure~4) at or near superior conjunction, but we cannot rule out the possiblity of eclipses at longer wavelengths.  Pulsars J2051$-$0827 and 47~Tuc~J show eclipses at radio wavelengths between 50-90\\,cm, but are detected at all orbital phases at 20\\,cm, scintillation permitting \\citep{sbl+96, clf+00}.  PSR~J1910$+$0004 in NGC6760 was detected at all orbital phases at 20\\,cm, but \\citet{dma+93} could not rule out short duration eclipses or eclipses at longer wavelengths.  Most formation theories for short-period binary MSPs in globular clusters involve dynamical interactions with primordial binaries which have been shown to exist in significant numbers \\citep{hmg+92}. \\citet{rpr00} describe a recent variant of the standard ``spin-up'' model \\citep[see][for a review]{pk94} using an exchange interaction between a neutron star (NS) and a hard primordial binary containing at least one relatively massive main sequence (MS) star ($\\sim$ 1-3\\,\\sm).  If the exchange interaction results in a binary containing the NS and the massive MS star, the system enters a common-envelope (CE) phase once the MS star evolves and fills its Roche lobe.  At the end of the CE phase, the NS emerges with a very low-mass companion in a very short period circular orbit.  Systems with orbital periods $P_b\\simlt8$\\,hours undergo evolution due to gravitational radiation and enter a second phase of mass transfer that spins the NS up to millisecond periods.  The scenario described by \\citet{rpr00} is possible only if the exchange interactions occurred at a time when massive MS stars were still present in the cluster.  The NSs and primordial binaries must also have undergone mass segregation and ended up near the cluster core before the MS stars evolved.  The timescale for mass segregation is the half-mass relaxation time, which for NGC6544 is $\\sim200$\\,Myr \\citep{har96}.  This is significantly less than the MS lifetime of a 1-3\\,\\sm\\ star and implies that this scenario may have produced \\psr.  Alternatives to ``spin-up'' models exist in which the NS is created either by accretion induced collapse of a white dwarf (WD) \\citep[e.g.][]{bg90}, or the coalescence of a pair of massive WDs \\citep{cl93}.  Pairs of primordial binaries interact to produce hard binaries which contain a massive WD or WDs that these models require. The NSs that result from these models are born spinning rapidly ($P_{spin} \\sim 2-10$\\,ms), have relatively weak ($10^{9}-10^{10}$\\,G) magnetic fields characteristic of MSPs, and may have low-mass companions in very short period orbits.  Interestingly, these systems could be the progenitors of the low-mass X-ray binaries (LMXBs) rather than their descendants \\citep{cmr93}.  Future observations of \\psr\\ will lead to improved measurements of rotational and orbital parameters, an estimate of or upper limit to the orbital eccentricity, and perhaps information about the circumstellar medium if eclipses are detected.  \\citet{rpr00} predict that many similar systems must exist in the globular cluster system --- some with orbital periods as short as $\\sim15$\\,min.  If recent successes such as the 20\\,cm searches of 47~Tucanae \\citep{clf+00} are any indication, this prediction will soon be put to the test.  The ever-increasing speed of computers and improved algorithms for binary pulsar detection \\citep[e.g.][]{ran00} will allow the analysis of extremely long observations (i.e. days or weeks) that should reveal even the weakest of pulsars in the tightest of binaries."
    },
    "0010/astro-ph0010075_arXiv.txt": {
        "abstract": "We have used the \\nan\\ Radio Telescope to obtain new global \\HI\\ data for 16 giant low surface brightness (LSB) galaxies. Our targets have optical luminosities and disk scale lengths at the high end for spiral galaxies ($L_{B}\\sim 10^{10}$~\\Lsun\\ and $h_{r}\\ga$6~kpc for $H_{0}$=75~\\kms~Mpc$^{-1}$), but they have diffuse stellar disks, with mean disk surface brightnesses $\\ga$1 magnitude fainter than normal giant spirals. Thirteen of the galaxies previously had  been detected in \\HI\\ by other workers, but the published \\HI\\ observations were either confused, resolved by the telescope beam, of low signal-to-noise, or showed significant discrepancies between different authors. For the other 3 galaxies, no \\HI\\ data were previously available. Several of the galaxies were resolved by the \\nan\\ \\am{3}{6} E-W beam, so global parameters were derived from multiple-point mapping observations. Typical \\HI\\ masses for our sample are $\\ga 10^{10}$~\\msun, with $M_{HI}/L_{B}=$0.3-1.7 (in solar units). All of the observed galaxies have published optical surface photometry, and we have compiled key optical measurements for these objects from the literature. We frequently find significant variations among physical parameters of giant LSB galaxies reported by various workers. ",
        "introduction": "\\subsection{The discovery of giant low surface brightness galaxies} In 1987, Bothun et al. reported the serendipitous discovery of the extraordinarily large low surface brightness (LSB) galaxy now known as Malin-1. In spite of having a projected $B$-band central surface brightness of only 26.5~mag arcsec$^{-2}$,  Malin-1 is the largest spiral galaxy known, with a disk scale length of 73~kpc (assuming $H_0$=75 \\kms\\ Mpc$^{-1}$), and an exceptionally high \\HI\\ mass ($\\sim 10^{11}$\\Msun) and optical luminosity ($M_{B}=-$23.1). Subsequently, through systematic searches of photographic survey plates, other galaxies with similar (albeit slightly less extreme) properties to Malin-1 have been uncovered (Bothun et al. 1990; Sprayberry et al. 1993; Sprayberry et al. 1995b). We hereafter refer to these as ``LSB Giants''. A handful of LSB Giants are also found in the UGC (Nilson 1973), NGC (Dreyer 1953), and ESO (Lauberts \\& Valentijn 1989) catalogues (see Gallagher \\& Bushouse 1983; Impey \\& Bothun 1989; Walsh et al. 1997; Pickering et al. 1997; Schombert 1998). Nonetheless, while  recent photographic and CCD surveys have uncovered large numbers of new small and medium-sized, moderate-to-low surface brightness  spiral galaxies (e.g., Schombert et al. 1992; Impey et al. 1996; O'Neil et al. 1997), LSB Giants have remained relatively rare. Having faint, diffuse disks, but sizes, \\HI\\ masses, and luminosities at the high end for disk galaxies, the LSB Giants occupy a unique realm of physical parameters space and may share evolutionary histories distinct from other LSB galaxies (e.g., Hoffman et al. 1992).  Since  a continuum of values exists for galaxy properties such as surface brightness, luminosity, and scale length, Sprayberry et al. (1995b) proposed to define LSB Giants as those objects meeting a ``diffuseness index'' criterion: $\\mu_{B}(0)+5{\\rm log}(h_{r})>27.6$\\footnote{Sprayberry et al. (1995b) formulated their diffuseness index based on a distance scale with $H_{0}$=100~\\kms~Mpc$^{-1}$. Here we have rewritten the diffuseness criterion for $H_{0}$=75~\\kms~Mpc$^{-1}$, which is assumed throughout this work.}, where $\\mu_{B}(0)$ is the extrapolated, deprojected $B$-band disk central surface brightness in magnitudes arcsec$^{-2}$, and $h_{r}$ is the disk scale length in kpc. Among seven LSB Giants described by Sprayberry et al. (1995b), mean properties include: $<B-V>=0.73\\pm0.05$, $<\\mu_{B}(0)>=23.23\\pm0.19$ mags arcsec$^{-2}$, and $<h_{r}>=13.0$~kpc. The colors of these LSB Giants are thus comparable to those of normal spirals (Sprayberry et al. 1995b), but are redder than typical colors of many small and moderate-sized LSB disks (e.g., McGaugh \\& Bothun 1994; Matthews \\& Gallagher 1997; de Blok et al. 1996; Beijersbergen et al. 1999). In addition, the LSB Giants are distinct from other more common LSB spirals in that they often have a significant bulge component (e.g., Gallagher \\& Bushouse 1983; Knezek 1993,1998), and frequently their centers harbor an active nucleus  (e.g., Schombert 1998).  The origin and evolutionary histories of LSB Giant galaxies are still enigmatic. Hoffman et al. (1992) have proposed a formation scenario whereby these systems form in very low density regions from rare, 3$\\sigma$ density fluctuations. They predict these galaxies should exhibit quiescent, unevolved, gas-rich disks, with rotation curves that flatten near $V_{max}\\sim$300~\\kms. Knezek (1993) has suggested an alternative scenario, based on Kormendy (1989), whereby LSB Giants may have dissipatively formed from massive, metal-poor dark matter halos.   \\subsection{The need for new \\HIit\\ observations} Testing formation and evolution scenarios for LSB Giants requires an accurate knowledge of the  neutral gas properties and linewidths of these galaxies. And only by combining such measures with optical data can we begin to build a picture of the star-formation histories of these systems and their relationship to other types of LSB galaxies.   Other motivations also exist for improved \\HI\\ observations. Hoffman et al. (1992) have argued that for the enormous disks of LSB Giants to remain quiescent over a Hubble time, they must be very isolated. Yet studies hint that LSB Giants are in fact less isolated than other LSB spirals, although redshift surveys in the vicinities of these objects are still incomplete (Sprayberry et al. 1995b). Pointed \\HI\\ observations in the vicinity of LSB Giants can thus reveal if these galaxies have any yet-undiscovered gas-rich neighbors.  Another important use of \\HI\\ data is for exploring the Tully-Fisher (TF) relation for giant LSB spirals. Sprayberry et al. (1995a) have shown that at least two of the presently known LSB Giant galaxies are extreme outliers from the TF relation defined by normal galaxies. This is unlike the bulk of moderate-sized, moderate luminosity LSB galaxies, which tend to follow TF (Sprayberry et al. 1995a; Zwaan et al. 1995; Verheijen 1997). It is of considerable interest therefore to assess from a larger sample whether LSB Giants deviate systematically from the TF relation.  While previous \\HI\\ observations have established that LSB Giants are in general very gas-rich (\\MHI$\\ga10^{10}$\\msun; e.g., Sprayberry et al. 1995b; Walsh et al. 1997; Pickering et al. 1997,1999), unfortunately existing \\HI\\ data for many LSB Giants are of dubious quality (i.e., the galaxy was confused or resolved by the telescope beam, the spectra are of low signal-to-noise, or measurements from different workers are highly discrepant; see also Table~5 and Section~\\ref{disc}).  For example, based on Arecibo 21-cm observations of 3 objects, Sprayberry et al. (1995b)  suggested that peculiar, asymmetric \\HI\\ profiles may be commonplace for LSB Giants. However, since the \\HI\\ extents of these galaxies were expected to be comparable to the size of the telescope beam, it is important to verify that these ``peculiar'' spectra do not result from some combination of source resolution and telescope mispointing. In addition, independent checks on derived \\HI\\ parameters are valuable since it is more difficult to accurately measure integrated fluxes and linewidths when the global \\HI\\ profiles are quite broad compared to the bandwidth used, and when sources are at large recessional velocities ($V_{r}\\ga$ 15\\,000~\\kms), where flux calibration can become increasingly uncertain. Finally, there are still a handful of known LSB Giants for which no \\HI\\ data have previously been obtained.  Quality single-dish \\HI\\ spectra for LSB Giants are also a useful precursor and complement to \\HI\\ aperture synthesis studies of these galaxies (e.g., Walsh et al. 1997; Pickering et al. 1997,1999). \\HI\\ mapping is of course a critical part of understanding the dynamics and gas distributions for these galaxies, but since many of the known examples of LSB Giants are rather distant ($V_{r}>$10\\,000~\\kms) and of relatively modest optical angular size ($D_{25}<2'$), such observations are challenging and benefit from careful planning based on prior \\HI\\ measurements. Moreover, because LSB Giants are generally expected to have  disks with relatively low \\HI\\ surface densities (e.g., Pickering et al. 1997), diffuse emission can be missed in aperture synthesis measurements, and total flux  and maximum rotational velocity measures from \\HI\\ pencil beam observations serve as an important check.  Based on the above motivations, we have used the \\nan\\ Radio Telescope to obtain new global observations of a sample of 16 LSB Giant galaxies for which existing global \\HI\\ measurements were incomplete, required reconfirmation, or were nonexistent. We measure integrated \\HI\\ line fluxes, linewidths, and recessional velocities, and  attempt to clear up conundrums surrounding several of these objects in the literature. ",
        "conclusions": "} \\begin{table*} \\caption{Comparison of published \\HI\\ line data with new measurements} {\\scriptsize \\begin{tabular}{lrrrrrl|lrrrrrl} \\hline \\vspace{-2 mm} \\\\ Ident. & $V_{HI}$ & $W_{50,c}$ & $W_{20,c}$ & $S_{HI}$ & Tel. & Ref. & Ident. & $V_{HI}$ & $W_{50,c}$ & $W_{20,c}$ & $S_{HI}$ & Tel. & Ref.\\\\ & km/s  & km/s & km/s & Jy km/s & & &   & km/s  & km/s & km/s & Jy km/s & & \\\\ \\vspace{-2 mm} \\\\ \\hline \\vspace{-2 mm} \\\\ UGC 1752   & 17836 & 384 &     &  3.93 & A & GH89 & Malin-1   & 24750 &     & 341 &  3.5  & A & Bo87 \\\\ & 17861 & 364 & 398 &  4.97 & N & New  &           & 24745 & 315 & 340 &  4.6  & B & IBpc \\\\ &       &     &     &       &   &      &           & 24705 & 295 & 355 &  2.5  & A & IBpc \\\\ UGC 2936   &  3809 &     & 490 &  8.36 & A & He83 &           & 24755 & & 322     &  2.5  & V & P97  \\\\ &  3823 &     & 490 &  8.4  & A & HG84 &           & 24784 & 269 & 308 &  1.68 & N & New \\\\ &  3816 &     & 513 & 11.7  & G & TC88 &           &       &     &     &       &   &      \\\\ &  3817 & 493 & 499 &  9.0  & A & S95  & PGC 45080 & 12264 &     &     &       & A & S95 \\\\ &  3811 & 484 &     & 13.29 & A & H99  &           & 12339 & 267 & 296 & 1.53  & N & New \\\\ &  3813 &     & 500 & 13.6  & V &  P99 &           &       &     &     &       &   &      \\\\ &  3811 & 461 & 488 & 11.20 & N &  New & UGC 9024  &  2285 &     &     &       &   & Bo85b  \\\\ &       &     &     &       &   &      &           &  2326 & 114 & 128 &  8.7  & A & L87   \\\\ UGC 3140   &  4650 &     &     &       & G & S78  &           &  2323 & 117 &     &  9.3  & A & H97    \\\\ &  4630 & 161 & 228 & 12.08 & N & Bt82 &           &  2323 & 115 & 131 &  8.4  & N & T98   \\\\ &  4622 & 125 & 148 & 12.30 & J & SD88 &           &  2321 & 117 & 126 & 11.38 & N & New   \\\\ &  4623 & 118 & 143 & 13.35 & N & New  &           &       &     &     &       &   &      \\\\ &       &     &     &       &   &      & F530-1    & 14339 & 488 &     &  1.7  & A & S92   \\\\ NGC 2770   &  1953 & 330 & 347 & 33.26 & G & FT81 &           & 14335 & 460 & 467 & 2.01  & N &  New \\\\ &  1950 & 334 & 357 & 37.84 & J & SD88 &           &       &     &     &       &   &     \\\\ &  1944 & 327 &     & 24.10 & A & C93  & F533-3    & 12660 & 406 & 430 &  2.5  & A & GH89  \\\\ &  1948 & 340 & 373 & 33.88 & W & RA96 &           & 12659 & 398 &     &  2.2  & A & S92    \\\\ &       & 324 &     & 35.32 & A & H97  &           & 12670 & 367 & 375 &  2.41 & N & New   \\\\ &  1947 & 319 & 343 & 36.90 & N & New  &          &       &     &     &       &   &     \\\\ &       &     &     &       &   &      & NGC 7589  &  8938 & 356 & 363 &       & A & S95  \\\\ F568-6     & 13827 & 386 &     &  3.48 & A & S92  &           &  8925 &     & 402 &  2.7  & V & P97   \\\\ & 13835 &     & 392 &  4.4  & V & P97  &           &  8866 & 345: & 395: & 2.90 & N & New   \\\\ & 13829 & 374 & 394 &  4.10 & N & New  &           &       &     &     &       &   &     \\\\ &       &     &     &       &   &      & PGC 71626 & 10022 &  91 & 118 &  2.6  & N & T98  \\\\ UGC 6614   &  6351 & 244 & 278 &  7.08 & A & Bo85a &          & 10016 &  83 & 115 & 3.86  & N & New   \\\\ &  6358 & 236 & 286 &  9.56 & A & HR89  &          &       &     &     &       &   &     \\\\ &  6353 &     & 287 & 15.0  & V & P97  &           &       &     &     &       &   &      \\\\ &  6351 & 248 & 277 & 15.79 & N & New  &           &       &     &     &       &   &      \\\\ \\vspace{-2 mm} \\\\ \\hline \\end{tabular} \\begin{tabular}{llllll} \\vspace{-2 mm} \\\\ \\multicolumn{6}{l}{References:} \\\\ \\vspace{-2 mm} \\\\ Bo85a & Bothun  et al. (1985a)          & Bo85b & Bothun et al. (1985b)           & Bo87  & Bothun et al. (1987)            \\\\ Bt82  & Bottinelli  et al. (1982)       & C93   & Chengalur et al. (1993)         & FT81  & Fisher \\& Tully (1981)          \\\\ GH89  & Giovanelli \\& Haynes (1989)     & HG84  & Haynes \\& Giovanelli (1984)     & H97   & Haynes  et al. (1997)           \\\\ H99   & Haynes  et al. (1999)           & He83  & Hewitt et al. (1983)            & Hu83  & Huchra et al. (1983)            \\\\ HR89  & Huchtmeier \\& Richter (1989)      & IBpc  & Impey \\& Bothun (1989), see HR89  & L87   & Lewis  (1987)                    \\\\ New   & This work                       & P97   & Pickering et al. (1997)         & P99   & Pickering et al. (1999)         \\\\ RA96  & Rhee \\& van Albada (1996)       & S92   & Schombert et al. (1992)         & S78   & Shostak      (1978)             \\\\ S95   & Sprayberry et al. (1995b)       & SD88  & Staveley-Smith \\& Davies (1988) & T98   & Theureau et al. (1998)          \\\\ TC88  & Tifft \\& Cocke (1988)           & &                                 \\\\ \\vspace{-2 mm} \\\\ \\multicolumn{6}{l}{Telescope codes:} \\\\ \\vspace{-2 mm} \\\\ A\\, & Arecibo 305m       & B \\,& Green Bank 43m     & E\\, & Effelsberg 100m    \\\\ G\\, &  Green Bank 91m    & J\\, & Jodrell Bank 64m   & N\\, & \\nan\\ 94m equiv    \\\\ V   & VLA                & W\\, & Westerbork         & &                    \\\\ \\vspace{-2 mm} \\\\ \\hline \\end{tabular} } \\label{hilit} \\normalsize \\end{table*}  \\subsection{The \\HIit\\ properties of LSB Giants} Not surprisingly, all of the LSB Giant galaxies observed in the present program are found to be very \\HI-rich systems, having \\mhi\\ values ranging from $2.5\\times 10^{9}$~\\msun--7.1$\\times 10^{10}$~\\msun (Table~4). The bulk of the sample galaxies are not only \\HI\\ rich in an absolute sense, but are also \\HI-rich both for their Hubble types and for their optical luminosities---six of the galaxies in the present LSB Giant sample have $M_{HI}/L_{B}\\ga$1 (in solar units; see Table~4). More typical values of $M_{HI}/L_{B}$ for high surface brightness spirals are: $\\sim$0.2 (type Sab-Sb); $\\sim$0.3 (type Sbc-Sc); $\\sim$0.4 (type Scd-Sd) (Roberts \\& Haynes 1994).  As expected for massive, \\HI-rich disks, broad \\HI\\ linewidths are also hallmarks of LSB Giants. Because a number of our program galaxies have rather low inclinations ($i\\le 40^{\\circ}$), it is difficult to estimate the true rotational velocities in some cases, but even many of the uncorrected linewidths tend to be quite broad (e.g., $W_{20,c}=$398~\\kms\\ for UGC~1752, which is seen nearly face-on). Only PGC~71626 seems to have a surprisingly narrow linewidth ($W_{20,c}$=115~\\kms) given its inclination ($i=61^{\\circ}$) and Hubble type (SBb or SBc).  The majority of the LSB Giants in our study exhibit relatively normal, classic, double-horned rotational profiles (Fig.~1). However, we do see a few unusual cases as well;  UGC~568 shows a peculiar, strongly asymmetric profile; PGC~45080 and NGC~7589 both have  rather squarish global profiles with no clear rotation horns; and finally, F530-1 has a double-horned profile that is strongly ``cleft'' in the middle. PGC~135754 may also have a similar cleft morphology, but our detection of this source is relatively weak. Among the 11 relatively normal, double-horned profiles, 3 cases show modest asymmetries (UGC~2936, NGC~2770, and F568-6). Thus we find no evidence that strong \\HI\\ asymmetries are a general feature of LSB Giants. We comment further on our individual spectra below.  \\subsection{Comparison of new \\HIit\\ data with previous measurements and comments on individual objects}  In order to compare the \\HI\\ parameters derived in the present work with past measurements, we have compiled global \\HI\\ parameters from the literature for the 13 of our sample galaxies which have previously been observed in the 21-cm line. Table~5 summarizes previously measured recessional velocities, linewidths, and integrated line fluxes for each case, and indicates the telescope at which the data were obtained, as well as the reference for the quoted parameters.  An examination of Table~5 shows that for LSB Giants observed by more than one group, significant discrepancies frequently exist between the measured parameters, particularly the global integrated line fluxes. We discuss each of the 16 targets observed in the present work in more detail below, including comments on these discrepancies. As part of our investigation of  our LSB Giant targets, we also searched the vicinity of  each of the 16 objects (i.e., an area about 1.5 times the \\nan\\ HPBW) for nearby galaxies which could possibly give rise to confusion in our observed \\HI\\ profiles. For this we employed the NED and LEDA databases, as well as optical images extracted from the Digitized Sky Survey (DSS). Results of these searches are also described below.  \\smallskip  \\noindent {\\bf UGC 568}: no \\HI\\ data have been previously published for this object. Our global spectrum appears quite peculiar, having a strong asymmetry, with a broad, single ``horn'' on the low velocity side of the profile. Published optical redshifts show a discrepancy of over 400~\\kms ($V_{opt}$=13\\,600~\\kms, Melnick \\& Quintana 1981; $V_{opt}$=13\\,198$\\pm$275~\\kms, Impey et al. 1996). Our new \\HI\\ velocity corresponds to the mean of the optical redshifts. Several other galaxies were found within our search area, but they do not seem candidates for an \\HI\\ confusion that could explain our peculiar global \\HI\\ profile: CGCG~384-029 is a $B_T$=15.0 mag Sbc spiral with an optical velocity 886~\\kms\\ higher than the LSB Giant (Dale et al. 1998; Katgert et al. 1998), while PGC~3280 is an S0 with a recessional velocity 1524~\\kms\\ lower than the LSB; MCGG~+00-03-027 is a 15.5 mag galaxy without known velocity, but it is classified as S0/a (LEDA) or Sa (NED), and is therefore expected to be gas-poor.  \\smallskip  \\noindent {\\bf UGC 1752}: Our new total integrated \\HI\\ flux ($S_{HI}=$5.13~\\jks) is somewhat larger than that measured at Arecibo by Giovanelli \\& Haynes (1989; $S_{HI}=3.93$~\\jks), suggesting this galaxy has extended \\HI\\ emission. \\smallskip  \\noindent {\\bf PGC 135657}: no \\HI\\ data have been previously published for this object. Our \\HI\\ detection has a recessional velocity 407~\\kms\\ larger than the optical velocity of $V_{opt}$=12\\,701$\\pm$275 \\kms\\ reported for PGC~135657 by Impey et al. (1996). In the present survey we detect additional marginal features at the edges of our bandpass, near $V_{HI}=$12\\,100~\\kms\\ and $V_{HI}=$13\\,300~\\kms. There are a number of small angular size galaxies in this field, but none appear as obvious candidates for interlopers. \\smallskip  \\noindent {\\bf UGC 2936}: The Arecibo spectrum of UGC~2936 published by Sprayberry (1995b) is highly asymmetric; however only a modest asymmetry is seen in our new data and in the VLA spectrum of Pickering et al. (1999). Our new integrated line flux of 11.23~\\jks\\ agrees to within errors with that of Tifft \\& Cocke (1988) ($\\sim$11.7~\\jks). In comparison, the Arecibo spectra by Hewitt et al. (1983), Haynes \\& Giovanelli (1984), and Sprayberry et al. (1995b) all  yield considerable lower integrated line fluxes ($\\sim$8.4-9.0~\\jks), while our new integrated flux value is somewhat lower than the Arecibo measurement  of Haynes et al. (1999 13.29~\\jks) and the VLA measurement of Pickering et al. (1999; 13.6~\\jks).  CO(1-0) and CO(2-1) line emission were detected from UGC~2936  at SEST and IRAM by Chini et al. (1996), making this one of the rare galaxies classified as an LSB Giant that has been detected in CO (see also Knezek 1993,1998). However, Schmelz et al. (1986) failed to detect OH line absorption from the object. \\smallskip  \\noindent {\\bf UGC 3140 (= NGC 1642)}: This galaxy forms a wide pair (\\am{8}{3} projected separation; $\\Delta V_{opt}$=332$\\pm$85 \\kms) with 15.0 mag Sb spiral UGC~3141. Their E-W separation is \\am{3}{3} and our \\nan\\ spectrum does not appear to be confused. Our new linewidth measurements are in good agreement with the Jodrell Bank measurements of Staveley-Smith \\& Davies (1988) and measurements we performed from new HIPASS data (see Barnes et al., in prep. for a description of HIPASS), but are lower than the values published by Bottinelli et al. (1982). No published \\HI\\ observations are available for UGC~3141. \\smallskip  \\noindent{\\bf NGC 2770}: Short Westerbork radio synthesis 21-cm line observations (Rhee \\& van Albada 1996; Broeils \\& Rhee 1997) indicate that the \\HI\\ distribution of NGC~2770 at a surface density level of 1 \\Msun\\ pc$^{-2}$ is about 1.7 times as large as the optical ($D_{25}$) diameter and that its mean \\HI\\ surface density averaged over the entire \\HI\\ disk $<$$\\sigma_{HI}$$>$ is 4.5 \\Msun\\ pc$^{-2}$, which is higher than typical values for  an Sc. Our new spectra show a slightly higher peak flux density on the low-velocity side of the global line profile, hinting at a possible weak asymmetry not seen in the Westerbork observations. \\smallskip  \\noindent {\\bf PGC 135754}: no \\HI\\ data have been previously published for this object and it was only weakly detected in the present study. Our new \\HI\\ velocity is 434~\\kms\\ higher than the optical value reported by Impey et al. (1996; 21335$\\pm$275~\\kms). \\smallskip   \\noindent {\\bf F568-6 (= Malin-2 = PGC 86622)}: \\HI\\ flux appears to have been missed in the Arecibo observations of F568-6 by Schombert et al. (1992). Our new value of $S_{HI}$=4.20~\\jks\\ shows good agreement with the value derived from the VLA observations of Pickering et al. (1997; 4.4~\\jks).  \\smallskip  \\noindent {\\bf UGC 6614}: Our new integrated \\HI\\ flux of 15.94~\\jks\\ is significantly larger than previous measurements from Arecibo (Bothun et al. 1985a; Haynes \\& Giovanelli 1989). However, Pickering et al. (1997)  measured $S_{HI}$=15.0~\\jks\\ from VLA data, consistent with our results. Our \\HI\\ diameter estimate for this object is very poorly constrained, as we find a Gaussian \\HI\\ model gives a poor fit to the data. This is reaffirmed by the VLA data of Pickering et al. (1997), which show a signficant central depression in the \\HI\\ distribution of this galaxy. \\smallskip   \\noindent {\\bf Malin-1 (= PGC 42102)}: One published Arecibo observation of Malin-1 (see Table~5) gives an integrated \\HI\\ line flux (2.5~\\jks; Bothun et al. 1987) comparable to the VLA value derived by Pickering et al. (1997), while a re-reduction of the same Arecibo data yielded $S_{HI}$=3.5 Jy \\kms\\ (Impey \\& Bothun 1989); finally  Impey \\& Bothun (1989) measure $S_{HI}$=4.6~\\jks\\ from a Green Bank 43m spectrum. Our new data indicate that the Arecibo and Green Bank fluxes previously reported by Impey \\& Bothun (1989) are likely to be significantly overestimated. In contrast our new integrated flux (1.80~\\jks) is somewhat lower than the VLA measurement of Pickering et al. (1997), although our values marginally agree to within errors, since at the redshifted frequency of Malin-1 ($\\sim$1312~MHz), our  absolute calibration uncertainties are expected to be as high as $\\sim\\pm$25\\% (see Section~\\ref{cal}). In terms of spectral morphology, our new global spectrum shows excellent agreement with that derived by Pickering et al.  No CO(1-0) line emission was detected in the central regions of Malin-1 by Radford (1992), with an estimated upper limit of 4.7$\\times 10^9$ \\Msun\\ to the H$_2$ mass, and Braine et al. (2000) recently reported an upper limit of $M_{H2}/M_{HI}<0.03$.  \\smallskip  \\noindent {\\bf PGC 45080}: Our new integrated line flux is smaller than the values reported by Sprayberry et al. (1995b), and our new spectrum lacks the distinct rotation horns seen in the Sprayberry et al. data. Given the small optical size of this object ($D_{25}$=\\am{0}{79}) it seems unlikely significant flux extended outside our beam. There are also no obvious interlopers in this field that would have caused confusion in the Sprayberry et al. spectrum, hence the discrepencies in the two spectra are probably due to the lower signal-to-noise of our present data. \\smallskip  \\noindent {\\bf UGC 9024}: Our new integrated flux value of 11.41~\\jks\\ is higher than past values measured at Arecibo or with single-pointing \\nan\\ measurements (Bothun et al. 1985; Lewis 1987; Haynes et al. 1997; Theureau et al. 1998), suggesting this galaxy has extended \\HI\\ emission. \\smallskip  \\noindent {\\bf F530-1 (= PGC 87136)}: We detect only a slightly higher integrated flux (2.07~\\jks) than the Arecibo value of Schombert et al. (1992; 1.7~\\jks).  Our spectrum shows an unusual ``cleft'' morphology, although our signal-to-noise is only modest due to the broad linewidth of this source. \\smallskip  \\noindent{\\bf F533-3 (= PGC 68495)}: No flux is clearly detected at 2$'$E in our mapping observations of this galaxy; this may be due to the broadness and relative weakness of the line profile and/or an \\HI\\ distribution that is very lopsided relative to the optical center of the galaxy. As a result, our \\HI\\ diameter estimate for F533-3 very uncertain. \\smallskip  \\noindent{\\bf NGC 7589}: NGC~7589 was mapped in \\HI\\ at the VLA by Pickering et al. (1997). Our new \\HI\\ profile hints at a possible small amount of additional flux on the low-velocity side of the profile compared with the VLA global spectrum; this is probably caused by confusion with F893-29, which was also detected in \\HI\\ at the VLA. F893-29 is  a 17.1 mag Sb? spiral (NED) or dwarf (Pickering et al. 1997), \\am{3}{3} SW of the LSB Giant, with a E-W separation of \\am{2}{2}. F893-29 has a systemic \\HI\\ velocity of  8768 \\kms\\ (Pickering et al. 1997), consistent with the velocity of the ``extra'' flux we observe associated with our global profile of NGC~7589. Our new integrated line flux (2.96~\\jks) agrees well with that reported by Pickering et al (1997; 2.7~\\jks), hence the flux contamination from this second source appears to be minimal. However our line profile widths appear to be overestimated by $\\sim$100~\\kms\\ due to this second source, hence corrections have been applied to the $W_{20}$ and $W_{50}$ measurements in Tables~2 \\& 4. \\smallskip  {\\scriptsize \\begin{table*} \\caption{Optical data from the literature} \\begin{tabular}{lcccccl} \\hline \\vspace{-2 mm} \\\\ Ident. & $V_{opt}$ & $B_{tot}$ & $\\mu(0)$ & $h_{r}$ & $M$ & Ref. \\\\ & km/s & mag & mag $''^{-2}$ & kpc & mag & \\\\ \\vspace{-2 mm} \\\\ \\hline \\vspace{-2 mm} \\\\ UGC 568    & 13197 & 14.76 & 23.0 & 17.1 & $-$21.53 & S95 \\\\ & 13198 & 14.9  & 19.9 &      & $-$21.42 & I96 \\\\ UGC 1752   &       & 15.5$^{a}$ & 22.7 & 20.5 & $-$21.38 & K93   \\\\ PGC 135657 & 12701 & 15.47 & 22.6 & 11.1 & $-$20.74 & S95\\\\ & 12701 & 15.3  & 20.9 &      & $-$20.92 & I96 \\\\ UGC 2936   &       & 13.97 &   22.23 & 9.24 & $-$19.58 & S95 \\\\ &  4065 & 15.1  & 22.1 &   & $-$18.42 & I96\\\\ &  3822 &       & 20.3$^{b}$ & 8.4 & $-$21.1$^{b}$ & P99\\\\ UGC 3140   &       & 13.34 & 20.9 & 3.9 & $-$20.61 & dJ96 \\\\ &       & 13.4$^{a}$ & 22.8 & 9.8 & $-$20.55 & K93 \\\\ NGC 2770   &       & 12.94 & 20.8 & 13.0 & $-$19.09 & K85 \\\\ PGC 135754 & 21335 & 16.40 & 22.5 & 11.2 & $-$20.99 & S95\\\\ & 21335 & 16.1  & 21.0 & & $-$21.22 & I96 \\\\ F568-6     &       & 14.57 & 23.2 & 21.0 & $-$21.79 & MB94  \\\\ &       &       & 22.1$^{b}$ & 18.3 & $-$23.6$^{b}$ & P97 \\\\ UGC 6614   &       &       & 22.9$^{b}$ & 13.6 & $-$22.3$^{b}$ & P97 \\\\ &       & 14.43 & 24.3 & 15.8 & $-$20.29 & MB94 \\\\ &       &14.6$^{a}$ & 22.3 & 9.1 & $-$20.04 & K93 \\\\ Malin-1    &       &       & 26.0$^{c}$ & 82.0 & $-$22.9$^{c}$ & P97 \\\\ &       &       & 26.4       & 73.2 &               & S93 \\\\ &       &       & 25.5$^{c}$ & 73.2 & $-$23.12      & IB89 \\\\ PGC 45080  &       & 17.48 & 22.8 & 6.45 & $-$18.65 & S95 \\\\ & 11674 & 17.0 & 22.4 & & $-$19.12 & I96 \\\\ UGC 9024   &       & 14.90 & 24.1 & 4.3 & $-$17.58 & dJ96 \\\\ &       & 14.74 & 24.4 & 7.4 & $-$18.46 & MB94 \\\\ F530-1     &       & 16.3$^{a}$ & 21.8 & 12.2 & $-$20.11 & K93  \\\\ F533-3     &       & 15.7$^{a}$ & 23.5 & 26.5 & $-$20.44 & K93  \\\\ NGC 7589   &       &       &  23.3$^{b}$ & 12.6 & $-$21.9$^{b}$ & P97\\\\ PGC 71626  &  9520 & 14.48 & 22.1 & 12.6 & $-$21.08 & S95\\\\ &  9521 & 14.1 & 20.0 & & $-$21.42 & I96 \\\\ \\vspace{-2 mm} \\\\ \\hline \\vspace{-2 mm} \\\\ \\multicolumn{7}{l}{Notes:} \\\\ \\vspace{-2 mm} \\\\ \\multicolumn{7}{l}{$^{a}$measurement within 26 mag arcsec$^{-2}$ isophote} \\\\ \\multicolumn{7}{l}{$^{b}$measurement in $R$ band} \\\\ \\multicolumn{7}{l}{$^{c}$measurement in $V$ band} \\\\ \\vspace{-2 mm} \\\\ \\multicolumn{7}{l}{References:} \\\\ \\vspace{-2 mm} \\\\ \\multicolumn{3}{l}{dJ96 de Jong (1996)}          & \\multicolumn{4}{l}{I96 Impey et al. (1996)}  \\\\ \\multicolumn{3}{l}{IB89 Impey \\& Bothun (1989)}  & \\multicolumn{4}{l}{K85 Kent (1985)}  \\\\ \\multicolumn{3}{l}{K93 Knezek (1993)}            & \\multicolumn{4}{l}{MB94 McGaugh \\& Bothun (1994)} \\\\ \\multicolumn{3}{l}{S93 Sprayberry et al. (1993)} & \\multicolumn{4}{l}{S95 Sprayberry et al. (1995b)}  \\\\ \\multicolumn{3}{l}{P97 Pickering et al. (1997)}  & \\multicolumn{4}{l}{P99 Pickering et al. (1999)}  \\\\ \\vspace{-2 mm} \\\\ \\hline \\end{tabular} \\label{reldet} \\end{table*} } \\normalsize   \\noindent {\\bf PGC 71626}: PGC~71626 was detected at a recessional velocity consistent with that recently reported by Theureau et al. (1998), but somewhat offset from the published optical velocity of 9520~\\kms\\ reported by Sprayberry et al. (1995b) and Impey et al. (1996). Our new integrated flux value of 3.94~\\jks\\ is higher than the value of 2.6~\\jks\\ reported by Theureau et al. (1998) based on a single-pointing \\nan\\ observation. An additional \\HI\\ source was detected in our spectrum at $V_{HI}=9753\\pm4$~\\kms, with $W_{20}=73$~\\kms\\ and $S_{HI}=1.52\\pm0.48$~\\jks. This may be contamination from the S? galaxy MCG~-01-59-026 (of unknown redshift), at a projected distance of \\am{4}{5} from PGC~71626, or due to an uncatalogued LSB dwarf visible on the DSS.  \\subsubsection{Optical parameters for the LSB Giants} All of the LSB Giant galaxies observed in the present survey have optical surface photometry available in the literature. To offer a more complete picture of the nature of the objects in our sample, we have compiled a summary of several optical and photometric parameters for each galaxy in Table~6.  Radial velocities quoted in Table~6 are heliocentric values derived from optical measurements. $B_{tot}$ values are the total $B$-band magnitudes derived from extrapolated exponential disk fits unless otherwise noted. $\\mu(0)$ is the extrapolated central disk surface brightness in the $B$-band (unless otherwise specified) and is {\\it uncorrected} for inclination and internal extinction. $h_{r}$ is the exponential disk scale length in kpc. $M$ is the absolute magnitude ($B$-band, unless otherwise noted) derived from the total magnitude in column 3, and assuming $D=V_{opt}/H_{0}$ if no distance was given in the original reference; no internal extinction corrections were applied. Corrections to the values in Table~6 for Galactic extinction, $k$-corrections, and cosmological surface brightness dimming were generally applied by the original authors.  The values given in Table~6 show that there is a great deal of scatter in the photometric parameters reported for LSB Giant galaxies. Surface photometry of LSB galaxies is always challenging, but it is further complicated for LSB Giants by the fact that many of these galaxies have significant bulges, leading to ambiguity in the bulge-disk decompositions and  hence vastly different derived scale lengths and central surface brightness values from different workers. Offsets between recessional velocities derived via optical and \\HI\\ observations are also common (cf. Table~5).  We include Table~6 partly as a caution to readers interested in understanding the global properties of Giant LSB galaxies that there still exist large uncertainties in the measured parameters for such objects, and to highlight the need for additional surface photometry (and \\HI\\ measurements) of such galaxies. Such data will be necessary before we can distinguish whether all LSB Giants such as the ones in the present study form a distinct class of object, as suggested by Hoffman et al. (1992), or whether some or most of these objects may form a natural and continuous extension of other low and moderate surface brightness disk galaxies. Well-established photometric parameters will also be key to placing the LSB Giant galaxies on the Tully-Fisher diagram."
    },
    "0010/astro-ph0010239_arXiv.txt": {
        "abstract": "We present high spectral resolution maps of near-infrared molecular hydrogen emission from the bipolar pre-planetary nebulae RAFGL 2688 and RAFGL 618, obtained with the NOAO Phoenix spectrometer. The measured velocity gradients along the polar axes of both nebulae indicate that the highest velocity gas lies closest to the central stars. These results support the suggestion that the polar H$_2$ emission regions of both nebulae contain shocked gas formed as fast ($\\sim50-150$ km s$^{-1}$), collimated, post-asymptotic giant branch (AGB) winds collide with slower-moving ($\\sim10-20$ km s$^{-1}$) material previously ejected while the central stars were still on the AGB. The kinematics of H$_2$ emission perpendicular to the polar axis of RAFGL 2688 are consistent with a model combining expansion along the equator at $5-10$ km s$^{-1}$ with rotation about the polar axis at $5-10$ km s$^{-1}$. The rapid onset of the common envelope phase of a close binary system may explain both the bipolar structure of RAFGL 2688 and the presence and complex kinematics of its shocked H$_2$ emission. ",
        "introduction": "The bulk of the mass lost by stars of initial mass $\\sim1-8$ $M_\\odot$ is shed while such stars are on the asymptotic giant branch (AGB).  Evidence is fast accumulating that, during late AGB or post-AGB evolutionary stages, the mass loss geometry of such stars changes from more or less spherically symmetric to axially symmetric, with the result that the vast majority of planetary nebulae (PNs) --- the endpoints of post-main sequence stellar evolution for intermediate-mass stars --- exhibit axisymmetric structures, ranging from elliptical to bipolar (see, e.g., reviews compiled by Kastner, Soker, \\& Rappaport 2000). Planetaries in the latter structural category are very likely to possess molecular envelopes (presumably the remnants of the ejected AGB envelope) that are readily detectable in the near-infrared rovibrational lines of H$_2$ (Kastner et al.\\ 1996). The available data suggest, furthermore, that the onset of near-infrared H$_2$ emission in PNs can be traced back to the pre-planetary nebula (PPN) phase but {\\it not} back to the AGB phase of evolution (Weintraub et al.\\ 1998). These observations suggest that further studies of H$_2$ emission from PPNs and PNs may offer insight into the transition from AGB star to PN and from spherical to axisymmetric mass loss.  The Egg Nebula (RAFGL 2688; Ney et al.\\ 1975) is perhaps the best-known example of such a transition object. Recent {\\it Hubble Space Telescope} (HST) optical and near-infrared images of the Egg reveal a striking juxtaposition of circularly and axially symmetric structures (Sahai et al.\\ 1998a, b). This puzzling combination of structures has been detected in HST images of several other PPNs and PNs (see, e.g., reviews compiled by Kastner, Soker, \\& Rappaport 2000). An outstanding feature of RAFGL 2688, however, is the ``quadrapolar'' morphology of 2.122 $\\mu$m H$_2$ emission detected in near-infrared images (Gatley et al.\\ 1988); this structure is seen most clearly in an HST/NICMOS image (Fig.\\ \\ref{fig:2688NICMOS}; Sahai et al.\\ 1998a). This image shows that the linear extents and bow-shock-like structures of the polar and equatorial plane H$_2$ emission regions are very similar, suggesting a common origin. Furthermore, the sharp, limb-brightened boundaries of the H$_2$ emission in all four ``lobes'' is suggestive of a fairly sudden event in the object's recent past.  The kinematics of molecular emission from RAFGL 2688 have been the subject of numerous investigations.  Bieging \\& Nguyen-Q-Rieu (1988, 1996) and Cox et al.\\ (2000) mapped the velocity distribution of rotational emission of various species with radio interferometers, and Cox et al.\\ (1997) imaged near-IR rovibrational H$_2$ emission at moderate velocity resolution.  Radio interferometric observations of HCN by Bieging \\& Nguyen-Q-Rieu yielded the rather surprising result of velocity gradients both along {\\it and perpendicular to} the polar axis of the system, a result confirmed by Cox et al. These gradients, which are of similar magnitude, were interpreted by Bieging \\& Nguyen-Q-Rieu (1988, 1996) as arising out of a combination of spherical expansion and rotation about the polar axis. Cox et al.\\ (1997, 2000) and Lucas et al.\\ (2000), however, intepret the H$_2$ kinematics of RAFGL 2688 in terms of a system of multiple jets.  The evolved bipolar nebula RAFGL 618 (Westbrook et al.\\ 1975) might be considered the direct evolutionary descendent of RAFGL 2688 in many respects. Both objects display molecule-rich circumstellar envelopes, carbon-rich circumstellar chemistries, and dusty, bipolar reflection nebulosities. Whereas RAFGL 2688 harbors an F-type central star and is not known to contain an H II region, RAFGL 618 possesses a B-type central star and compact H II region, suggesting that the latter nebula has evolved farther toward the PN stage. Like the Egg Nebula, RAFGL 618 displays bright near-infrared H$_2$ emission, but in the case of RAFGL 618 the H$_2$ emission appears to be confined to a region along its polar axis (Latter et al.\\ 1995).  While the ``quadrapolar'' structure of the H$_2$ emission in the Egg appears to be unique, it may be a short-lived evolutionary phase common to many or most AGB stars as they go through the transition to PPNs.  In RAFGL 618, however, the H$_2$ morphology also is unusual as emission is found only along the polar axes, whereas most PPN in which H$_2$ is detected exhibit the strongest H$_2$ along the equatorial plane (Kastner et al. 1996). Thus, the kinematics and morphology of H$_2$ emission from these two objects are strongly deserving of more careful study. To investigate the H$_2$ velocity gradients of RAFGL 2688 in detail, we obtained a series of high-resolution, long-slit spectra in the 2.12 $\\mu$m region with the NOAO\\footnote{National Optical Astronomy Observatories is operated by Associated Universities for Research in Astronomy, Inc., for the National Science Foundation.} Phoenix near-infrared spectrometer (Hinkle et al.\\ 1998). We also obtained a Phoenix spectrum of 2.12 $\\mu$m H$_2$ emission from RAFGL 618. As we now describe, examination and comparison of the H$_2$ kinematics of these two objects, as revealed in the Phoenix spectroscopic data, offer clues to their detailed structures and recent evolutionary histories. ",
        "conclusions": "\\subsection{A kinematical model of H$_2$ emission from RAFGL 2688}  We now construct a detailed, empirical model of the H$_2$ kinematics of RAFGL 2688. In formulating this model, we are guided by our observations that (1) the polar lobes are characterized by velocity gradients in which the fastest moving material is found closest to the star, and the slowest moving material is found at the tips of the H$_2$ emission regions; and (2) the H$_2$ emission seen projected along the equatorial plane contains a substantial velocity gradient east to west and a second, smaller velocity gradient in the E lobe that runs north to south.  \\subsubsection{Polar lobe H$_2$ kinematics}  The Gaussian shapes of the spatially integrated H$_2$ emission line profiles of the polar lobes of RAFGL 2688 (Fig.\\ \\ref{fig:2688spectra}a) are similar to the molecular emission line profiles detected in the radio regime from species such as CO and HCN (e.g., Bieging \\& Nguyen-Q-Rieu 1996).  The H$_2$ line profiles can be understood in the context of the detailed morphology of H$_2$ emission seen in the NICMOS image (Fig.\\ \\ref{fig:2688NICMOS}). In this image the N and S lobes appear as limb-brightened cavities. The emission peaks of the line profiles correspond to the brightest regions of the lobes --- their limbs (i.e., outlines) --- whereas the line wings are produced by emission from within the fainter lobe interiors, which is dominated by material in the near sides (blue wings) and far sides (red wings) of the lobes.  The peak of the line profile therefore indicates the line of sight component of the H$_2$ velocity at the lobe limb, and we can infer the deprojected N and S lobe H$_2$ velocities as a function of position given an estimate of $i$, the inclination of the polar axis of the system out of the plane of the sky. Previous results (Yusef-Zadeh et al.\\ 1984; Sahai et al.\\ 1998a) suggest $i\\sim15^\\circ$. The velocities we measure at the tips of the lobes ($\\sim 5$ km s$^{-1}$; Fig.\\ \\ref{fig:2688poles}) then suggest that the gas in this region is expanding away from the central star at a velocity $\\sim 20$ km s$^{-1}$. We further infer that the H$_2$ velocities in the regions of the lobes within $\\sim1''$ of the inferred position of the central star rise to $\\sim 70$ km s$^{-1}$.  For simplicity, we assume this behavior can be described by an inverse power law of the form \\begin{equation} v(\\Delta\\alpha,\\Delta\\delta) = v_0 (\\frac{r_0}{r})^p \\sin{i} \\end{equation} for (RA, dec) offsets ($\\Delta\\alpha,\\Delta\\delta$) that lie within $45^\\circ$ of the projected polar axis of RAFGL 2688, where $v_0$ is the velocity at the tips of the polar lobes, $v(\\Delta\\alpha,\\Delta\\delta)$ is the observed radial velocity at projected angular radius $r = ((\\Delta\\alpha)^2 + (\\Delta\\delta)^2)^{1/2}$ from the central star, and $i$ is the inclination of the polar axis out of the plane of the sky. Based on the foregoing discussion of line profiles, we make the further simplification that the polar H$_2$ emission is dominated by emission from the lobe limbs.  \\subsubsection{Equatorial plane H$_2$ kinematics}  {\\it If one postulates that all of the east-west H$_2$ emission is confined to a narrow region along the equatorial plane,} then a component of azimuthal velocity must be present in the material responsible for this emission. Such a structure might be analogous to that of the equatorial ``skirt'' observed in $\\eta$ Car (Smith, Gehrz, \\& Krautter 1998). Although the kinematics of the two objects appear to differ in detail (Zethson et al.\\ 1999), the H$_2$ morphology in the NICMOS image (Fig.\\ 1) is consistent with the confinement of the H$_2$-emitting material in RAFGL 2688 to the plane of the system: knot E2 is brighter than knot E4, as expected if the former is directed toward the observer and the latter is directed away from the observer; while E1 and E3, which mark the perpendiculars to the polar axis (see Weintraub et al.\\ 2000, their Fig.\\ 6), are of similar intensity, as expected if we view these structures through similar columns of intervening material.  We therefore assume that the equatorial plane H$_2$ kinematics consist of a combination of azimuthal (rotation) and radial (expansion) velocity components. We calculate these velocities from the relation \\begin{equation} v(\\Delta\\alpha,\\Delta\\delta) = (v_e \\sin{\\phi} + v_r \\cos{\\phi}) \\cos{i} \\end{equation} for (RA, dec) offsets ($\\Delta\\alpha,\\Delta\\delta$) that lie within $45^\\circ$ of the projected equatorial plane of RAFGL 2688, where $v_e$ is the expansion velocity, $v_r$ is the rotational velocity about the polar axis, and $\\phi$ is the azimuthal angle with respect to the polar axis of RAFGL 2688, given by \\begin{equation} \\phi = \\tan^{-1} {(\\frac{\\Delta x}{\\Delta y \\; \\sin{i}})}. \\end{equation} Offsets ($\\Delta x, \\Delta y)$ are then related to ($\\Delta\\alpha,\\Delta\\delta$) via a rotation of the coordinate system by the projected position angle of the polar axis on the plane of the sky.  \\subsubsection{Comparison with data}  Model parameters $v_0$ and $p$ are relatively well constrained by Fig.\\ \\ref{fig:2688poles}, from which we infer $v_0 \\sim 20$ km s$^{-1}$ and $p \\sim 0.7$. It is more difficult to constrain the values of the parameters governing the equatorial plane H$_2$ radial velocities, $v_e$ and $v_r$. To do so, we compared model velocity field images ($v(\\Delta\\alpha,\\Delta\\delta)$) calculated for a range of values of $v_e$ and $v_r$ with velocity centroids calculated from the Phoenix data cube (see, e.g., Bieging \\& Rieu 1996). We find qualitative and rough quantitative agreement between model and data for values of both $v_e$ and $v_r$ in the range $5-10$ km s$^{-1}$, with the additional constraint $v_e + v_r \\sim 15$ km s$^{-1}$. An example of the results for a representative successful model is displayed in Fig.\\ \\ref{fig:2688vcomp}. In this figure the model velocity centroid image is calculated from Eqs.\\ 1 and 2 for values of $v_e = 5$ km s$^{-1}$ and $v_r = 10$ km s$^{-1}$. There is clear qualitative agreement between the model and observed velocity images for these parameter values, in the sense that the overall distribution of redshifted and blueshifted emission is captured by the model. Furthermore, this model reproduces the important details of the observed H$_2$ velocity distribution previously described. Specifically, the model recovers the magnitudes and positions of the observed blueshifted velocity extrema near (RA,dec) offsets $(+1'',+2'')$ in the N lobe and $(+5'',-3'')$ in the E lobe, as well as the magnitudes and positions of redshifted velocity extrema near offsets $(-1'',-2'')$ in the S lobe and at $(-4'',+3'')$ in the W lobe.  Empirically, we find these parameter values --- $v_0=20$ km s$^{-1}$, $p=0.7$, $v_e = 5-10$ km s$^{-1}$, and $v_r = 5-10$ km s$^{-1}$ --- provide the best agreement between model and data in terms of overall qualitative appearance of the velocity images as well as in the location {\\it and} magnitude of the velocity extrema. However, this model cannot be considered a ``best fit'' to the data. Rather, the foregoing comparison of calculated and observed velocity fields offers an indication of the magnitude of an azimuthal velocity component relative to the components of radial expansion both parallel and perpendicular to the polar axis.  \\subsection{Alternative models}  The kinematic model described in \\S 4.1 is not unique. For example, on the basis of high spatial and spectral resolution CO maps, which display kinematics very similar to those seen in H$_2$ emission, Cox et al.\\ (1997, 2000) and Lucas et al.\\ (2000) have proposed that the velocity gradients parallel and perpendicular to the polar axis result from a multipolar system of jets whose directions of motion are everywhere directed {\\it radially outward} from the central star. Since the entire fan-like eastern H$_2$ emission region is blueshifted, the Cox et al.\\ model requires that these outflows must be directed well ``above'' the equatorial plane of the system, toward the observer, while the outflows responsible for the western H$_2$ emission region are more tightly confined and directed ``below'' the equatorial plane, away from the observer.  Hence, in this model, the remarkable orthogonality of the H$_2$ and CO emission morphologies is purely a result of viewing angle. Multiple symmetry axes are also suggested by observations of radio continuum emission (Jura et al.\\ 2000). Chaotic precession of bipolar jets might explain these observations (see, e.g., Livio 2000 and references therein). The very similar linear extents of the orthogonal N-S and E-W molecular emission regions appear to pose a problem for any model invoking precessing jets, however.  Bieging \\& Nguyen-Q-Rieu (1988, 1996) have proposed a model in which the radio molecular line emission from the projected equatorial plane of RAFGL 2688 originates in an expanding, rotating equatorial disk or torus. Though qualitatively similar to the model proposed in \\S 4.1, our model does not require that the H$_2$ emission arises from a disk. Rather, the E-W H$_2$ lobes, like the polar (N-S) H$_2$ lobes (see below), trace the interaction between material ejected along the equatorial plane at two different epochs; indeed, the interferometric radio molecular line observations indicate that the molecular jets detected in CO may be responsible for shocks that produce the H$_2$ emission, based on the striking correspondence between the CO and H$_2$ emission morphologies (Lucas et al.\\ 2000). There is no particular reason to assume that the previously ejected material forms a complete torus around the star and, indeed, Cox et al.\\ (2000) and Morris \\& Sahai (2000) have argued, on the basis of radio and mid-infrared imaging, respectively, that no such structure exists. However, the optical and infrared morphology of the nebula --- in particular, the apparent occultation of the southern H$_2$ lobe and reflection nebulosity within $\\sim2''$ of the central star --- does suggest the presence of a large ($\\sim5''$ radius), optically thick structure.  \\subsection{Origin of the shocked H$_2$ emission}  \\subsubsection{The transition from spherical to axial symmetry}  As noted by Sahai et al.\\ (1998a), the detailed morphology of near-IR H$_2$ emission from RAFGL 2688 strongly supports previous suggestions that the emission arises in shocks formed by the interaction of recently-developed, fast-moving winds with slower-moving material. This relatively slow-moving material likely was ejected during the AGB phase of the progenitor star. Assuming J shocks are the origin of the gas heating that leads to H$_2$ emission then, according to model calculations (e.g., Burton, Hollenbach, \\& Tielens 1992), the shock fronts at the H$_2$ lobe tips are moving at a minimum velocity $v_s \\sim 10$ km s$^{-1}$ in the frame of the preshocked (AGB) wind. This limit corresponds roughly to the energy threshold for excitation of the $v=1-0$ $S(1)$ H$_2$ line. Hence, if the velocities inferred for the tips of the polar lobes and for the equatorial region, $\\sim 20$ km s$^{-1}$ and $\\sim 15$ km s$^{-1}$, respectively, consist of a superposition of (minimum) shock and AGB wind velocities, we infer that the AGB wind was ejected at $v_{\\rm AGB} \\sim 5-10$ km s$^{-1}$, which is typical of {\\it low-mass} carbon stars. On the other hand, the nascent bipolar structure and strong H$_2$ emission of RAFGL 2688 hint at a relatively massive progenitor (Kastner et al.\\ 1996) which may, in turn, suggest a larger AGB outflow velocity ($v_{\\rm AGB} \\sim 20$ km s$^{-1}$; e.g., Kastner et al.\\ 1993). If so, it would appear that, at their leading edges, the shocks which excite H$_2$ have imparted very little momentum to the former AGB wind.  In either case it is apparent that, well behind the leading edges of the shocks that excite the H$_2$ in the polar lobes, considerable wind acceleration has taken place. The maximum speed for non-dissociative shocks is $\\sim25$ km s$^{-1}$ (Burton et al.\\ 1992). Thus, within $\\sim1''$ ($\\sim1000$ AU) of the central star of RAFGL 2688, the outflow velocity of the gas is at least $\\sim45$ km s$^{-1}$, which is much larger than the outflow velocities of AGB stars. In RAFGL 618, meanwhile, the inferred polar outflow velocities very near the central star are at least $\\sim100$ km s$^{-1}$ (neglecting projection effects), assuming the shocks are traveling at $\\sim25$ km s$^{-1}$.  We conclude that the velocity gradients along the polar axes of both RAFGL 2688 and RAFGL 618 trace rapid transitions from the ``slow,'' spherically symmetric winds of their AGB progenitors to faster, collimated, post-AGB winds. A likely mechanism for such a transition is that late AGB and post-AGB ejecta relatively close to the central star are accelerated with increasing efficiency as the star is ``unveiled'' and circumstellar material is exposed to a progressively hotter photosphere and, potentially, a stronger stellar magnetic field.  Indeed, RAFGL 2688 has been observed to be increasing steadily in optical magnitude (Gottlieb \\& Liller 1976). These conditions would appear to be conducive to the formation of H$_2$ shocks during the post-AGB but pre-planetary nebula phase of bipolar nebulae, a scenario consistent with the conclusions of Weintraub et al.\\ (1998). Such shocks could then play a central role in the transition from spherically symmetric to axisymmetric outflow, as the ``young,'' fast (but still predominantly molecular) wind driven by the emerging post-AGB star excavates bipolar cavities within the previously ejected red giant wind. The displacement from the central star to the tips of the H$_2$ lobes ($\\sim6000$ AU), combined with the deprojected velocities at the lobe tips, $v_0 \\sim 20$ km s$^{-1}$, suggests the H$_2$ lobes have a dynamical age of $1500$ yr, which then serves as an upper limit to the post-AGB ``unveiling'' of the central star.  \\subsubsection{The Egg Nebula as endpoint of common envelope binary evolution}  A presently popular theory accounts for the structure of bipolar planetary nebulae as due to the interaction of a close binary system, in which the secondary diverts mass loss from and/or spins up the envelope of the primary (e.g., Soker 1997, 1998). According to a specific formulation of the binary progenitor model (Soker 1998), the relatively massive progenitors of bipolar PNs undergo a phase during the AGB evolution of the primary in which the secondary resides just outside the envelope of the primary and diverts mass loss into the equatorial plane. This phase is followed by a short-lived common envelope phase that, presumably, rapidly spins up the envelope of the primary; the carbon star V Hya, which displays $V \\sin{i} \\sim 13$ km s$^{-1}$, may be undergoing such a phase at present (Barnbaum et al.\\ 1995).  Many, though certainly not all, of the observed characteristics of RAFGL 2688 appear to be explained by such a model. Specifically, its notorious system of concentric broken arcs would result from episodic, spherically symmetric mass loss during a prolonged period of quasi-single star AGB mass loss preceding the common envelope phase; the subsequent formation of bipolar structure would be the result of the increasingly close proximity of a companion during the late AGB evolution of the primary; and the H$_2$ shocks would mark the rapid termination of a common envelope phase and the resultant sudden ejection of the remaining AGB star envelope. Hence, the H$_2$ emission regions are produced by bullet-like ejecta that have collided with material that was recently ejected --- much of it along the equatorial plane --- during the close binary phase preceding common envelope formation.  It would appear that no existing model can explain satisfactorily the rich phenomenology of the Egg Nebula. The molecular kinematics of its equatorial region remain particularly problematic for present models, whether or not a component of azimuthal velocity is present along the equatorial plane of the system. A pure radial outflow model cannot explain the striking orthogonality of the molecular emission from RAFGL 2688; furthermore, such a model is, at present, reliant on {\\it ad hoc} physical mechanisms for the presence of multiple symmetry axes (e.g., Frank 2000). On the other hand, the azimuthal velocities of the equatorial H$_2$ emission suggested by our model and the models of Bieging \\& Nguyen-Q-Rieu (1988, 1996) are far too large to be explained by Keplerian rotation; an alternative source of angular momentum is required.  As proposed by Bieging \\& Nguyen-Q-Rieu, this angular momentum might be provided by magnetic fields that are corotating with the star. It is of considerable importance to establish whether sufficiently large fields can be generated via interaction with a close companion star (Bieging \\& Nguyen-Q-Rieu 1988, 1996) or by dynamo activity in the primary post-AGB star itself (e.g., Soker 2000; Blackman, Frank, \\& Welch 2000). The latter mechanism could be particularly effective if the central star has just undergone a common envelope phase, such that highly magnetically active interior layers of the star drive the molecular jets seen in CO and H$_2$."
    },
    "0010/astro-ph0010525_arXiv.txt": {
        "abstract": "Using high-resolution cosmological N-body simulations, we study how the density profiles of dark matter halos are affected by the filtering of the density power spectrum below a given scale length and by the introduction of a thermal velocity dispersion. In the warm dark matter (WDM) scenario, both the free-streaming scale, \\fsl, and the velocity dispersion, \\vthw, are determined by the mass \\mw\\ of the WDM particle. We found that \\vthw\\ is too small to affect the density profiles of WDM halos. Down to the resolution attained in our simulations ($\\sim0.01$ virial radii), there is not any significant difference in the density profiles and concentrations of halos obtained in simulations with and without the inclusion of \\vthw. Resolved soft cores appear only when we increase artificially the thermal velocity dispersion to a value which is much higher than \\vthw. We show that the size of soft cores in a monolithic collapse is related to the tangential velocity dispersion. The density profiles of the studied halos with masses down to $\\sim0.01$ the filtering mass \\mfs\\ can be described by the Navarro-Frenk-White shape; soft cores are not formed. Nevertheless, the concentrations of these halos are lower than those of the CDM counterparts and are approximately independent of mass. The cosmogony of halos with masses $\\lesssim\\mfs$ is not hierarchical: they form through monolithic collapse and by fragmentation of larger structures. The formation epoch of these halos is slightly later than that of halos with masses $\\approx\\mfs$. The lower concentrations of WDM halos with respect to their CDM counterparts can be accounted for their late formation epoch.  Overall, our results point to a series of advantages of a WDM model over the CDM one. In addition to solving the substructure problem, a WDM model with $\\fsl \\sim 0.16$ Mpc ($\\mw \\approx 0.75\\ \\kev$; flat cosmology with $\\Omega_{\\Lambda}=h=0.7$) also predicts concentrations, a Tully-Fisher relation, and formation epochs for small halos which seems to be in better agreement with observations relative to CDM predictions. ",
        "introduction": "The damping of small-scale modes in the power spectrum of density fluctuations has been proposed to overcome potential observational difficulties at small scales of the hierarchical cold dark matter (CDM) scenario for structure formation (\\eg\\ \\cite{AFH98}; \\cite{Moore99b}; \\cite{SomDol00}; \\cite{Hogan99}; \\cite{KL99}; \\cite{HD2000}; \\cite{WC2000}; \\cite{CAV}). The damping in the power spectrum may be produced either during its primordial generation (in the biased scalar inflationary models, for example) or due to Landau and free-streaming damping, which erase galactic or sub-galactic density fluctuations when the dark matter particles are warm (e.g., \\cite{BPP82}); although the free-streaming damping could also be present in CDM models if CDM particles are non-thermally produced (\\cite{Lin00}). In this paper we study warm dark matter (WDM) models; however, it is important to remark that our results are also valid for other models where the power spectrum is damped at small scales.  A potential shortcoming of the CDM scenario,  is the large amount of substructure (satellites) that is predicted for Milky-Way like halos with respect to observations under the assumption that at each small halo a dwarf galaxy is formed (\\cite{Kauffmann93}; Klypin et al. 1999; \\cite{Moore99a}; but see e.g., \\cite{BKW2000} for an alternative solution). The cosmological N-body experiments of Col\\'{\\i}n et al. (2000, hereafter \\Colinetal) have shown that the observed satellite circular velocity function of Milky Way and Andromeda can be reproduced if the power spectrum is exponentially damped (filtered) at scales $\\sim 0.1-0.2$ Mpc (a flat cosmological model with $\\Omega_\\Lambda = 0.7$ and $h = 0.7$ was used). A free-streaming scale \\fsl\\ of $0.1 - 0.2\\ \\mpc$ is attained for particle masses \\mw\\ of $\\sim 0.6 - 1\\ \\kev$ (\\cite{SomDol00}). Interestingly, \\cite{NSDCh2000} derived a lower limit for \\mw\\ of 750 eV from the restriction that the predicted power spectrum should reproduce the observed properties of the Lyman$-\\alpha$ forest in quasar spectra.  In \\Colinetal\\ it was also found that the concentrations of satellite WDM halos decrease as \\fsl\\ increases, being these concentrations lower than those of the corresponding CDM halos. Unfortunately, these small halos were poorly resolved and we could not explore in detail the inner density profiles of halos with masses below \\mfs. Studying the density profiles of WDM halos with high resolution is necessary, in particular because another potential problem of the CDM scenario is related namely to the inner density profile and concentration of small halos. It seems to be a discrepancy between numerical simulation results and observations of dwarf and low surface brightness (LSB) galaxies (\\eg\\ \\cite{Moore94}; \\cite{FP94}; \\cite{Burkert95}; \\cite{dBM97}; de Blok et al. 2001) although other observational studies have challenged this result (van den Bosch et al. 2000; Swaters, Madore \\& Trewhella 2000; van den Bosch \\& Swaters 2000).  In the present paper the density profiles of WDM halos with masses close to or up to $\\sim 100$ times smaller than \\mfs\\ are studied with a mass resolution $2-3$ orders of magnitude larger than in \\Colinetal. Our conclusions will be based partly on simulations with a power spectrum corresponding to $\\mw\\sim 0.6$ KeV. However, in order to study with high resolution the halos with masses below \\mfs, we need to increase \\mfs. Thus, we will also use simulations for $\\mw\\sim 0.13$ KeV ($\\mfs\\sim 10^{14} \\msunh$) and assume that the structure of halos should depend on the halo mass-to-\\mfs\\ ratio rather than on the specific value of \\mfs. The cosmogony of halos with masses $<\\mfs$ is not hierarchical. Halos with masses close to or larger than \\mfs\\ by factors of $5-10$ assemble a large fraction of their present-day mass by a coherent and nearly monolithic collapse, whereas halos with masses smaller than \\mfs\\ form from the gravitational fragmentation of larger pancake-like structures, as in the top-down scenario (Zel'dovich 1970). Firmani et al. (2000b) emphasized that the density profiles of dark halos formed through a monolithic collapse depend significantly on the velocity dispersion of the collapsing particles (see also Aguilar \\& Merrit 1990); this velocity dispersion may be primordial (thermal relict) or can be acquired during an inhomogeneous gravitational collapse (dynamical).  The non-zero thermal velocities of warm particles (warmons) should produce minimal shallow halo cores because of the phase packing limit (\\eg\\ Gunn \\& Tremaine 1979). However, this thermal velocity dispersion, \\vthw, for warmons with masses of $\\sim 1\\ \\kev$ seems to be too small to produce a noticeable core (Hogan \\& Dalcanton 2000). In \\Colinetal\\ this velocity was neglected, leaving uncertain the effect it could have on the structure of simulated halos (Sellwood 2000). Here we will include different thermal velocity values in the simulations in order to explore their influence on the halo density profile. An analytical (dynamical) approach will be also used to study the ``hot'' monolithic collapse and clarify the conditions required to form shallow cores.  After the completion of this paper, a similar paper by Bode, Ostriker, \\& Turok (2000) appeared in the preprint list of Los Alamos National Laboratory. Bode et al. study the cosmogony and structure of halos for models with damped power spectra in a cosmological box. They also find that halos with masses below the filtering mass are less concentrated and that these halos form later, relative to CDM. In addition to this, they also describe the spatial distribution and the mass function of these small halos; in this sense, their study complement ours.  In section \\S 2 we present the WDM models to be explored in this paper. In \\S 3 we  briefly describe the numerical simulations. In \\S 4 we present the results from our experiments aimed to study the density profiles and concentrations of halos with masses below the mass corresponding to the filtering scale. The influence of the velocity dispersion over the inner density profile of halos formed by monolithic collapse is explored in \\S 5.1. In \\S 5.2 we present results from our WDM cosmological simulations including the thermal velocity dispersion of warmons. In \\S 6.1 we describe and discuss how structure formation proceeds in simulations with a power spectrum suppressed at small scales. The viability of the WDM scenario is disscused in \\S 6.2. Finally, our conclusions are given in \\S 7. ",
        "conclusions": "We have carried out high-resolution N-body cosmological simulations with the aim to study the effect on the halo density profiles produced by (i) the damping of the power spectrum at small scales and (ii) the introduction of a relict thermal velocity dispersion, as well as to explore the viability of the WDM scenario. We have also studied hot monolithic collapse by means of an analytical model and top-hat N-body simulations. Our main conclusions are:  -For the halos studied here, with masses close or up to $\\sim100$ times smaller than the filtering mass \\mfs, the density profiles are on average well described by the NFW shape; the density profiles were well resolved down to $\\sim0.01$ \\rv. The only differences between halos with masses below \\mfs\\ and their CDM counterparts is that the former have lower concentrations and their innermost density profiles are not as steep as $r^{-1.5}$. Thus, though the cosmogony of the halos in both cases is different, the final virialized structures are not dramatically different.  -The c$_{1/5}$ or c$_{\\rm NFW}$ concentrations of halos more massive than a few times the filtering mass \\mfs\\ are similar to those of their CDM counterparts. However, as the mass decreases, the concentrations on average remain almost constant (slightly decrease), while for CDM halos the concentration increase monotonically. The scatter of the concentration for a given mass is large in both cases. The difference in the concentrations of the small halos can be explained because the relatively late formation epoch of small halos in the simulations with damped power spectrum at small scales.  -The relict thermal velocity dispersion of warmons, \\vthw, does not affect the density profiles of WDM halos, at least down to 0.01 \\rv. For our simulations with \\fsl=0.2 and 1.7 Mpc (\\modI\\ and \\modVI, respectively) we used a thermal velocity two times larger than \\vthw\\ and we did not find any significant difference in the halo concentrations and density profiles with respect to the case with \\vth=0. For the high-resolution run \\modVI\\ we also experimented with \\vth= 4 and 16 times \\vthw, finding resolved soft cores only for the last case (with radii at $\\sim0.03-0.04$ \\rv).  -The inner structure of dark matter halos formed through monolithic collapse can be affected {\\it only} if the particles have a significant (tangential) velocity dispersion \\vrms. The penetration degree towards the halo center is determined by the pericenter of the orbiting particle, or equivalently, the angular momentum forces the particles to avoid migration towards inner regions. From N-body simulations of the collapse of top-hat spheres with different amounts of \\vth, we have found that the core radius scales linearly with \\vth\\ up to velocities much larger than \\vthw. In order that WDM halos with masses close to \\mfs\\ form noticeable cores, \\vth\\ should be much larger than \\vthw.  We conclude that a flat $\\Lambda$WDM model with $\\fsl\\approx 0.16$ Mpc ($\\mw\\approx 750$ eV), $\\Omega_{\\Lambda}\\approx0.7$, and $h\\approx 0.7$, has several advantages over its CDM model counterpart. The problem of excess of substructure is solved (\\Colinetal), the halos of dwarf and LSB galaxies are less concentrated than in the CDM scenario as observations suggest (Fig. 9), and the disk angular momentum problem is alleviated (\\cite{SomDol00}). Furthermore, our preferred WDM model describes better than CDM the TF relation of dwarf galaxies (Fig. 10) and the formation epochs of these galaxies (\\S 6), as well as their large-scale distribution (Bode et al. 2000). On the other hand, this model is not apparently in conflict with measurements of the power spectrum of Lyman$-\\alpha$ forest (Narayanan et al. 2000) and with the reionization constraints (Bode et al. 2000). Nonetheless, the WDM model is ruled out if more observations confirm the existence of soft halo cores as well as the independence on mass of their densities; the crucial test is at cluster scales. No doubt, exciting questions remain to be answered in the near future."
    },
    "0010/astro-ph0010655_arXiv.txt": {
        "abstract": "\\rightskip 0pt \\pretolerance=100 \\noindent In the error box of \\source, the brightest as-yet unidentified EGRET source at intermediate Galactic latitude, we find a weak, ultrasoft X-ray source at energies $E < 0.3$~keV in the \\ro\\ All-Sky Survey.  Deep optical imaging at the location of this source, as pinpointed by an observation with the \\ro\\ HRI, reveals a blank field to a limit of $V > 25.2$. The corresponding lower limit on $f_X/f_V$ is 300, which signifies that the X-ray source is probably a thermally emitting neutron star. Considering our previous complete multiwavelength survey of the \\source\\ region which failed to find any other notable candidate for identification with \\source, we propose that this X-ray source, \\xray, is a rotation-powered $\\gamma$-ray pulsar which is either older or more distant than the prototype Geminga.  We see marginal evidence for variability between two \\ro\\ HRI observations.  If real, this would indicate that the X-ray emission has an external origin, perhaps due to intermittent heating of the polar caps by a variable particle accelerator. \\xray\\ could even be an old, recycled pulsar, which may nevertheless have a high $\\gamma$-ray efficiency. ",
        "introduction": "The nature of the persistent high-energy ($> 100$ MeV) $\\gamma$-ray sources in the Galaxy remains an enigma three decades after their discovery.  Ever since the identification of the mysterious $\\gamma$-ray source Geminga as the first radio quiet but otherwise ordinary pulsar (see review by Bignami \\& Caraveo 1996), it has been argued that rotation-powered pulsars likely dominate the Galactic $\\gamma$-ray source population (e.g., Yadigaroglu \\& Romani 1997), and that many of them will be radio quiet. That Geminga is a bright EGRET source at a distance of only $\\sim 160$~pc (Carveo et al. 1996) begs for it not to be unique.  Several excellent neutron-star candidates for EGRET sources have emerged from X-ray observations in recent years. One is present in the supernova remnant CTA1 which is close to 3EG~J0010+7309 (Brazier et al. 1998). Another is in the $\\gamma$-Cygni supernova remnant which is coincident with 3EG~J2020+4017 (Brazier et al. 1996). An intriguing 34~ms X-ray pulsar with a Be star companion lies in the error circle of 3EG~J0634+0521 (Kaaret et al. 2000). X-ray nebulae that are inferred to contain pulsars have been detected in the error circles of the EGRET sources 2EG~J1811--2339 (Oka et al. 1999) and 3EG~J1420--6038 (Roberts \\& Romani 1998; Roberts et al. 1999). Radio-quiet neutron stars have been found which are not apparently EGRET sources, but they are important as members of the growing class of cooling neutron stars (RX~J185635--3754, Walter 2000; PKS~1209--52, Zavlin et al. 2000). Neutron-star candidates discovered by \\ro\\ were reviewed by Caraveo, Bignami, \\& Tr\\\"umper (1996) and by Motch (2000).  As the brightest of the as-yet unidentified intermediate-latitude EGRET sources at $(\\ell, b)=(89^{\\circ},25^{\\circ})$, \\source\\ is a good {\\it a priori} target for the next such identification. We recently completed an exhaustive search for a counterpart of \\source. The observations included deep radio, X-ray, and optical surveys, as well as optical spectroscopic classification of every active object within or close to its 99\\% confidence error ellipse (Mirabal et al. 2000, hereafter Paper~1). In summary, we identified optically all but one of the {\\it ROSAT\\/} and {\\it ASCA\\/} sources in the region of \\source\\ to a flux limit of $\\sim 5 \\times 10^{-14}$~erg~cm$^{-2}$~s$^{-1}$, which is $10^{-4}$ of the $\\gamma$-ray flux, without finding any suggestive evidence for a possible counterpart among the identified sources.  We also proposed that the one unidentified X-ray source, \\xray, is the most promising candidate for identification with \\source\\ principally because the {\\it absence} of an optical counterpart for it is the standard signature of an isolated neutron star, or perhaps a more exotic compact object.  The recent release of the decade-old \\ro\\ All-Sky Survey ({\\it RASS\\/}) data (Voges et al. 2000) provides an important new tool for detecting neutron stars by means of their soft, thermal X-ray emission, and thus for furthering the search for the counterparts of nearby EGRET sources. In this Letter, we report new results from the {\\it RASS} on the previous candidate \\xray\\ which greatly enhance its credentials as the likely pulsar counterpart of \\source.  We also present a detailed study of the X-ray and optical objects in its immediate neighborhood which establishes that \\xray\\ is indeed undetected to a limit of $V > 25.2$, as would be expected for a neutron-star counterpart of \\source. ",
        "conclusions": "The latest analysis of the EGRET observations of \\source\\ leads to the conclusion that it shows no evidence for long-term variability (Reimer et al. 2000).  Its spectrum can be fitted by a relatively flat power law of photon index --1.7 from 70~MeV to 4~GeV, with a turndown above 4~GeV. Such temporal and spectral behavior are more consistent with a rotation-powered pulsar than a blazar, which is the other major class of EGRET source. The soft X-ray source \\xray\\ located in the error box of the \\source\\ presents properties that resemble a neutron star, and could be the second case of a nearby, radio-quiet $\\gamma$-ray pulsar.  The detection of \\xray\\ in the {\\it RASS} 7.5~yr prior to the final HRI observation eliminates an alternative hypothesis mentioned in Paper~1, that it might have been a luminous soft X-ray transient caused by the tidal disruption of a star by a supermassive black hole in a distant galaxy, and unrelated to \\source. Such events are expected (and observed) to last only a few months (e.g., Komossa \\& Bade 1999; Komossa \\& Greiner 1999), while this source has persisted for at least 7.5~yr. Furthermore, our deep optical imaging failed to find a candidate galaxy at this location.  The available data from radio through $\\gamma$-rays are thus consistent with the hypothesis that \\xray\\ is the actual counterpart of \\source\\ and a more distant or older cousin of the Geminga pulsar. The implications of a neutron-star interpretation of \\xray\\ were discussed at length in Paper~1.  Although the intrinsic flux of the source is highly uncertain because of the unknown column density, the latter is probably not greater than $3 \\times 10^{20}$~cm$^{-2}$ or we would not be seeing such a soft spectrum.  Then \\xray\\ is evidently at least 10--40 times fainter than Geminga.  Thus it is either more distant than Geminga ($d > 160$~pc, Caraveo et al. 1996), or cooler ($T < 5 \\times 10^5$~K, Halpern \\& Wang 1997). But a cooling neutron star of age $< 10^6$~yr should not be {\\it too} distant, because to place it $> 400$~pc above the Galactic plane at $b = 25^{\\circ}$ would require a kick velocity at birth of $> 500$~km~s$^{-1}$. A reasonable upper limit on the distance is therefore 1~kpc, which we note implies an isotropic $\\gamma$-ray luminosity of $6 \\times 10^{34}\\,(d/1\\,{\\rm kpc})^2$ ergs~s$^{-1}$, rather more than the spin-down power $I\\Omega\\dot \\Omega$ of Geminga ($3.3 \\times 10^{34}$ ergs~s$^{-1}$).  If the actual spin-down power of \\source\\ is determined, it will place an additional constraint on the distance and beaming factor.  The only observation that possibly complicates this interpretation is the marginally significant long-term (and non-monotonic) variability by about a factor of two.  Such behavior is not consistent with surface thermal radiation of the heat of formation from the interior. However, it is possible that thermal emission from older neutron stars is dominated by the heated polar caps impacted by inflowing relativistic particles from the acceleration regions which are responsible for their pulsar action.  Indeed, such reheating has been shown to be the source of the X-rays from the nearby recycled millisecond pulsar PSR~J0437--4715 (Zavlin \\& Pavlov 1998). If under such conditions the particle acceleration regions in the pulsar were intermittent, the thermal X-rays resulting from this external heating would also be variable. In principle, \\source\\ could be as powerful as Geminga at $\\gamma$-ray energies, but less conspicuous at soft X-ray wavelengths because its surface has cooled except for the $\\sim 1$~km radius polar caps which are being reheated directly by the accelerator. The most efficient $\\gamma$-ray pulsars are expected to be the ones like Geminga, operating closest to their death lines (Chen \\& Ruderman 1993). A millisecond pulsar typically possesses spin-down power and magnetospheric gap voltage similar to Geminga and other ordinary pulsars.  Since there is also evidence that the magnetospheric X-rays from Geminga are variable (Halpern \\& Wang 1997), either an exotic middle-aged pulsar or a maximally efficient recycled pulsar are plausible scenarios for \\source."
    },
    "0010/astro-ph0010463_arXiv.txt": {
        "abstract": "We present a simple novel method for determining the orbital parameters of binary pulsars. This method works with any sort of orbital sampling, no matter how sparse, provided that information on the period derivatives is available with each measurement of the rotational period of the pulsar, and it is applicable to binary systems with nearly circular orbits. We use the technique to precisely estimate the hitherto unknown orbital parameters of two binary millisecond pulsars in the globular cluster 47~Tucanae, 47~Tuc~S and T. The method can also be used more generally to make first-order estimates of the orbital parameters of binary systems using a minimal amount of data. ",
        "introduction": "\\label{sec:introduction}  The discovery of a pulsar whose period changes significantly usually indicates that it is a member of a binary system.  It is then important to determine the Keplerian orbital parameters of the system in order to investigate the astrophysics of the two stars and to obtain a coherent timing solution for the rotation of the pulsar.  The latter provides precise astrometric information as well as the period derivative which may give information on the magnetic flux density and age of the pulsar or, if it is in a globular cluster, on the gravitational field of the cluster \\cite{and92,phi92b}.  The usual procedure for obtaining orbital parameters involves fitting a Keplerian model to a series of period measurements.  Such a procedure works well, provided that it is possible to determine the rotational period on several occasions during a single orbit. However, there are often circumstances where this is not the case, such as where interstellar scintillation permits only sparse positive detections of a pulsar \\cite{clf+00}.   We have therefore sought a way of employing the hitherto unused information contained in the measured orbital accelerations to obtain a full, coherent solution for the Keplerian orbital parameters.  In what follows, we present a simple procedure for estimating the orbital parameters of a binary that is completely independent of the distribution of the epochs of the individual observations, provided that the period derivatives are known in each observation. In section \\ref{sec:general}, we present the equations for the period and acceleration of a pulsar in an eccentric binary system as a function of its position in the orbit. We also provide an analysis of the circular case and show how the orbital parameters can be obtained from a plot of the acceleration of the pulsars against rotational period. In section \\ref{sec:app}, we demonstrate the application of the procedure to a number of pulsars in the globular cluster 47~Tucanae which have known orbital parameters, and then to 47~Tuc~S and T, neither of which hitherto had an orbital solution.  In section \\ref{sec:refining}, we show how to improve the estimates of the orbital parameters, and proceed to refine the orbits of 47~Tuc~S and T. ",
        "conclusions": "The procedure outlined in section \\ref{sec:general} and Appendix A is useful even when the binary's orbit is well-sampled, because it provides automatically a ``first guess'' of the orbital parameters, which can then be used in conventional period analysis. The quality of this guess obviously depends on phase coverage and the precision of the measurement of accelerations. The procedure presented in this paper is, however, the only one capable of giving any estimates of the orbital parameters when the data samples the orbital period sparsely. For this purpose, the epochs of the individual period/acceleration measurements are completely irrelevant as long as the measurements have reasonable orbital phase coverage; their values are only needed to make the appropriate period corrections to the barycentre of the solar system."
    },
    "0010/astro-ph0010149_arXiv.txt": {
        "abstract": "s{ We study the evolution of magnetized clusters in a cosmological environment using magneto-hydro dynamical simulations. Large scale flows and merging of subclumps generate shear flows leading to Kelvin-Helmholtz instabilities, which, in addition to the compression of the gas where the magnetic field is frozen in, further amplify the magnetic field during the evolution of the cluster. Therefore, well-motivated initial magnetic fields of $\\langle B^2\\rangle^{1/2}=10^{-9}\\,{\\rm G}$ reach the observed $\\sim\\mu{\\rm G}$ field strengths in the cluster cores at $z=0$. These magnetized clusters can be used to study the final magnetic field structure, the dynamical importance of magnetic fields for the interpretation of observed X-Ray properties, and help to constrain further processes in galaxy clusters like the population of relativistic particles giving rise to the observed radio halos or the behavior of magnetized cooling flows. } ",
        "introduction": "Observations consistently show that clusters of galaxies are pervaded by magnetic fields of $\\sim\\mu{\\rm G}$ strength.  Coherence of the observed Faraday rotation across large radio sources demonstrates that there is at least a field component that is smooth on cluster scales. The origin of such fields is largely unclear. Models invoking individual cluster galaxies for field generation and amplification generally yield field strengths too low by an order of magnitude.  We used the cosmological MHD code described in Dolag et al. (1999) to simulate the formation of magnetised galaxy clusters from an initial density perturbation field. Our main results can be summarised as follows: (i) Initial magnetic field strengths are amplified by approximately three orders of magnitude in cluster cores, one order of magnitude above the expectation from flux conservation and spherical collapse. (ii) Vastly different initial field configurations (homogeneous or chaotic) yield results that cannot significantly be distinguished. (iii) Micro-Gauss fields and Faraday-rotation observations are well reproduced in our simulations starting from initial magnetic fields of $\\sim10^{-9}\\,{\\rm G}$ strength at redshift 15. Our results show that (i) shear flows in clusters are crucial for amplifying magnetic fields beyond simple compression, (ii) final field configurations in clusters are dominated by the cluster collapse rather than by the initial configuration, and (iii) initial magnetic fields of order $10^{-9}\\,{\\rm G}$ are required to match Faraday-rotation observations in real clusters.  We used these magnetized clusters to study the final magnetic field structure, the dynamical importance of magnetic fields for interpretation of observed X-Ray properties and start to constrain further processes in galaxy clusters like the population of relativistic particles causing the observed radio halos or the behavior of magnetized cooling flows. ",
        "conclusions": ""
    },
    "0010/astro-ph0010112_arXiv.txt": {
        "abstract": "If a light gluino exists, the lightest gluino-containing baryon, the \\OSO, is a possible candidate for self-interacting dark matter.  In this scenario, the simplest explanation for the observed ratio $\\Omega_{dm}/\\Omega_b \\approx 6-10$ is that $m_{S^0} \\sim 900\\;$\\MeVcs; this is not at present excluded by particle physics. Such an \\OSO could be present in neutron stars, with hyperon formation serving as an intermediate stage. We calculate equilibrium compositions and equation of state for high density matter with the \\OSO, and find that for a wide range of parameters the properties of neutron stars with the \\OSO are consistent with observations. In particular, the maximum mass of a nonrotating star is $1.7-1.8\\;M_\\odot$, and the presence of the \\OSO is helpful in reconciling observed cooling rates with hyperon formation. ",
        "introduction": "The interiors of neutron stars offer a unique meeting point between astrophysics on one hand and nuclear and particle physics on the other. The macroscopic properties of neutron stars, such as mass, radius, spin rate and thermal history depend on the microscopic nature of matter at very high densities. Neutrons stars are, in this sense, cosmological laboratories of physics at densities exceeding the nuclear saturation density ($\\rho_0\\simeq0.16\\;$\\fmc). One aspect of matter at supernuclear densities is the possible appearance of hadronic degrees of freedom beyond nucleons. The principle candidates are hyperons, meson condensates and a deconfined quark phase (see, e.g., Glendenning 1996, Prakash et al.~1997 for reviews).  In this work we wish to illuminate another possibility, the appearance of an $R$-baryon. We are motivated by the observation of \\citet{SS_SIDM} that the observed properties of dark matter (DM) on scales of order 10-100 kpc are better described by a theory with self-interacting DM than by conventional, weakly interacting DM. Since the required cross-section to mass ratio, $\\sigma_{XX}/m_X\\sim 1$b/GeV, is characteristic of hadronic physics at low energies, a light stable neutral hadron would be a welcome feature of physics-beyond-the standard model, and such particles arise naturally in theories with a very light gluino.  In some sense, this is the minimal extension of the standard model needed to produce the required DM particle, and therefore merits examination.  A supersymmetric version of the standard model contains a massless gluino (\\gl), the spin-1/2 superpartner of the gluon;  therefore it is endowed with a corresponding chiral symmetry called $R$-invariance. When supersymmetry is broken, the $R$-invariance may or may not break. If it does not, the gluino mass arises only from loops sensitive to the Higgs vevs (which do break $R$-invariance) and is naturally of order $100\\;$\\MeVcs. Although considerable effort has been devoted to excluding it, a very light gluino remains a possibility -- mainly because of the difficult of making firm, first-principles predictions for non-perturbative phenomena in QCD. The question deserves very serious scrutiny because the existence of a light gluino would resolve several puzzles. These issues and the constraints imposed by experiment are reviewed by Farrar (1998, 2000). See also \\citet{fg00ltgl} and references therein.  The lightest supersymmetric ``$R$-hadrons'' are the spin-1/2 \\gl$g$ bound state ($R^0$ or glueballino) and the spin-0 baryon $uds$\\gl~ bound state, the \\OSO. The \\OSO is thus a boson with baryon number unity and a strangeness charge $-1$. The strong color hyperfine interaction unique to the $uds$\\gl~in a flavor singlet state makes the mass of the \\OSO much less than that of all other $R$-baryons \\citep{Farrar84,BuccFarr85}, and probably less than 2 GeV$c^{-2}$, which would make it stable through conservation of $R-$parity \\citep{Farrar96}.  In the very early universe the \\OSO remains in equilibrium with nucleons via processes such as $R^0 N \\leftrightarrow \\SO K$, down to $T \\approx 35$ MeV (Farrar, Spergel \\& Steinhardt 2000; see also in Wandelt et al.~2000). Hence  for a suitable \\OSO mass, slightly less than the proton mass, standard freezeout estimates lead to a density ratio $\\Omega_{\\SO}/\\Omega_B \\approx 6-10$, as required if the \\OSO is to be the dark matter.  To posit a new baryon lighter than a nucleon is quite shocking!  But it is known that the color hyperfine force is very strong in this channel, and there are innumerable examples of the inadequacy of phenomenological models of non-perturbative QCD phenomena. The possibility of such a light \\OSO should not be dismissed on theoretical grounds, and we must seek observational constraints. One possible astrophysical constraint is whether this new baryon can form in neutron stars, and if so, would it imply unacceptable neutron star properties?  Nucleons cannot convert directly into the \\OSO, even if the latter is less massive. Conservation of $R-$parity requires that $R$-hadrons are produced in pairs, forbidding direct conversion of individual nucleons into \\OSO particles.  The weak neutral current process $nn\\rightarrow{\\rm S}^0{\\rm S}^0$ requires simultaneous creation of two units of strangeness ($\\Delta S=-2$).  Strangeness-changing neutral currents ($\\Delta S = \\pm 1, \\Delta Q = 0$) are highly suppressed in the standard model due to the GIM mechanism. Thus, $nn\\rightarrow{\\rm S}^0{\\rm S}^0$ is doubly GIM suppressed; crude estimates indicate that its rate is of order the inverse Hubble time \\citep{FSScoming}. However processes such as $\\Lambda\\Lambda\\rightarrow{\\rm S}^0{\\rm S}^0$ ($\\Delta S=0$) and even $\\Sigma^-p \\rightarrow {\\rm S}^0{\\rm S}^0$ (charged current, $\\Delta S=-1$) are not GIM suppressed, so that if the conditions in interiors of neutron stars allow for hyperons to appear (as is predicted by many works on this issue, see, e.g., Glendenning 1985, Weber \\& Weigel 1989, Knorren, Prakash \\& Ellis 1995, Schaffner \\& Mishustin 1996, Balberg \\& Gal 1997, Huber et al.~1998, Baldo, Burgio \\& Schulze 2000), these hyperons could serve as a doorway for \\OSO formation, which might affect observable properties of neutron stars.  Our goal in this {\\it letter} is to explore whether a light \\OSO can form in neutron stars and assess whether the resulting global properties of these stars are compatible with observations. In \\S~\\ref{sect:interactions} we extend an equation of state for high density matter with hyperons to include the \\OSO.  The equilibrium composition with the \\OSO is found in \\S~\\ref{sect:composition}, followed by a discussion of the corresponding properties of neutron stars in \\S~\\ref{sect:EoSandNS}. ",
        "conclusions": "\\label{sect:concdis}  Our primary conclusion is that the existence of a light ($m_{S^0}\\lesssim m_p$) \\OSO particle is consistent with the fundamental observed properties of neutron stars, namely, their inferred masses.  This conclusion applies for a wide range of models for high density matter with the \\OSO.  In particular, the maximum mass of a static neutron star under nominal assumptions regarding the \\OSO is found to be about $1.8\\;M_\\odot$. We note that since it suppresses hyperon accumulation, the appearance of the \\OSO yields a stiffer equation of state and higher maximum mass than obtained with nucleons and hyperons without \\OSO.  Therefore, identification of a relatively high mass neutron star would be easier to reconcile with an \\OSO component than in its absence (where hyperons accumulate undisturbed). Some indication for the existence of such massive neutron stars is implied by recent analyses, although uncertainties are still significant \\citep{vKW}. We emphasize that {\\it current} data on neutron star masses is not restrictive enough to decisively constrain the properties of high density matter.  The presence of \\OSO particles could have a significant and welcome effect on the cooling rate of neutron stars, if they do indeed replace hyperons in the cores. Hyperons can participate in the direct Urca cooling process ($B_1\\rightarrow B_2 e\\nu$) while the \\OSO cannot. The threshold fraction required for hyperons to drive direct Urca cooling processes is of the order of 0.01 \\citep{PPLP92}, so neutron stars with hyperons are susceptible to rapid cooling, which seems inconsistent with observations \\citep{Schaabal96}.  Hyperon direct Urca cooling can be partially suppressed if the $\\Lambda$ hyperons pair to a $^1{\\rm S}_0$ superfluid state \\citep{SBSB98}, but the extent of suppression is sensitive to the details of the equilibrium composition and gap energy \\citep{Pageal00}. On the other hand, \\OSO formation would prevent the hyperon direct Urca, or restrict it only to the highest mass stars, so that very moderate cooling rates are attainable. Future measurements of surface thermal emission from young neutron stars with the {\\it CHANDRA} and {\\it XMM} satellites may lead to stronger constraints on cooling rates; if so this will yield valuable insight into the internal structure of high density matter.   The research of GRF is supported in part by NSF-PHY-99-96173. The idea that the \\OSO could be a dark matter candidate, and constraints on such a scenario, were developed by GRF in collaboration with D.~Spergel and P.~Steinhardt; that work will be reported in greater depth elsewhere \\citep{FSScoming}; a brief description can be found in \\citep{wandelt00}. SB and TP thank NYU for hospitality which facilitated this research."
    },
    "0010/astro-ph0010438_arXiv.txt": {
        "abstract": "{We present a moderately-deep $JK$ photometry for three selected areas of the dwarf irregular galaxy IC\\,1613. The color-magnitude diagrams contain a mixture of red supergiants, asymptotic giant branch stars and the brightest red giant stars. The  red supergiants are massive (20 - 25$\\cal M_\\odot$)  and  young --- with ages between  8 and 25 Myr.  The most important result is the evidence of the decreasing density of the intermediate age AGB population in the vicinity of the HII regions in the galaxy. We also find age differences between AGB stars in the main body of the galaxy and those near the HII regions in the North-East.  The former span a range in ages between 1 and 10 Gyr, while the latter are younger than 1 Gyr.  Using  the period-luminosity relation derived by Madore \\& Freedman (1991) and $JK$ magnitudes of the Cepheid variable V20, we calculated $(m-M)_{K} = 24.37\\pm0.2$. The recently discovered Nova (King et al. 1999) was identified in Field III. Its presence of our images and its brightness questioned its classification as a nova.} ",
        "introduction": "Studies of bright stars in nearby galaxies provide a direct link to their history of star formation.  Photometric investigations of red supergiants (RSG) and  asymptotic giant branch (AGB) are of particular interest, as these stars probe the evolution during early and intermediate epochs.  Carrying out such studies at near-infrared (NIR) wavelengths have a number of advantages.  NIR observations are less sensitive to both the reddening towards and within galaxies.  They allow us to overcome the effects of line blanketing, which can affect the spectral energy distributions of moderately metal-rich giants at optical wavelengths.  In addition, they are the best choice for determining bolometric corrections and for the analysis of the bolometric luminosity function of the luminous, cool evolved stars.  Several recent studies of the red stellar content in nearby spiral and irregular galaxies exploiting these advantages include Davidge (1999: M\\,33), Davidge (1998: NGC\\,55, NGC\\,300, and NGC\\,7793), and Cioni et al. (1999: LMC).  Here, we present NIR observations of IC\\,1613. IC\\,1613 is a faint, irregular galaxy within the Local Group. Recently, Cole et al. (1999) studied the history of star formation in the central part of the galaxy using HST $VI$ photometry.  They found a dominant old stellar population --- red giant branch and red clump stars, and a prominent sequence of intermediate age AGB stars. Cole et al. (1999) deliberately chose their region for study to avoid known HII regions.  In this paper, we present $JK$ photometry of the three selected areas of IC\\,1613.  We chose our fields to extend the area covered by Cole et al. (1999) towards the active HII areas. The goal of this work is to analyze the bright, red stellar content and to determine the distance of IC\\,1613. ",
        "conclusions": "In summary, we find substantial evolved stellar populations in our three fields in IC\\,1613, including the region of ongoing star formation. The brightest stars are massive (20 - 25$\\cal M_\\odot$) red supergiants with $K$ magnitudes between $13 - 13.5$ and $(J-K) \\approx 1$.  Based upon a comparison with the Padua isochrones (Bertelli et al. 1994), these brightest RSGs are very young, with ages between  8 and 25 Myr.  The main feature of the NIR CMDs is the rich intermediate age AGB population. In the central part of the galaxy (our Fields II and III), the tip of AGB stars can be located at $K = 16.20$. There is an indication of lack of ABG stars in Field I (centered on most active HII region of the galaxy) that are redder than $(J-K)_{0} = 1$.  All of the AGB stars in Field I are younger than 1 Gyr, whereas those in the Fields II and III span ages from 1 to 10 Gyr. Using  the PL relation derived by Madore \\& Freedman (1991) along with our $J$ and $K$ magnitudes for the Cepheid variable V20 we calculate a distance modulus $(m-M)_{K} = 24.37\\pm0.2$. In Field III we identify an object at the position of the recently discovered Nova 1999 (King et al. 1999).  Its presence in our IR frames one year before discovery indicate that this star is too bright to be a classical Nova."
    },
    "0010/astro-ph0010324_arXiv.txt": {
        "abstract": "s { The knowledge of the redshift of multiple images in cluster-lenses allows to determine precisely the total projected mass within the Einstein radius. The observation of various multiple images in a same cluster is opening new possibilities to constrain the curvature of the universe. Indeed, although the influence of $\\Omega_m$ and $\\Omega_\\lambda$ on the images formation is of the second order, observations of many multiple images at different redshifts formed by a regular cluster-lens should allow to constrain very accurately the mass distribution of the cluster and to start to be sensitive to the cosmological parameters entering the diameter angular distances.  We present, analytical expressions and numerical simulations that allow us to compute the expected error bars on the cosmological parameters provided an HST/WFPC2 resolution image and spectroscopic redshifts for the multiple images. Numerical tests on simulated data confirm the rather small uncertainties we could obtain this way for the two popular cosmological world models: $\\Omega_m=0.3{\\pm 0.24}$, $\\Omega_\\lambda=0.7{\\pm 0.5}$ or $\\Omega_m=1.{\\pm 0.33}$, $\\Omega_\\lambda=0.{\\pm 1.2}$. Our method can be applied from now on, on real clusters.} ",
        "introduction": "Recent works on constraining the cosmological parameters using the CMB and the high redshift supernovae seem to converge to a new  ``standard cosmological model'' favouring a flat universe with $\\Omega_m\\sim 0.3$ and $\\Omega_\\lambda\\sim 0.7$: White~\\cite{White} and references therein. However these results are still uncertain and depend on some physical assumptions, so the flat $\\Omega_m=1$ model is still possible (Le Dour {\\it et al.}~\\cite{LeDour}). It is therefore important to explore other independent techniques to constrain these cosmological parameters.  In cluster gravitational lensing, the existence of multiple images -- with known redshifts -- given by the same source allows to calibrate in an absolute way the total cluster mass deduced from the lens model. The great improvement in the mass modeling of cluster-lenses that includes the cluster galaxies halos (Kneib {\\it et al.}~\\cite{Kneib96}, Natarajan \\& Kneib~\\cite{Natarajan}) leads to the hope that clusters can also be used to constrain the geometry of the Universe, through the ratio of angular size distances, which only depends on the redshifts of the lens and the sources, and on the cosmological parameters. The observations of cluster-lenses containing large number of multiple images lead Link \\& Pierce~\\cite{Link} (hereafter LP98) to investigate this expectation. They considered a simple cluster potential and on-axis sources, so that images appear as Einstein rings. The ratio of such rings is then independent of the cluster potential and depends only on $\\Omega_m$ and $\\Omega_\\lambda$, assuming known redshifts for the sources. According to them, this would allow marginal discrimination between extreme cosmological cases. But real gravitational lens systems are more complex concerning not only the potential but also off-axis positions of sources. They conclude that this method is ill-suited for application to real systems.  We have re-analyzed this problem building up on the modeling technique developed by us. As demonstrated below, we reach a rather different conclusion showing that it is possible to constrain $\\Omega_m$ and $\\Omega_\\lambda$ using the positions of multiple images at different redshifts and some physically motivated lens models.  Troughout this paper we have assumed $H_0=65$ km s$^{-1}$ Mpc$^{-1}$, however the proposed method is independant of the value of $H_0$. ",
        "conclusions": "Following the work of LP98, we discussed a method to obtain informations on the cosmological parameters $\\Omega_m$ and $\\Omega_\\lambda$ while reconstructing the lens gravitational potential of clusters with multiple image systems at different redshifts.  This technique gives degenerate constraints, $\\Omega_m$ and $\\Omega_\\lambda$ being negatively correlated, with a better constraint of the matter density. With a single cluster in a typical lensing configuration we can expect the following error bars : $\\Omega_m=0.3{\\pm 0.24}$, $\\Omega_\\lambda=0.7{\\pm 0.5}$. To perform that, several general conditions must be fulfilled:  $\\ast$ a cluster with a rather regular morphology, so that a few parameters are needed to describe the gravitational potential ; this is not so restrictive because we saw that a bimodal cluster can also provide a constraint,  $\\ast$ ``numerous'' systems of multiple images, probing each part of the cluster,  $\\ast$ a good spatial resolution image (HST observations) of the cluster and arcs to compute relatively precise -- 0.1'' -- (flux weighted) images positions,  $\\ast$ spectroscopic precision for the different redshifts that should be also regularly distributed from $z_l$ to high values -- this requires deep spectroscopy on 8-10m class telescopes due to the faintness of the multiple images .  It is important to notice that one cluster could provide one constraint on the geometry of the whole universe. And it is possible to combine data from different clusters to tighten the error bars. Combining the study of about 10 clusters would lead to meaningful constraints. The dashed lines confidence levels in the Fig. \\ref{3fam} are the result of a numerical simulation made with 10  {\\it identical} clusters.  Actually the degeneracy depends only on the different redshifts involved that we will have various sets of when applying the method to real configurations. This should lead to a more reduced area of allowed cosmological parameters. We are encouraged by more and more known observations including systems with multiple sources and we plan to apply this technique to clusters like MS2137-23, MS0440+02, A370, AC114 and A1689."
    },
    "0010/astro-ph0010054_arXiv.txt": {
        "abstract": "We have determined elemental abundance ratios in the core of the Perseus cluster for several elements. These ratios indicate a central dominance of SN Ia ejecta similar to that found for A496, A2199 and A3571 (Dupke \\& White 2000a). Simultaneous analysis of {\\sl ASCA} spectra from SIS1, GIS 2\\&3 shows that the ratio of Ni to Fe abundances is $ \\sim 3.4 \\pm 1.1 $ times Solar within the central 4$^\\prime$. This ratio is consistent with (and more precise than) that observed in other clusters whose central regions are dominated by SN Ia ejecta.  Such a large Ni over-abundance is predicted by ``Convective Deflagration'' explosion models for SN Ia such as ``W7'' but is inconsistent with delayed detonation models. We note that with current instrumentation the \\nika\\ line is confused with \\fekb\\ and that the Ni over-abundance we observe has been interpreted by others as an anomalously large ratio of \\fekb\\ to \\feka\\ caused by resonant scattering in the \\feka\\ line. We argue that a central enhancement of SN Ia ejecta and hence a high ratio of Ni to Fe is naturally explained by scenarios that include the generation of chemical gradients by suppressed SN Ia winds or ram-pressure stripping of cluster galaxies. It is not necessary to suppose that the intracluster gas is optically thick to resonant scattering of the \\feka\\ line. ",
        "introduction": "The existence of heavy elements in the intracluster medium (ICM) has been known since the discovery of the $\\sim 6.7$ keV \\feka\\ line in the Perseus cluster (Mitchell et al. 1976) and indicates that part of the ICM is not primordial, i.e. was processed in stars and injected into the ICM. If the emitting plasma is assumed to be optically thin then the iron line flux can be converted into an elemental abundance. When this is done the inferred ICM iron abundances are typically in the range 0.3--0.4 Solar (Mushotzky \\& Loewenstein 1997). The measurements of abundances of iron and other elements are key data used in constraining models of the formation and evolution of the ICM. Therefore, it is important to check that line fluxes can be unambiguously converted to elemental abundances.  A key assumption is that the plasma is optically thin. However, Gilfanov, Sunyaev \\& Churazov (1987) pointed out that this may not always be true for the \\feka\\ line. Resonant scattering (the absorption and immediate re-emission of an \\feka\\ line photon by an Fe ion) could be significant in the cores of clusters. If so, the line emission from the cluster core would be reduced because of photons being scattered out of the line of sight and the measured Fe abundance would be an underestimate of the true intracluster gas abundance.  The simplest test for resonant scattering in the \\feka\\ line is to compare the line flux with that from the \\fekb\\ line. The optical depth in the \\fekb\\ line is $\\sim 20\\%$ of the \\feka\\ so that if the \\fekba\\ ratio is anomalously high then resonant scattering must be taken into account.  Such anomalous ratios have been observed with low significance in data from the HEAO 1 (Mitchell \\& Mushotzky 1980) and {\\sl Tenma\\ } (Okumura \\etal 1988) satellites. Recently, these results have been placed on a much firmer basis using observations of several clusters using {\\sl ASCA} (Akimoto \\etal 1997; Yamashita \\etal 1999) and of the Perseus cluster using \\sax\\ (Molendi \\etal 1998). The best data are from Molendi \\etal\\ who show that the \\fekba\\ ratio is anomalously high for the inner 6 $^\\prime$ of the Perseus cluster. Outside that radius the ratio decreases to the expected value for an optically thin plasma. They deduce from this that the Fe abundance is underestimated by about a factor of 2 in the center of the Perseus cluster.  However, at the resolution of current X-ray spectrometers the \\fekb\\ line is confused with the \\nika\\ line so that a measurement of the \\fekba\\ ratio requires an assumed flux in the \\nika\\ line. The results reported above assumed that the ratio of Ni and Fe abundances is similar to the Solar value (0.038 by number - photospheric value; Anders \\& Grevesse 1989). To make the \\fekba\\ ratio consistent with the optically thin case requires a Ni/Fe abundance ratio several times that of Solar. To see whether this is reasonable it is necessary to consider how heavy elements arrived in the ICM.  $Einstein$ FPCS spectroscopy (Canizares et al.\\ 1982) and more recent {\\sl ASCA} spectroscopy (Mushotzky \\& Loewenstein 1997; Mushotzky et al.\\ 1996) suggest that global intracluster metal abundances are consistent with ejecta from Type II supernovae (SN II). This implies that the bulk of the heavy elements were injected into the ICM by protogalactic winds. This is further supported by analysis of ICM energetics (White 1991). Since SN II do not produce much Ni we do not expect Ni/Fe abundance ratios large enough to restore the \\fekba\\ ratio to the optically thin value. However, there are recent indications that SN Ia ejecta can contribute with a significant fraction (up to 50\\%) of the ICM metals (Ishimaru \\& Arimoto 1997; Fukazawa et al.\\ 1998; Nagataki \\& Sato 1998, Finoguenov \\& Ponman 1999, Allen et al. 2000).  Clues to metal enrichment mechanisms in clusters can be derived from the analysis of clusters with central abundance enhancements (abundance gradients). Analysis of spatially resolved X-ray spectra has shown that in several clusters the metal abundance is centrally enhanced (Koyama, Takano \\& Tawara 1991; White et al.\\ 1994; Fukazawa et al.\\ 1994; Ulmer et al.\\ 1987; Ponman et al.\\ 1990; Kowalski et al.\\ 1993; Arnaud, et al.\\ 1994; Ikebe et al.\\ 1997; Ezawa et al.\\ 1997; Xu et al.\\ 1997; Pislar et al.\\ 1997; Dupke 1998; Dupke \\& White 2000a,b; Irwin \\& Bregman 1999; Finoguenov, David \\& Ponman 1999; White 1999; Allen et al. 1999). There are several mechanisms which may cause central abundance enhancements in the ICM including: 1) the radial profile of intracluster gas density is shallower than that of metal-injecting galaxies, i.e., if metals trace the galaxies that injected them into the ICM and these galaxies are more concentrated in the center than the intracluster gas one would expect to find a general abundance enhancement towards the clusters' central regions; 2) mass loss from the stars in central dominant galaxies may accumulate near the cluster center; 3) ram-pressure stripping of the metal-rich gas in cluster galaxies by intracluster gas is more effective at the center, where the intracluster gas density is highest; 4) secular SN Ia winds in central dominant galaxies which are partially suppressed due to the galaxy location at the bottom of the cluster gravitational potential, where the intracluster gas density is highest (Dupke \\& White 2000b). Some of the above mentioned mechanisms could deposit substantial amounts of SN Ia enriched material near the clusters' center, therefore creating a ``chemical gradient''. Since SNe Ia and II have different elemental mass yields, analysis of the elemental abundance ratios provides information on the fractional contribution of gas enriched by SN Ia and II. The analysis of abundance ratios in the central regions of clusters having abundance gradients indeed indicates a dominance of SN Ia ejecta (Dupke \\& White 2000a,b see below).  As mentioned above, SN Ia metal injection mechanisms such as ram-pressure stripping or post-protogalactic suppressed SN Ia winds may contribute significantly to ICM enrichment. SN Ia can generate large amounts of Ni. ``Convective Deflagration'' models such as W7 (Nomoto, Thieleman, \\& Yokoi 1984; Thieleman, Nomoto \\& Yokoi 1986) predict a Ni/Fe number abundance ratio $\\sim 5$ times Solar. Alternative slow flame-speed (delayed detonation) models produce a lower Ni/Fe abundance ratio closer to that from SN II ($\\approx$ 1.6). Recently, Dupke \\& White (2000a) and Dupke (1998) have found evidence that in the central regions of A496, A2199, and A3571 the ICM is enriched predominantly by SN Ia ejecta. In these objects the ratio of Fe abundance to that of $\\alpha$-process elements increases with decreasing radius from the cluster center. These results suggest that, in clusters for which the abundance profile is centrally enhanced, SN Ia ejecta provide most of the total iron mass in the central regions (see also Allen et al. 1999).  In the outer regions the SN Ia contribution is smaller than in the cluster's center but not negligible (e.g. Nagataki \\& Sato 1998). In A496 the excess iron that generates the central abundance gradient is fully accounted for by the excess SN Ia ejecta that produce a chemical gradient (Dupke \\& White 2000b). In all three clusters mentioned above the Ni/Fe abundance ratio is consistent with that predicted by the SN Ia ``Convective Deflagration'' W7 model, i.e., the Ni/Fe ratio is several times Solar (assuming that the \\fekba\\ ratio takes its optically thin value). The analysis of A1060 (the only cluster with flat abundance distribution in Dupke's sample (1998)) shows no significant abundance ratio trend in its radial distribution (including the Ni/Fe ratio). However, since the number of clusters with detailed measured abundance ratio distributions is relatively small, it is not clear whether this ``chemical gradient'' is always true for clusters with central abundance gradients.   This work extends the analysis of Dupke \\& White (2000a) to the Perseus cluster. This cluster is closer and brighter than A496, A2199, and A3571 so we can determine elemental abundance ratios with higher precision. The Fe abundance is known to be centrally enhanced in Perseus (Ulmer \\etal 1987; Ponman \\etal 1990; Kowalski et al.\\ 1993; Arnaud \\etal 1994; Molendi \\etal 1998). If gradients in abundance and abundance ratios are related there should be a significant contribution from SN Ia ejecta in the regions where the abundance is enhanced. We argue below that the central regions of the Perseus cluster do show evidence for SN Ia ejecta dominance, that the Ni/Fe abundance ratio is that predicted by the W7 models, and hence there is no evidence for resonant scattering in the \\feka\\ line and the measured Fe abundance is reliable. This result also argues against delayed detonation models for SN Ia and illustrates the power of X-ray observations of clusters of galaxies to learn about supernova physics. We note that there are no measurements of Ni abundances from observations of Galactic supernova remnants because these objects are too cool to provide significant counts in the Ni K lines. ",
        "conclusions": "In this work we have shown that \\begin{enumerate} \\item The central region (0-4$^\\prime$) of the Perseus cluster is significantly enriched with ejecta from SN Ia, which produce more than half ($\\sim 67\\%$) of the total iron. \\item The observed Ni/Fe ratio is in excellent agreement with the observed ratios of other elements involving Fe, Si, Ni and O in predicting the SN Ia Fe mass fraction in the central regions of Perseus . The Ni/Fe ratio observed is $\\sim$ 3.4 . \\item The Ni/Fe ratio is consistent with the theoretically predicted values for SN Ia models with fast flame-speed (W7) but not with delayed detonation explosion models. \\item The average Si/Fe ratio in Perseus is significantly ($>$90\\% confidence) higher than in the central regions, which indicates that the central Fe abundance gradient is accompanied by a chemical gradient, i.e., the SN Ia contamination has a radial decline in Perseus. \\item It is not necessary to invoke resonant scattering in the \\feka\\ line since the high \\fekb/\\feka\\ ratio can be totally attributed to substantial ICM contamination from SN Ia ejecta. \\end{enumerate}  Although the high central Ni abundances can be fully explained by SN Ia enrichment, the current data do not allow a precise quantification of the relative contribution of resonant scattering, which can be done only with very high resolution X-ray spectroscopy. If resonance scattering is important it will always provide the largest effect at the peak density of the cluster. Therefore, clusters with high resonant scattering optical depth should show, observationally, a Ni abundance profile decreasing with radius (within the resolution of current X-ray spectrometers). Therefore, additional discriminating evidence between the two scenarios can be provided by observation of other clusters that show no central SN Ia ejecta dominance and, at the same time, have large resonant scattering optical depth."
    },
    "0010/astro-ph0010262_arXiv.txt": {
        "abstract": "As presented by Nice et al. (2000), long--term timing of the eclipsing binary PSR B1744--24A shows that the orbital period of this system  decreases with a time--scaleof only $\\sim$ 200 Myr. To explain the much faster orbital period decay than that predicted by only emission of the gravitational waves ($\\sim$ 1000 Myr) we propose that the orbital evolution of this system is also driven by magnetic braking . If magnetic braking is to explain the rapid decay of the orbit, then $\\lambda$ characterizing the effectiveness of the dynamo action in the stellar convection zone in the magnetic stellar wind formula must be equal to 1. ",
        "introduction": "The formation and evolution of a very low--mass binary system consisting of a neutron star (millisecond pulsar) and a 0.1 $M_\\odot$ companion is not yet well understood. One of the main observational features of these systems is that they are eclipsing. One interesting eclipsing binary system, PSR B1744--24A, is in the globular cluster Terzan 5. The neutron star is an 11.56 ms pulsar in a 1.8 h orbit with a low--mass companion (Lyne et al. 1990). The duration of the eclipse in this system is very variable and is never less than one--third of the orbital period. Lyne et al.(1990) estimated that the energy flux from the pulsar at the companion surface (isotropic radiation is assumed) is $<$ 2$\\times 10^{10}$ erg s$^{-1}$cm$^{-2}$ which is near  the critical value when the irradiation may influence the structure of a low--mass star (Podsiadlowski 1991). However, this value is at least a factor of 15 less than for another eclipsing millisecond binary pulsar system PSR B1957+20 (Fruchter et al. 1990). This small irradiative flux value and great variations in duration of the eclipse lead Lyne et al. (1990) to the conclusion that in this system Roche lobe overflow may occur.  Recent  long--term timing observations of the eclipsing binary pulsar PSR B1744--24A at the VLA  and Green Bank, show that the orbital period of PSR B1744--24A has decreased, with a time--scale of only $P_{orb}/\\dot{P}_{orb}$$\\sim$200 Myr. This is five times faster  than if the orbital period decay is driven only by the gravitational wave radiation (Nice et al. 2000).  The key question of interest is: how did this system evolve?  In this short note we  show that the new observational data for the orbital period decay of PSR B1744--24A may give us a unique opportunity to test the magnetic braking mechanism if the  observed orbital period decay is  secular. ",
        "conclusions": "If the observed orbital period decay is secular and our suggestion is correct that the orbital evolution of PSR 1744--24A is driven by magnetic braking then this may be the first case when the orbital period decay by a magnetic braking mechanism has been measured.  According to our model calculations the best fit to the observed orbital period decay is obtained when $\\lambda$ is near to one.  Which means that the value proposed by  Smith (1979)  $\\lambda$=1.78 underestimates, that and proposed by Skumanich (1972) $\\lambda$=0.73 sligtly,  overestimates  the magnetic braking efficiency."
    },
    "0010/astro-ph0010168_arXiv.txt": {
        "abstract": "We present color magnitude diagrams of four blue massive and compact star clusters in M31: G38, G44, G94, and G293.  The diagrams of the four clusters reveal a well-populated upper main sequence and various numbers of supergiants.  The U-B and B-V colors of the upper main sequence stars are used to determine reddening estimates of the different lines of sight in the M31 disk. Reddening values range from $E_{B-V} = 0.20\\pm0.10$ to $0.31\\pm0.11$. We statistically remove field stars on the basis of completeness, magnitude and color. Isochrone fits to the field-subtracted, reddening-corrected diagrams yield age estimates ranging from $63\\pm15$ Myr to $160\\pm60$ Myr. Implications for the recent evolution of the disk near NGC 206 are discussed. ",
        "introduction": "Studies of the stellar content of extra-galactic globular clusters (hereafter GCs) started in the Magellanic Clouds (e.g. \\citeNP{monograph}).  These observations led to the recognition of several blue GCs whose ages were found to range from $\\sim$10$^8$ - 10$^9$ years (\\citeNP{arp1958}; \\citeNP{arp1959}; \\citeNP{hodge1961}; \\citeNP{hodge1984}; \\citeNP{brocato1989}; and many others).  The Galaxy appears not to have such young massive and compact clusters, though other more distant spirals do \\cite{larsen}.  If they are young GCs, observations may uncover valuable clues about the formation and evolution of GCs.  Regardless of their classification, these clusters are certainly useful probes of the reddening in their portion of the galaxy because they contain upper main sequence stars at a common distance, which have nearly the same color in B-V independent of metallicity.   Using photoelectric photometry, \\citeN{vandenbergh} observed that some of the objects in M31 classified as GCs were very blue, including the four clusters discussed here: G38, G44, G94, and G293 (designations are from \\citeNP{sargent}).  Integrated U, B, and V photometry of hundreds of clusters was done with photographic plates by \\citeN{battistini}, confirming the blue colors of these four and several others.  With the Ultraviolet Imaging Telescope aboard the Astro-1 mission, \\citeN{bohlin} detected 43 M31 GCs in the ultraviolet, including G94, and its UV flux was consistent with its being a young cluster.  \\citeN{huchra1991} completed spectroscopic metallicity measurements for 150 GCs in M31, including G38.  Its metallicity measurement has high uncertainty and may be significantly underestimated because it was determined without the knowledge of its young age.  The first M31 globular cluster resolved into stars was studied from the ground by \\citeN{heasley1988}.  With sub-arcsecond seeing observations from Mauna Kea, they resolved the red giant branch of G1. More recently, the populations of star clusters in M31 have been resolved with from space (e.g. \\citeNP{ajhar1996}, \\citeNP{rich1996}, \\citeNP{pecci1996}, \\citeNP{holland1997}, \\citeNP{jablonka2000}). These studies have used images from the Wide Field and Planetary Camera (WFPC-2) aboard the Hubble Space Telescope (HST) to construct color-magnitude diagrams (CMDs) of M31 star clusters in order to determine their metallicities, reddening values, and ages; in some cases, comparisons were made between the cluster and field populations (e.g. \\citeNP{jablonka2000}).  We are using the same instrument to make CMDs of four of the massive and compact young clusters: G38, G44, G94 and G293.  We begin in section 2 with a description of our observations and reduction techniques, followed in section 3 by the details of our field subtraction technique.  In section 4 we discuss the determination of the reddening values and luminosity determinations for each cluster, and section 5 describes the age determinations. Finally, in section 6 we provide a summary of our results. ",
        "conclusions": "We have constructed CMDs for four massive and compact star clusters in the disk of M31.  The CMDs reveal an upper main sequence stellar population showing that these are indeed young clusters.  The reddening values determined using the color of the upper main sequence stars are comparable to typical reddening values determined for globular clusters in the M31 halo, suggesting that three of these clusters may be closer to the near side of the disk.  The fourth case shows higher reddening, suggesting that it is located deeper in the disk.  None of the reddening values are extremely high, which suggests that these clusters have formed in regions of relatively low dust content.  Age determinations from statistical matching of CMD properties as well as those from isochrone fitting give consistent ages for all four clusters.  These ages range from $63\\pm15$ Myr to $160\\pm60$ Myr.  The ages of G38 and G44 show that they formed at the same time as the well-studied surrounding Cepheid population, strengthening the conclusions of {\\citeN{magnier1997}.  The clusters' luminosities and half light radii imply that they are more massive and compact than any young clusters in our Galaxy; however, their luminosities are quite similar to young clusters found in many other galaxies \\cite{larsen}. Apparently, these massive and compact young clusters are common to most galaxies, including the Magellanic Clouds and M31, and it is very curious that our Galaxy contains no similar clusters."
    },
    "0010/astro-ph0010442_arXiv.txt": {
        "abstract": "A radio source of the Fanaroff--Riley type II with the angular size of 11.3 arc min is identified with an optical galaxy at $z=0.229$. Thus, the projected linear extent of the radio structure is 3.14 Mpc which makes it the third largest classical double radio source known after 3C236 and WNB2147+816. The high-frequency VLA observations, a galaxy identification and its optical spectroscopy are reported. The equipartition magnetic field and energy density in the source is calculated and compared with corresponding parameters of other giants known, indicating that either parameter of the source investigated is extremely low. On the other hand, the age estimate of relativistic electrons and the advance speed of the lobe material are comparable to the respective parameters characterizing other low-luminosity giant sources, as well as much smaller and brighter 3CR sources. The environment analysis suggests that J1343+3758 lies in a significantly poor region of intergalactic medium. ",
        "introduction": "In this letter, we report a discovery of the third largest radio galaxy in the Universe. Known from years, the record holder radio galaxy 3C236 with the linear extent of 5.65 Mpc (if $H_{o}=50$ km~s$^{-1}$~Mpc$^{-1}$ and $q_{o}=0.5$ which we assume hereafter) was mapped with the WSRT over 20 years ago (cf. Willis, Strom \\& Wilson 1974; Strom \\& Willis 1980). The second largest, recently discovered, 3.56-Mpc radio galaxy is WNB2147+816 (Palma et al. 2000). Proceeding with our investigation of low-luminosity giant radiosource candidates (cf. Jamrozy \\& Machalski 1999; Machalski \\& Jamrozy 2000), selected from the VLA surveys: FIRST (Becker, White \\& Helfand 1995) and NVSS (Condon et al. 1998), we focussed our attention on the source J1343+3758. This source can be discerned in the 325-MHz WENSS survey (Rengelink et al. 1997) and 1.4-GHz VLA NVSS survey as a 11.3 arc min large FRII-type radio source. The source consists of the two extended radio lobes, possibly connected with a very dimmed bridge. At the outer edges of both lobes there are more compact bright regions clearly visible on the relevant FIRST map J134300+38071E.COADD.1, however from that FIRST map alone one could not be sure that these bright regions belong to the same source. The region in the NE lobe contains a very bright hot spot. The source was also detected during the low-resolution sky surveys at 151 and 232 MHz (6C2: Hales, Baldwin \\& Warner 1988; Zhang et al. 1997, respectively). In the 408 MHz and 4.85 GHz surveys (B3: Ficarra, Grueff \\& Tomassetti 1985; GB6: Gregory et al. 1996, respectively), only its SW lobe was detected above the survey's limit.  The FIRST map may suggest a radio core with 1.4-GHz flux density of about 1.5 mJy at the J2000 position: RA $13^{\\rm h}42^{\\rm m}54\\fs 53$, Dec. $+37\\degr 58\\arcmin 18\\farcs 1$, but the rms noise on this map (0.15 mJy beam$^{-1}$) produced other spots of comparable intensity. Therefore, we have made additional high-frequency VLA observations which allow us to confirm its radio core and firmly identify the source with a galaxy at the redshift of 0.229. This VLA observations are described and the radio spectrum of the entire source and its lobes are determined in Sect.~2. The optical spectroscopy of the identified galaxy is described in Sect.~3. In Sect.~4, the equipartition magnetic field and energy density within the lobes as well as the lifetime of relativistic electrons at the frequency of 325 MHz and the mean advance velocity of the lobes, are calculated and compared with the respective values found for other giants as well as for much smaller 3CR sources by Ishwara-Chandra \\& Saikia (1999). Finally, the environment density is estimated by the simple counts of galaxies around the identified galaxy and the calculation of particle density around the radio lobes. ",
        "conclusions": ""
    },
    "0010/astro-ph0010397_arXiv.txt": {
        "abstract": "We present striking results from {\\sl Rossi X-ray Timing Explorer (RXTE)} observations of the 21-hr low mass X-ray binary \\source, showing five complex dips in unprecedented detail. For the first time, dipping is detected up to 15 keV. Prominent flares are also observed in the light curves, limited to energies above $\\sim$8~keV.  Spectra selected by intensity during dip episodes can be well fit with a two-component model consisting of a point-like blackbody from the neutron star and progressive covering of an extended Comptonized region, presumably an accretion disk corona (ADC), corrected for photons scattered into and out of the X-ray beam by a interstellar dust halo. We find that the outer regions of the absorber are highly ionized and that electron scattering is totally responsible for the X-ray attenuation during shallow dipping. The timescales of dip ingress and egress indicate that the envelope of material absorbing the ADC has smaller angular size than the ADC itself, and that the ADC is likely limited to a height-to-radius ratio of 10\\%, rather than being spherical in extent.  In addition, we have analyzed $\\sim$4.5 yrs of {\\sl RXTE} All Sky Monitor (ASM) coverage to derive the first accurate orbital ephemeris for \\source, with phase zero (the time of dip centers) well-described by the relation 245~0088.63918(69) $+$ N $\\times$ 0.869907(12) (JD). ",
        "introduction": "The study of X-ray dipping sources has played a critical role in the understanding and visualization of how a low mass X-ray binary (LMXB) might look to a nearby astronautical observer; it is generally accepted that the dips are due to occultations of the central source by a thickened region of the accretion disk rim where the gas stream from the companion impacts upon the outer disk. However, the dips themselves also provide a wealth of physical information not limited to their role in defining the system geometry. From their irregular, ragged appearance and cycle-to-cycle differences in width and depth we know that the disk structure is dynamic. Also, spectral fitting results show that dips are associated with, but cannot be completely described by, increases in absorption of $N_H$ $\\approxgt$ 10$^{23}$~cm$^{-2}$.  \\source\\ is one of the most unusual members of this class; its persistent emission is the brightest (6$\\times$10$^{37}$~erg~s$^{-1}$), its dip profiles seemingly the most erratic, and its 21-hr orbital period is the longest -- $\\sim$5--25 times longer than other dipping sources, corresponding to a much greater stellar separation and a larger accretion disk radius. Dipping is deep, $\\sim$ 75\\% in the 1--10~keV band, and can at times reach a stable lower level (Watson et al.\\ 1985; Church \\& Ba\\l uci\\'nska-Church 1995) suggesting that one spectral component is entirely removed. The source also exhibits strong flaring in which the X-ray flux can increase by 30\\% or more over timescales of a few thousand seconds. In this regard, it may provide an interesting overlap between the dipping LMXBs, generally assumed to be atoll sources by virtue of their persistent emission level and noise characteristics, and the Z-track sources with their well-known flaring branches (e.g. Hasinger \\& van der Klis 1989).  The depth, duration, and spectral evolution in dipping vary considerably between members of the dipping class. An additional complication is the fact that several sources exhibit an unabsorbed component of the spectrum, e.g. XB1916$-$053 (Smale et al.\\ 1988, 1992), XB0748$-$676 (Parmar et al.\\ 1986) and XB1254$-$690 (Courvoisier et al.\\ 1986). More recently, it has been demonstrated that a unified approach describes the dipping sources (Church \\& Ba\\l uci\\'nska-Church 1995; Church et al.\\ 1997, Ba\\l uci\\'nska-Church et al.\\ 1999).  Their model consists of point-like blackbody emission from the neutron star, plus extended Comptonized emission from the Accretion Disk Corona (ADC), and a ``progressive covering'' description of absorption. In this description, the absorber on the outer edge of the accretion disk (White \\& Swank 1982) moves progressively across the emission regions, so that at any stage of dipping, part of the Comptonized emission is absorbed and part is not, giving rise to the observed unabsorbed component. As a point source, the blackbody is rapidly absorbed once the envelope of the absorber reaches the neutron star.  In previous observations of \\source\\ with {\\sl EXOSAT} and {\\sl Ginga}, a blackbody plus bremsstrahlung model was used to parameterize the spectra (Jones \\& Watson 1989). Subsequently, Church \\& Ba\\l uci\\'nska-Church (1995) showed that the light curve at energies $>$5~keV was dominated by flaring, which can strongly modify the spectrum and make the spectral investigation of dipping difficult. By selecting sections of non-dip and dip data without apparent flaring, the above two-component model was found to fit the data well, showing that in deep dipping the blackbody component was totally absorbed, and the Comptonized component relatively little absorbed. However, with only one non-dip and one deep dip spectrum it was not possible to determine the extent of absorption of the Comptonized component.  A more recent {\\sl BeppoSAX} observation revealed the dust scattered halo of the source demonstrated by Angelini et al.\\ (1997). Modeling of the radial distribution of count rate in the MECS at a series of energies below 5~keV allowed an optical depth to dust scattering of $\\rm {2.4\\pm 0.4}$ at 1~keV to be derived, the data being consistent with a cross-section varying as $E^{-2}$ (Ba\\l uci\\'nska-Church et al.\\ 2000a).  Spectral models of non-dip and dip data included the effects of this interstellar dust scattering, and good fits were obtained with the above two-component model, in which the blackbody had temperature $kT$ = 1.31$\\pm$0.07~keV, and the Comptonized emission was approximated by a cut-off power law with photon index of $\\Gamma$ = 2.0$^{+0.5}_{-0.7}$ and cut-off energy $\\sim$12 keV.  While more sensitive than previous studies, the {\\sl BeppoSAX} observation also included only a single dip episode.  Previous observations of \\source\\ have been limited either by the short duration of the observations or by the signal-to-noise ratio of the data; a thorough investigation requires a baseline of several orbits, coupled with a collecting area and efficiency large enough to study the spectral variability on short timescales. In this paper, we present light curves obtained with {\\sl RXTE} covering 4.5 contiguous orbits of \\source, during which both strong dipping and strong flaring took place. We derive an accurate system ephemeris, and perform a detailed study of the spectral variability in dips.  A companion paper will contain our analysis of the evolution and physics of the flares (Ba\\l uci\\'nska-Church et al.\\ 2000b, hereafter Paper II).   \\begin{figure}[htb] \\figurenum{1} \\begin{center} \\begin{tabular}{c} \\psfig{figure=f1.ps,width=8cm,angle=-90} \\end{tabular} \\caption{The X-ray light curve of \\source, 1999 September. The energy range covered is 2--20~keV, and the time binning is 64 sec. } \\end{center} \\end{figure} ",
        "conclusions": "We have shown that spectral evolution during dipping in \\source\\ is well-described by a combination of the point-source blackbody plus extended Comptonization emission model, with a ``progressive covering'' treatment of the absorption process. To this basic model it was necessary to add a broad Gaussian line at $\\sim $6.4~keV, and terms to describe interstellar dust scattering both out of and into the line-of-sight.  The blackbody temperature $kT$, and power law index $\\Gamma$, and cut-off energy $E_{CO}$ from these {\\sl RXTE} fits (Table 2) are in good agreement with the values determined from the {\\sl BeppoSAX} fitting, of $kT$ = $\\rm {1.37\\pm 0.07}$ keV, $\\Gamma $ = $\\rm {2.0^{+0.5}_{-0.7}}$ and $E_{CO}$ $\\sim $12 keV (Ba\\l uci\\'nska-Church et al.\\ 2000a).  During the {\\sl RXTE} observations, the total luminosity in the 1--30~keV band was $\\rm {1.47\\times 10^{38}}$ erg s$^{-1}$ at a distance of 15 kpc (Christian \\& Swank, 1997), approximately twice that seen during the {\\sl BeppoSAX} observation, and thus complete consistency of emission parameters should not necessarily be expected. The bolometric blackbody luminosity was $\\rm {2.3\\times 10^{37}}$ erg s$^{-1}$, or 15\\% of the total luminosity, smaller than the 36\\% measured by {\\sl BeppoSAX} (Ba\\l uci\\'nska-Church et al.\\ 2000a). Luminosity considerations are discussed further in Paper II.  In dipping, the blackbody component is rapidly and completely removed once the neutron star is covered.  The clear occurrence of saturated dipping in most of the dips observed indicates that in addition to the removal of the blackbody, absorption of the Comptonized emission also reaches a stable level. This must be due either to due to an absorber of smaller angular extent than the ADC by a fixed fraction, or to the presence of blobs within the absorber envelope such that $\\sim $20\\% of the absorber is transmitting in less dense regions in between blobs of higher density. The Gaussian line varied in dipping in a way that could be described by giving it the same covering factor as the ADC. There are two contributions to the excess at low energies: the dust-scattered halo components which are shown separately in Figure 6, and also the part of the extended Comptonized emission that is not covered at any level of dipping. Table 3 shows the covering fraction rising to a maximum value of $\\sim $80\\% in the best-fit model, unlike other dipping sources such as XB1916$-$053 (Church et al.\\ 1997) where dipping reaches a depth of 100\\% at all energies below 10 keV and the covering fraction similarly reaches 100\\%, proving that the angular extent of all emission regions is less than that of the absorber. In the present case, there may be two reasons why this does not happen: either the absorber is somewhat smaller in angular extent than the ADC extended emission, or more likely, the absorber in the outer accretion disk is not uniformly dense within its envelope, but blobby, as indicated by the strong variability within dipping. We investigate this possibility further below.  We can obtain further information on the ADC from the duration of dip ingress and egress. The sharp ingress to deep dipping, and the fast variability within dipping, take place on timescales of $\\sim $32~s, and this is clearly associated with absorption of the point-source blackbody component.  However, the long, shallow shoulders of the dips map a gradual process of dip ingress and egress which is consistent with the extended ADC being covered by relatively less dense absorber before the absorber envelope reaches and obscures the neutron star. The duration of the ingress shoulder is $\\rm {\\sim 12\\pm 2}$ ks from the present observation, and we can also derive a duration of $\\rm {\\sim 13\\pm 2}$ ks from studying the EXOSAT light curve (Church \\& Ba\\l uci\\'nska-Church 1995).  Some asymmetry can be seen in the shoulders, and we use the shorter ingress time corresponding to the leading edge of the bulge on the accretion disk, not the trailing edge, which would give a longer shoulder depending on the absorber dimensions rather than the dimensions of the ADC.  For an absorber of larger angular extent than the ADC, the ingress time $\\Delta t$ is the time taken by the leading edge to cross the ADC given by the velocity of material in the outer disk: \\[\\rm {2\\,\\pi\\, r_{disk}/P \\;=\\; d_{ADC}/\\Delta t,}\\]  \\noindent where $\\rm {r_{disk}}$ is the disk radius, P is the orbital period and $\\rm {d_{ADC}}$ is the diameter of the ADC. In the event that the absorber has somewhat smaller angular extent than the ADC, as is a possible interpretation of spectral fitting, the ingress time gives the diameter of the absorber, and the source region could be as much as 20\\% larger. Using a period of 20.87 hr from our ephemeris work above, and a mass of $\\rm {1.4 M_{\\sun}}$, a value of $\\rm {1.0\\times 10^{11}}$ cm can be calculated for the accretion disk radius (Frank et al.\\ 1987).  Using our observed $\\Delta t$, we derive from this an ADC radius of $\\rm {5.0\\times 10^{10}}$ cm. The average ingress time from the EXOSAT and {\\sl RXTE} data gives $\\rm {5.3\\times 10^{10}}$ cm. If the absorber has smaller angular size than the ADC, then the above calculation provides the absorber radius, so that the ADC could be up to 20\\% larger than this, on the basis of the covering fraction only reaching $\\sim $80\\% in deepest dipping.  We can use the above radius of the ADC to make some comparisons with simple ADC theory. The maximum radius of an ADC that can be supported in hydrostatic equilibrium $\\rm {r_{eq}}$ is given by the condition that kT $<$ $\\rm {GM_xm/r_{eq}}$ where $\\rm {M_x}$ is the mass of the neutron star and m the mass of a proton.  This provides a simple formula: \\[\\rm {r_{eq} \\simeq {1.6\\times 10^{11}\\; M_x\\over T_{ADC}\\; M_{\\sun}}\\;\\; (cm) }\\] where $\\rm {T_{ADC}}$ is the temperature of the ADC in units of 10$^7$ K. We can use the Comptonization cut-off energy $E_{CO}$\\ obtained in the present analysis of 12 keV to place limits on the electron temperature $kT_e$ of 4--12 keV, corresponding to an optical depth to electron scattering $\\tau <$1 or $\\tau >$1 respectively. For low optical depth, $kT_e \\sim E_{CO}$, but for higher optical depth, even if Comptonization is not saturated, $kT_e$ will be a factor of 2.5--3.0 smaller than $E_{CO}$ (e.g.\\ Dove et al.\\ 1997).  Using these limits on electron temperature, we find $\\rm {r_{eq}}$ = $\\rm {1.6-5.3\\times 10^{10}}$~cm, the smaller value corresponding to the higher temperature. Comparing these with $\\rm {r_{ADC}}$ $\\sim $ $\\rm {5\\times 10^{10}}$ cm from our dip ingress calculation, we can see that there is better agreement with $\\rm {r_{eq}}$ calculated assuming a higher optical depth to electron scattering. We thus have evidence that the ADC is maintained in hydrostatic equilibrium. The ADC is likely to be ``thin'', with a height-to-radius ratio of the order of 10\\% at its outer edge, as opposed to being spherical. For example, for an ADC radius of $\\rm {5\\times 10^{10}}$ cm, it is unlikely that the absorber on the outer disk would be able to cover a spherical ADC extending to this distance in the vertical direction."
    },
    "0010/astro-ph0010383_arXiv.txt": {
        "abstract": "Real physical systems are often maintained off equilibrium by energy or matter flows. If these systems are far from equilibrium then the thermodynamical branch become unstable and fluctuations can lead them to other more stable states. These new states are often endowed with higher degrees of organization. In order to explore whether an energy-flow in combination with self-gravity can lead to complex, inhomogeneous structures, like observed in the interstellar medium (ISM), we perform $N$-body simulations of self-gravitating systems subjected to an energy-flow.  Moreover we perform some simple ``gravo-thermal'' N-body experiments and compare them with theoretical results. We find negative specific heat in an energy range as predicted by Follana \\& Laliena (1999). ",
        "introduction": "An energy-flow through a system is related to an entropy-flow. If the entropy-flow leaving the system is larger than the entering one, the system evacuates its internal, by irreversible processes produced entropy to the outer world. Consequently order is created inside the system, diverging from the classical thermodynamical equilibrium of a closed system. Far from equilibrium the system is no longer characterized by an extremum principle, thus losing its stability. Therefore perturbations can lead to long range order, through which the system acts as a whole. Such a behavior is well known in laboratory hydrodynamics and chemistry. The underlying concepts such as ``dissipative structures'' and ``self-organization'' were extensively studied (see e.g. Nicolis \\& Prigogine 1977). But in spite of their popularity they are until now only little studied in the context of non-equilibrium structures in self-gravitating astrophysical systems. Therefore we take up some ideas of these concepts and build a simple model of an open self-gravitating system. With this model we want to check if an energy-flow can maintain a self-gravitating system in an statistically stable state, out of thermodynamical equilibrium and if gravitation in combination with an energy flow can create structures with a higher degree of order. ",
        "conclusions": ""
    },
    "0010/astro-ph0010660_arXiv.txt": {
        "abstract": "{The luminous infrared-loud quasar \\object{IRAS~13349+2438} was observed with the {\\it XMM-Newton} Observatory as part of the Performance Verification program.  The spectrum obtained by the Reflection Grating Spectrometer (RGS) exhibits broad ($v \\sim 1400 ~\\rm{km~s}^{-1}$ $FW\\!H\\!M$) absorption lines from highly ionized elements including hydrogen- and helium-like carbon, nitrogen, oxygen, and neon, and several iron L-shell ions (\\ion{Fe}{xvii -- xx}).  Also shown in the spectrum is the first astrophysical detection of a broad absorption feature around $\\lambda = 16 - 17$ \\AA\\ identified as an unresolved transition array (UTA) of $2p - 3d$ inner-shell absorption by iron M-shell ions in a much cooler medium; a feature that might be misidentified as an \\ion{O}{vii} edge when observed with moderate resolution spectrometers.  No absorption edges are clearly detected in the spectrum.  We demonstrate that the RGS spectrum of \\object{IRAS~13349+2438} exhibits absorption lines from at least two distinct regions, one of which is tentatively associated with the medium that produces the optical/UV reddening. ",
        "introduction": "\\object{IRAS~13349+2438} is an archetypal highly-polarized radio-quiet quasar at a redshift of $z = 0.10764$ (Kim et al.\\ \\cite{kim95}).  Since its identification as an infrared-luminous quasar (Beichman et al.\\ \\cite{beichman86}), this source has been extensively studied in the optical, infrared, and X-ray bands.  In a detailed investigation of the optical and infrared spectra and polarization measurements, Wills et al.\\ (\\cite{wills92}) demonstrated that the nuclear spectrum exhibits two distinct components; a highly-reddened component and a highly-polarized component that suffers much lower extinction.  Based on these observational facts, Wills (\\cite{wills92}) constructed a simple and elegant picture of the nuclear region of \\object{IRAS~13349+2438} in which the direct AGN radiation is attenutated by a thick dusty torus, while the observed highly-polarized light is produced by scattering in an extended bipolar region, either by warm electrons or by small dust grains.  \\object{IRAS~13349+2438} was detected in the {\\it ROSAT} All-Sky-Survey (Walter \\& Fink \\cite{walter93}; Brinkmann \\& Siebert \\cite{brinkmann94}), and has been the target of extensive pointed PSPC observations with {\\it ROSAT} (Brandt, Fabian, \\& Pounds \\cite{brandt96}), and with {\\it ASCA} (Brinkmann et al.\\ \\cite{brinkmann96}; Brandt et al.\\ \\cite{brandt97}).  The soft X-ray spectrum obtained with the PSPC shows a lack of absorption by cold material indicated by the observed optical reddening, and suggests the presence of a warm, dusty medium along the line of sight (Brandt, Fabian, \\& Pounds \\cite{brandt96}).  In a more recent investigation of the complex X-ray properties of \\object{IRAS~13349+2438}, Siebert, Komossa, \\& Brinkmann (\\cite{siebert99}) self-consistently accounted for the effects of dust embedded in the warm absorbing medium, and concluded that single zone models, both with and without internal dust, do not provide adequate fits to the combined, {\\it ROSAT}, {\\it ASCA}, and optical data sets.  In particular, they find that a dust-free warm absorber model formally gives the best fit to the X-ray data, and conclude that the X-ray absorption and optical reddening must arise in spatially distinct regions.  However, owing to the moderate spectral-resolving-power capabilities of the available detectors on {\\it ROSAT} and {\\it ASCA}, and the likely cross-calibration uncertainties, the precise nature of the soft X-ray spectrum has remained controversial.  In this Letter, we present results from the first high-resolution X-ray observation of \\object{IRAS~13349+2438} with the {\\it XMM-Newton} Observatory.  The spectrum obtained with the Reflection Grating Spectrometer (RGS) shows a wealth of discrete spectral features, including the first astrophysical detection of inner-shell $2p -3d$ absorption by M-shell iron ions in the form of an unresolved transition array (UTA).  From a detailed analysis of the rich absorption spectrum, we measure the column density and velocity field of the line-of-sight material.  We show that the spectrum contains absorption features from regions with two distinct levels of ionization.  The column density of the lower ionization component is consistent with the observed optical reddening, and we tentatively associate this component with the dusty torus. ",
        "conclusions": ""
    },
    "0010/astro-ph0010456_arXiv.txt": {
        "abstract": "Molecular outflows from young protostars are widely believed to be collimated by magnetic fields, but there has been little observational evidence to support this hypothesis. Using the new technique of millimetre-wavelength spectropolarimetry, we demonstrate the existence of a magnetic field in the NGC2024-FIR5 outflow lobe. The 1.3 mm J=2--1 transition of carbon monoxide (CO) is polarized at a level of approximately 1 \\%, in a direction within 10--15$^{\\circ}$ of the outflow axis.  This agrees with theoretical models where the magnetic field channels the outflowing gas, and shows that the process can be effective as far as 0.1 pc from the protostar. ",
        "introduction": "Many models predict that the outflows from young protostars are magnetically collimated (e.g. Uchida \\& Shibata 1985; Pudritz 1986; Shu et al. 1994;  Fiege \\& Henricksen 1996). The collimation may take place within a few stellar radii, or be maintained at much larger distances, for example by alignment of the magnetic field with streamlines in the extended protostellar envelope. These theories have previously been very difficult to test, as the outflow emission is weak compared to that of the ambient cloud. Detection of the Zeeman effect in molecular lines is thus very difficult, and the more sensitive technique of mapping of polarized thermal dust emission (e.g. Hildebrand et al. 1996) cannot be used, as there is no velocity information to separate the envelope and outflow.  This problem can now be addressed with spectropolarimetry in the millimetre-wave regime, a recently developed technique (e.g. Greaves et al. 1999) utilising the Goldreich-Kylafis effect. In the presence of a magnetic field, Zeeman splitting of the rotational levels of molecules occurs, and rotational transitions can then be polarized, due to imbalances in the magnetic sublevel populations (Goldreich \\& Kylafis 1981). The percentage polarization is sensitive to optical depth and excitation conditions, since both polarizations saturate if $\\tau \\gg 1$, and the sub-level populations equalise if collisional transition rates significantly exceed radiative rates. However, with suitably chosen molecular lines, spectropolarimetry can be used to map magnetic fields {\\it as a function of velocity}. There is an ambiguity in the field direction, since the polarization direction may be either parallel or perpendicular to the net field in the plane of the sky, but this can be resolved for simple velocity structures (Kylafis 1983).  Detections of magnetic fields using this technique have previously been published by Glenn et al. (1997) for the envelope of the evolved star IRC+10216, and by Greaves et al. (1999) for two molecular clouds and the `2 pc ring' around the Galactic Centre. These are all bright sources and the deepest observations so far are those of Girart, Crutcher \\& Rao (1999), who used the BIMA array to map the environment of the NGC1333-IRAS4A protostellar system. Surprisingly, they did not find a field clearly aligned with the outflow axis, but did detect strong polarization (2--10\\%) in a region within the flow where the magnetic field appears to be deflecting the gas.  In this {\\it Letter}, we present results of a similar experiment for the molecular outflow from the vicinity of the massive young source NGC2024-FIR5. This outflow has been extensively mapped (e.g. Richer et al. 1992) and extends several arcmin to the south, but has apparently no northern lobe. We have used the James Clerk Maxwell Telescope to search for polarized CO J=2--1 rotational emission (at 1.3 mm wavelength) in the bright, highly collimated, red-shifted lobe. Polarization was detected at a low level ($\\approx$ 1~\\%), and the results are discussed in the context of the outflow structure. ",
        "conclusions": "The observed polarization direction is a direct diagnostic of the orientation of the magnetic field in the plane of the sky. Kylafis (1983) has shown that the polarization will be either perpendicular or parallel to the field direction, even for quite low field strengths (B $\\sim$ 1 $\\mu$G). Further, if the outflow is assumed to have a one-dimensional velocity field, it is possible to distinguish between these two orientations. For large $\\tau$(CO), the polarization will generally be seen parallel to the field, unless the field is at a large angle to the velocity gradient (Kylafis 1983). The criterion for the field and polarization to be parallel is cos$^2 \\alpha > 1/3$ ($\\alpha < 54.7^{\\circ}$), where $\\alpha$ is the angle between the velocity gradient and the field.  For NGC2024, Richer et al. (1992) have found that the flow is accelerated, with the fastest moving material further from the star, so we assume that the dominant velocity gradient is along the outflow axis. Then the most probable interpretation of the data is that the magnetic field lies close to the flow axis, i.e. $\\alpha \\sim 0^{\\circ}$, and the polarization will be produced near the flow axis, as observed.  It is unlikely that the p vectors are perpendicular to the field, as then the field would lie across the flow axis ($\\alpha \\sim 90^{\\circ}$), and this would impede the flow by channeling ionized molecules sideways.  Magnetic channeling of the outflow is not implausible on energy grounds. For a gas cylinder, the magnetic energy is given by $B^2 R^2 / 8$ for a unit length (e.g. Chandrasekhar \\& Fermi 1953), and the outflow kinetic energy will be $M v^2 / 2$. Adopting an outflow radius (cf. Figure~1) of $R \\sim 0.03$~pc or $1 \\times 10^{17}$~cm, the mass per unit length is $M \\sim 3 \\times 10^{14}$ g, and a lower limit on the bulk outflow speed is $v \\sim 10$~km s$^{-1}$ (cf. Figure~2). This gives a required magnetic field strength, for the two energies to be equivalent, of $\\geq 350 \\mu{\\rm G}$. However, since the magnetic field affects only {\\it orthogonal} flow, then the energy required is reduced when the field and flow are nearly aligned. Adopting an outflow opening angle of $\\pm 5-10^{\\circ}$ (cf. Fig.~ 4 of Richer et al. 1992), the cross-wise velocity is only 1--2~km~s$^{-1}$ and B $\\geq 35-70 \\mu{\\rm G}$ should provide sufficient energy to keep the flow well collimated. This is not excessive, given that Crutcher et al. (1999) have measured a lower limit to the line-of-sight field component of 80 $\\mu{\\rm G}$, in the main cloud emission near our North point.  In summary, the spectropolarimetry observations have shown that the outflow gas is in fact magnetized, and that the net field direction is aligned within a few degrees of the flow axis, as predicted in most models. If the magnetic field is channeling the outflowing gas, this process must be effective at quite large distances from the star (1 arcmin, or about 0.1 pc at a distance of 415 pc). We see little sign of `magnetic deflections' as in the NGC1333-IRAS4A outflow observed in CO J=2--1 spectropolarimetry by Girart et al (1999), using the BIMA array with a $9 \\times 6''$ synthesized beam. The NGC2024 outflow is relatively straight, although the small differences in our North and South polarization directions do appear to follow slight bends in the flow at the limits of the errors."
    },
    "0010/astro-ph0010504.txt": {
        "abstract": "We derive new stability criteria for purely MHD instabilities in rotating jets, in the framework of the ballooning ordering expansion. Quite unexpectedly, they involve a term which is linear in the magnetic shear. This implies that cylindrical configurations can be destabilized by a negative magnetic shear as well as by a favorable equilibrium pressure gradient, in distinction with the predictions of Suydam's stability criterion, which suggests on the contrary that the shear is always stabilizing.  We have used these criteria to establish sufficient conditions for instability. In particular, the magnetic shear can always destabilize jets with vanishing current density on the axis, a feature which is generically found in jets which are launched from an accretion disk. We also show that standard nonrotating jet models (where the toroidal field dominates the poloidal one), which are known to be unstable, are not stabilized by rotation, unless the plasma $\\beta$ parameter and the strength of the rotation forces are both close to the limit allowed by the condition of radial equilibrium.  The new magnetic shear-driven instability found in this paper, as well as the more conventional pressure-driven instability, might provide us with a potential energy source for the particle acceleration mechanisms underlying the high energy emission which takes place in the interior of AGN jets. ",
        "introduction": "Jets are commonly observed in connection with Active Galactic Nuclei (AGNs) and Young Stellar Objects (YSOs). These jets are cylindrically collimated over remarkably long distances in comparison with their radial extents; the confinement provided by the magnetic tension of the toroidal field in current carrying jets has long been recognized as an efficient way to achieve this type of collimation (as shown, e.g., by Chan and Henriksen 1981, Blandford and Payne 1982, after an initial suggestion by Benford 1978). However, it is well-known from the field of thermonuclear fusion that toroidally confined plasma configurations are generically unstable, and instabilities in this context are able to destroy these configurations on very short time-scales (for a general background on MHD instabilities, in particular in the context of thermonuclear fusion, see Bateman 1978, and  Freidberg 1987). Furthermore, jets are subject to other instabilities, most notably the Kelvin-Helmholtz one, which pose similar threats to their survival. On the other hand, there is ample evidence of instability in the observed jets, both directly (lateral displacements of the jet beam, bright knots...) and indirectly (models of jet synchrotron emission rely on particle acceleration through e.g. shock waves or turbulence), but the observed jet survival implies that (for reasons which are still unclear), instabilities in real jets lead to internal reorganisation and turbulence rather than to disruption.  Most of the literature on jet stability has focused on the Kelvin-Helmholtz instability (see, e.g., Birkinshaw 1991 and references therein), both for hydrodynamical and MHD jets, and for a variety of jet equilibria. Two types of modes are produced by this instability: ordinary surface modes and reflected body modes. Surface waves are effectively confined to the jet interface with the external medium, while body modes appear and become dominant only for beam velocities in excess of the fast magnetosonic velocity (up to a factor of order unity). Modes growth rates decrease when the Mach number increases. A longitudinal magnetic field has a stabilizing influence due to magnetic tension; in particular, sub-Alfv\\'enic (up to a factor of order unity) flows are completely stabilized. Radiative effects can either enhance or reduce the growth rate of the instability depending on the steepness of the temperature dependence of the cooling function; in the presence of radiative cooling, the surface waves are apparently the most dangerous for jet survival (Hardee and Stone 1997; Stone {\\it el al.} 1997).  Besides the Kelvin-Helmholtz instability, driven by velocity gradients, jets can be unstable with respect to purely MHD processes. In the context of ideal MHD, these instabilities are usually divided into pressure-driven and current-driven instabilities (see e.g. Freidberg 1987 for details). Pressure-driven instabilities are related to gradients of the equilibrium pressure and to magnetic field line curvature. The excited modes include the so-called saussage mode, and are usually divided into interchange modes and ballooning modes; they share some common features with the Rayleigh-Taylor and Parker instabilities. Current-driven modes originate in the current parallel to the magnetic field, and include the so-called kink instabilities; the most dangerous is usually the $m=1$ kink mode ($m$ being the azimuthal wave-number).  Much less attention has been devoted to the analysis of these MHD instabilities, most probably because in superfast jets (whose beam velocity exceeds the fast magnetosonic one), the kinetic energy exceeds the magnetic energy, and the Kelvin-Helmholtz instability is expected to dominate. Indeed, this expectation is borne out both in superfast and transmagnetosonic jets (superalfvenic, but subfast beam velocities) for current-driven instabilities, whose growth rates are always substantially smaller than for the Kelvin-Helmholtz instability, at least for force-free jet equilibrium configurations (Appl and Camenzind 1992; Appl 1996). These authors also show that the jet current is stabilizing for supermagnetosonic flows, whereas it is destabilizing for transmagnetosonic ones.  This paper focuses on pressure- and magnetic shear-driven instabilities, in the framework of the ballooning ordering expansion. Pressure instabilities have been most actively studied in fusion research (Kadomtsev 1966; Coppi {\\it et al.} 1979; Dewar and Glasser 1983; see Freidberg 1987, and references therein). In astrophysics, they have been considered in the context of solar (e.g., Hood, 1986) and magnetospheric physics (see Ferri\\`ere {\\it et al.} 1999 for a recent and synthetic overview of the situation in this field). Although these instabilities have been virtually ignored in the context of MHD jets (with the exception of Begelman 1998), it is important to characterize their conditions onset there; indeed, in spite of their local nature, they are able to produce large-scale disruptions over a dynamical time-scale in fusion devices. Furthermore, our analysis is also motivated by the following considerations. Small-scale Kelvin-Helmholtz and current-driven instabilities are stabilized by magnetic tension. Furthermore, the Kelvin-Helmholtz instability in the transmagnetosonic regime is confined to the jet boundary. These features don't make them favorable energy sources for the high energy emission which takes place in jet interiors. By contrast, the pressure- and magnetic shear-driven instabilities considered here usually have a small-scale component transverse to the magnetic field lines, and are expected to occur in the type of magnetic configurations which are typically considered for jet inner regions.  In this paper, we clarify somewhat the relation between the various classical stability criteria found in the literature, and we assess the role of the magnetic shear and of the rotation of the jet on the onset of jet instability, in particular in jet interiors. In the process, we point out the possible and unexpected destabilizing role of the magnetic shear. This paper is organized as follows. In Section 2, we recall the MHD equations and introduce our notations; we also briefly recall the origin of the ballooning ordering expansion, and present the perturbation equations in this approximation (their derivation is performed in the Appendix); finally, we introduce the type of modes we analyse in this paper (which are interchange and not ballooning modes) and specialize the ballooning equations to cylindrical systems. In Section 3, we focus on the mode dispersion relation and discuss its properties for non-rotating and rotating jets. Section 4 discusses our results and concludes this paper. ",
        "conclusions": "In this paper, we have focused on the stability of rotating MHD jets with respect to inhomogeneities connected to pressure gradients and magnetic shear, in the framework of the ballooning ordering. The most remarkable finding of our analysis is the possible destabilizing influence of the magnetic shear pointed out in section 3. Indeed, the shear appears quadratically in Suydam's well-known criterion, which applies to the stability of cylindrical columns, whereas the stability criterion of the preceding section is linear in magnetic shear. We point out that, had we not relaxed the eikonal condition in section 2, we would have recovered Suydam's criterion (see Freidberg 1987, Chapter 10). This unexpected behavior of the magnetic shear implies that the inner regions of jets launched from accretion disks are generically unstable, as argued in section 3.  We have also recovered that for jet models dominated by a toroidal magnetic field, jet interiors are likely to be subject to this type of instability, a result already pointed out by Begelman (1998). Furthermore, we find that rotation has little influence on this conclusion; rotation can only stabilize jets when it is close to the limit allowed by the radial equilibrium condition, and for values of $\\beta$ close to the same limit.  Our results on pressure-driven instabilities agree with those found in the literature, most notably Begelman (1998) and Ferri\\`ere {\\it et al.} (1999), once appropriate limits are taken. Begelman's study is restricted to $B_\\theta/B_z \\gg 1$, which removes the effect of the magnetic shear from the analysis. Ferri\\`ere {\\it et. al.} (1999) do not rely on the ballooning expansion, keep the full six-dimensional dispersion relation, and discuss in detail the effects of rotation and gravity, besides gas pressure and field curvarture. They give a detailed and interesting synthesis of gravitational and centrifugal instabilities in magnetospheric studies. However, to achieve such a generality, they rely on a particular coordinate representation of the magnetic field, which is well-adapted to planetary magnetospheres, but forbids also the presence of magnetic shear.  Jet models which are not dominated by the toroidal magnetic field have been proposed in the literature. One example is the force-free jet model of Appl and Camenzind, 1992, which has $B_\\theta\\sim B_z$, and belongs to a class of configurations known as ``reversed field pinch\" in the fusion research context. These configurations are known to be stable with respect to pressure-driven instabilities for moderately low values of $\\beta$ (and, more generally, to possess good stability properties). This apparently conflicts with our results on destabilization by the shear. However, we point out that destabilization occurs because $B_\\theta$ varies faster than $r$ in the jet inner region. This variation is induced by the behavior of the jet matter and field rotation, a feature which is absent from fusion control devices.  In this analysis as in all others we are aware of, we have assumed that the jet equilibrium structure is invariant along the jet axis. This need not be the case, as examplified in various self-similar solutions (Sauty and Tsinganos, 1994;  Contopoulos and Lovelace, 1994), and more recenly in a complete analytic solution displaying a more complex magnetic field structure (Bogoyalenskij, 2000). The relevance of the available stability analyses to such configurations is unclear; furthermore, these configurations are possibly unstable with respect to appropriately chosen ballooning modes. In any case, this question should be addressed before firm conclusions can be drawn.  We have not tried to derive precise growth rates, as these depend on the radial structure of the most unstable modes, the determination of which is outside the scope of the approximation scheme we have adopted. However, order of magnitude considerations show that these instabilities should develop within a few dynamical time-scale ($\\sim a V_A^{-1}$ where $a$ is the characteristic radial size of the considered region).  Let us also return to the question of overstability, which was briefly alluded to in section 2 and 3. Our dispersion relation shows that $\\omega^2$ is real for alfvenic modes in the $\\kpa=0$ limit, implying that $\\omega$ is real when $\\omega^2>0$. This property remains true when $\\omega(\\kpa)$ is computed as a series in $\\kpa$, to all orders in this series if it is true to lowest order (this can be shown by induction from the full dispersion relation, because it involves only real terms). Therefore the stability property of the Alfv\\'en mode is obtained in the $\\kpa=0$ limit. In the same way, the dispersion relation of Ferri\\`ere {\\it et. al.} (1999) [their Eq.~(27)] implies that the fast magnetosonic mode is stable in the absence of magnetic shear in the perpendicular WKB approximation. This suggests that the fast mode is probably not destabilized in the asymptotic limit examined in this paper.  We point out again that, in spite of their partly local nature, pressure-driven instabilities, and magnetic shear-driven ones as well, are likely to produce large-scale disruptions of MHD jets. How real jets survive these instabilities as well as Kelvin-Helmholtz and current-driven instabilities is a still open and pressing problem, especially that 3D instabilities seem more vigorous than there 2D counterparts (see, e.g., Bodo {\\it et al.}, 1998). However, progress on this topic has recently been accomplished. For the Kelvin-Helmholtz instability, recent 3D simulations have shown that for initial magnetic Reynolds numbers smaller than $\\sim 50$, the shear layer broadens, the turbulence level decreases and the flow becomes quasi-laminar and stable in the end (Ryu {\\it et al.}, 2000); as jets are not expected to be drastically out of equipartition, and as the Lorentz force provides the origin of their acceleration, this probably ensures their stability against a possible disruption by the Kelvin-Helmholtz instability. Also, the kink instability seems to lead to helicoidal equilibria with a redistribution of the current density rather than to disruption (Lery {\\it et al.}, 2000). Finally, it has recently been recognized in the fusion literature that pressure-driven instabilities can be stabilized by large equilibrium gradients, a somewhat counterintuitive result (Terry, 2000).  One of the interesting questions raised by this work concerns the possible link between the type of instabilities we consider and the particle acceleration mechanism which is responsible for the high energy emission seen in AGN jet interiors. For example, in the context of the two-flow model, most of the high energy emission comes from a highly relativistic pair plasma occupying the inner region of a slower jet which contains most of the mass and provides most of the energy. This model relies on energy injection to the pair plasma from the bulk of the jet through turbulence (Renaud and Henri, 1998); note also that the existence of a pair plasma in AGN jets has recently received an interesting observational support (Wardle, {\\it et. al.}, 1998; Hirotani {\\it et. al.}, 1998). In such a picture of the high energy emission, the required turbulence would most likely be provided by MHD instabilities in the nonlinear regime, as other sources look unpromising. Indeed, a purely hydrodynamical origin in regions of increasing angular velocity with radius, as advocated by Richard and Zahn (1999) is not found in numerical simulations (Balbus {\\it et al.}, 1996), and happens in laboratory Couette experiments (performed in general with narrow gap widths between the inner and outer cylinder) only when the outer cylinder completely dominates the rotation (Coles, 1965), a situation which has little connection with disks and jets; furthermore, the magneto-rotational instability (Balbus and Hawley, 1991) does not work in this regime, and the velocity shear is ineffective in driving the Kelvin-Helmholtz instability in the subalfvenic portion of the jet in jet interiors.  To conclude, let us finally point out that pressure-driven modes are also known to be avalanche-like instabilities, due to the existence of a critical gradient, and this feature raises important question concerning the action of this instability on jet profiles, as well as their role in jet variability.  \\appendix"
    },
    "0010/astro-ph0010510_arXiv.txt": {
        "abstract": "\\noindent The newly discovered supernova remnant \\snr\\ (RX~J0852.0--0462), along the line of sight to the Vela SNR, was observed with ASCA for 120 ks. We find that the X-ray spectrum is featureless, and well described by a power law, extending to three the class of shell-type SNRs dominated by nonthermal X-ray emission. Like G347.3$-$0.5, this low-latitude remnant displays discrete regions of enhanced emission along the rim as well as faint nonthermal emission from the interior. We derive limits on the thermal content of the remnant emission, although the presence of the Vela SNR compromises our ability to seriously constrain a low temperature component. Limits placed on the amount of Sc-K emission are compared with the expected flux based upon the reported $^{44}$Ti emission from \\snr. We also report on an unresolved X-ray source surrounded by diffuse emission near the center of the remnant. The properties of the source are not well determined, but appear consistent with the interpretation that the source is a neutron star surrounded by a synchrotron nebula. Alternatively, the source may be associated with one of two stars located within the positional error circle, but this appears somewhat unlikely. ",
        "introduction": "Young supernova remnants (SNRs) are believed to be the prime accelerators of cosmic rays, at least up to the ``knee'' of the cosmic ray spectrum at energies of $\\sim 10^3$~TeV, beyond which the spectrum steepens. However, the direct evidence for particle acceleration to such energies is scarce. Radio emission from shell-type SNRs is the result of synchrotron radiation from shock accelerated electrons, but these electrons have typical energies of order 10~GeV. It is only the recent discoveries of nonthermal X-ray emission from shell-type SNRs that have finally provided evidence of particles at energies of 10--100~TeV. The X-ray emission from SN~1006 (Koyama et al. 1995) and G347.3$-$0.5 (Koyama et al. 1997, Slane et al. 1999) is {\\it predominantly} nonthermal, while that from Cas~A, Kepler, Tycho, and RCW 86 (Allen, Gotthelf, \\& Petre 1999) is predominantly thermal, but contains a nonthermal component as well. However, it may also be the X-ray emission that poses the most significant constraint on accepting SNRs as the source of cosmic rays at energies near the knee. The X-ray fluxes from all remnants fall below the extrapolation of their radio spectra, implying a steepening or cut-off of the electron spectrum at higher energies. The implied spectral cut-off energies appear to be well below the knee of the cosmic ray spectrum (Reynolds \\& Keohane 1999). It is thus of considerable importance to identify new examples of shell-type SNRs for which the X-ray emission is largely nonthermal.  RX~J0852.0-0462 was discovered by Aschenbach (1998) using data from the ROSAT All-Sky Survey. Situated along the line of sight to the Vela SNR, the emission stands out above the soft thermal emission from Vela only at energies above $\\sim 1$~keV. Data from the ROSAT PSPC indicates a hard spectrum with either a hot thermal component with $kT \\sim 2.5$~keV or a power law component with photon index $\\Gamma \\sim 2.6$ (Aschenbach 1998). The remnant is circular with a diameter of $\\sim 2^\\circ$; the emission is particularly enhanced along the northern, western, and southeastern limbs. Hereafter, we use the designation \\snr\\ for this newly identified SNR.  As first noted by Aschenbach (1998) the radio emission from \\snr\\ is quite weak ($S_\\nu = 30 - 50$ Jy at 1 GHz) with faint, limb-brightened emission similar to the X-ray morphology (Combi, Romero, \\& Benaglia 1999; Duncan \\& Green 2000). SN1006 and G347 are also faint radio emitters, perhaps suggesting a general characteristic of remnants for which the X-ray flux is dominated by nonthermal processes.  Iyudin et al. (1998) reported the COMPTEL detection of $^{44}$Ti from the source GRO~J0852-4642, which was tentatively associated with \\snr. If correct, this association would be profound; the very short $^{44}$Ti lifetime ($\\tau \\approx 90$~y) would imply a very young SNR, and the large observed size would require that the remnant be very nearby as well. Estimates based on the X-ray diameter and $\\gamma$-ray flux of $^{44}$Ti indicate an age of $\\sim 680$~y and a distance of $\\sim 200$~pc (Aschenbach, Iyudin, \\& Sch\\\"{o}nfelder 1999). The hard X-ray spectrum would seem to support this scenario as well. The inferred temperature would imply a rapid shock that is consistent with a young SNR. However, as we show here, the hard X-ray emission is not from hot, shock-heated gas; it is nonthermal. Further, separate reanalysis of the COMPTEL data finds that the detection of \\snr\\ as a $^{44}$Ti source is only significant at the $2 - 4 \\sigma$ confidence level (Sch\\\"onfelder et al.~2000). In the absence of such emission, and given that the X-ray emission is nonthermal, the nearby distance and young age may need to be reexamined.  \\centerline{\\psfig{file=f1.eps,width=8.8cm}} \\small\\parindent=3.5mm {\\sc Fig.}~1.--- ASCA GIS image of G266.2--1.2 ($E = 0.7 - 10$~keV). The image consists of a mosaic of 7 individual fields. Contours represent the outline of the Vela SNR as seen in ROSAT survey data with the PSPC. \\normalsize ",
        "conclusions": "The ASCA observations of \\snr\\ reveal that the soft X-ray emission from this SNR is dominated by nonthermal processes. This brings to three the number of SNRs in this class, and provides additional evidence for shock acceleration of cosmic rays in SNRs. Because of the bright and spatially varying background caused by the Vela SNR, limits on the thermal emission from \\snr\\ are difficult to establish. As a result, its evolutionary state is poorly constrained. However, the ASCA data reveal a larger column density for this remnant than for Vela, indicating that \\snr\\ is at a larger distance, and perhaps associated with the star formation region located at a distance of $1 - 2$~kpc. The very strong ratio of nonthermal to thermal X-ray emission argues for a low density environment, possibly suggesting evolution in a stellar wind cavity. Such a scenario would indicate a massive progenitor, from which one could expect a relic neutron star. The point source AX~J0851.9-4617.4 could represent this source, with the diffuse emission around the source being an associated synchrotron nebula, albeit a very faint one. Higher resolution X-ray observations of this source are of considerable importance in order to assess its relationship to the SNR.  We note that TeV $\\gamma$-ray emission has been detected from both SN~1006 and G347.3$-$0.5, confirming the presence of very energetic electrons in these remnants. Similar observations of \\snr\\ are clearly of considerable interest as well."
    },
    "0010/astro-ph0010276_arXiv.txt": {
        "abstract": "The strengths of magnetic fields in interstellar gas clouds are obtained through observations of the circular polarization of spectral line radiation. Irregularities in this magnetic field may be present due to turbulence, waves or perhaps other causes, and may play an essential role in the structure and evolution of the gas clouds. To infer information about these irregularities from the observational data, we develop statistical relationships between the rms values of the irregular component of the magnetic field and spatial variations in the circular polarization of the spectral line radiation. The irregularities are characterized in analogy with descriptions of turbulence---by a sum of Fourier waves having a power spectrum with a slope similar to that of Kolmogorov turbulence. For comparison, we also perform computations in which turbulent magnetic and velocity fields from representative MHD simulations by others are utilized. Although the effects of the variations about the mean value of the magnetic field along the path of a ray tend to cancel, a significant residual effect in the polarization of the emergent radiation remains for typical values of the relevant parameters. A map of observed spectra of the 21 cm line toward Orion A is analyzed and the results are compared with our calculations in order to infer the strength of the irregular component of the magnetic field. The rms of the irregular component is found to be comparable in magnitude to the mean magnetic field within the cloud. Hence, the turbulent and Alfven velocities should also be comparable. ",
        "introduction": "The breadths of spectral line radiation from interstellar gas clouds at radio wavelengths are recognized to be much greater than would be caused by Doppler shifts resulting from thermal motions alone at the temperatures of the atoms and molecules in the gas. These breadths are believed to be an indication of disturbances (most likely a form of turbulence or waves) which are fundamental to the structure and evolution of interstellar gas clouds and star formation (e.g., V\\'azquez-Semadeni et al. 2000, and reviews in Lada \\& Kylafis 1999 and in Franco \\& Carraminana 1999). Although the conclusions of earlier investigations that relate detailed information about the spectral lines specifically to the properties of turbulence have been questioned (see Dubinski, Narayan, \\& Phillips 1995), recent efforts along these lines are encouraging (Miesch \\& Scalo 1995; Lis et al. 1998). Of particular importance is the role of waves or turbulence in causing a magnetic pressure that acts parallel to the direction of the mean magnetic field lines in the clouds. This has long been viewed as a promising way to understand why the rate of collapse of interstellar clouds along the field lines is not more rapid than is indicated by the observational data. Magnetic disturbances have been thought to decay more slowly than ordinary disturbances, and hence more likely to persist and provide the component of the pressure that seems to be necessary (Arons \\& Max 1975). However, the premise that the magnetic disturbances are longer lived has been questioned in recent investigations (MacLow et al. 1998; Stone, Ostriker, \\& Gammie 1998; Padoan \\& Nordlund 1999). In any case, irregularities in the magnetic fields are expected to be associated with the bulk motions which are reflected by the non-thermal breadths of spectral lines. At some level, these should be evident in the observational data.  Information about the direction of the magnetic field in the interstellar gas has long been inferred from the linear polarization of starlight that results from selective absorption by dust grains. However, the polarizing power of the grains that absorb starlight apparently is greatly reduced in gas clouds of even modest densities (e.g., Goodman 1996). Though such polarization data are extensive, they do not thus seem to be sensitive indicators for the irregularities of the magnetic field in dense clouds. In contrast, there is no doubt that the magnetic field in interstellar clouds is evident in the circular polarization of spectral lines at radio frequencies through the Zeeman effect. Although such data are not so extensive as is desirable for extracting information on irregularities in the field, advances in the observational capabilities are continuing. It is thus useful to examine quantitatively the sensitivity of the circular polarization of spectral lines to plausible irregularities in the magnetic field in interstellar clouds. In particular, the polarization associated with a ray of radiation is a sum of contributions by the gas all along the path of the ray. The effects of deviations of the field from its mean value thus tend to cancel for the radiation that emerges from the cloud and to reduce its sensitivity as an indicator of irregularities. The goal of our investigation is to determine what information remains in the circular polarization of radiation that emerges from an irregular medium. Though not the emphasis of our investigation, an important challenge is to identify observational quantities for which the predictions distinguish between actual MHD phenomena and a simple generic description of the irregularities in the fields. Finally, we note that the emission of linearly polarized radiation has been observed from grains in dense clouds. This linearly polarized radiation holds promise to become a valuable tool (e.g., Hildebrand 1996; Rao et al. 1998) as more sensitive and extensive maps are obtained.  In Section 2, we describe how the irregularities are characterized and how the calculations of the circular polarization are performed. We focus on the quantity that commonly is obtained from the observational data---the single value $B_{\\rm f}$ of the magnetic field that best fits the circular polarization (Stokes $V$ intensity) when the Zeeman splitting is much smaller than the breadth of the spectral line. By comparing the statistical properties of the $B_{\\rm f}$ which are inferred from the observations with the properties of the $B_{\\rm f}$ which we calculate, we propose to gain information about the actual strength of the irregular component of the field within interstellar clouds. As an example, data for the 21 cm spectral line toward Orion A are analyzed in this manner in Section 3. ",
        "conclusions": "A number of deficiencies can be imagined in the simplifying idealizations of our calculations. Nevertheless, the conclusion appears to be firm that plausible turbulent magnetic fields are likely to have a significant effect on the magnetic field strengths that are inferred from the observation of the circular polarization of spectral lines at radio wavelengths. That is, their effect is unlikely to ``average out''. Clearly the exact ratio depends upon specific considerations outlined in the foregoing. One-third is a representative estimate for the ratio $(B_{\\rm f} )_{\\rm sd} /B_{\\rm rms} $ of the rms deviations of the observationally inferred field to the rms of the irregular (or turbulent) magnetic field that actually exists in the cloud. As expected, there is a tendency for deviations from the average to cancel when the contributions are summed along the paths of the rays of radiation. When the opacity function $f(s)$ is a function of location $s$ because it depends upon matter density or other considerations, $(B_{\\rm f} )_{\\rm sd} /B_{\\rm rms} $ will naturally tend to increase. This increase can be viewed as a result of effectively reducing the number of elements along the path that contribute which, in turn increases the scatter simply due to statistics. In MHD simulations, there should be some correlation between the matter density and the strength of the magnetic field. This also tends to increase $(B_{\\rm f} )_{\\rm sd} /B_{\\rm rms} $, though we suspect that the effect is small based on our examination of the correlation between density and field strength. The results for the strong field (S) case in Figure 1 might seem to be in conflict with this conclusion. We suspect, however, that the relatively larger values of $(B_{\\rm f} )_{\\rm sd} /B_{\\rm rms} $ in this case are due mainly to the ``statistical'' effect of greater contrasts due to the anisotropic variations in the density.  In a larger sense, the idealizations here are an incomplete description for star forming clouds because gravity is absent in these idealizations. Gravitational attractions will tend to cause clustering of matter, and hence more extreme variations in $f(s)$ than considered in our calculations. In contrast, the results when $f(s)$ is constant in Figures 1 and 2 suggest that purely statistical sampling provides velocity and magnetic fields that may be adequate to understand the circular polarization of clouds where gravity is not a major effect and when the finite resolution of the telescope is considered. This is not an altogether happy conclusion since it potentially limits our prospects for identifying the physical phenomena for which the non-thermal line breadths and variations in the magnetic fields are indicators.  Our interpretation that the irregular component of the magnetic field in the Orion A cloud is similar in magnitude to the average magnetic field ($B_{\\rm rms}= \\left|\\left\\langle B_{\\rm f}^{\\rm obs}\\right \\rangle\\right|$) implies that the magnetic field is quite disordered in this cloud. That the bulk motions of the gas are able to deform the field to this degree tends to indicate that the turbulent kinetic energy is at least comparable with the energy in the magnetic field, and thus that the turbulent velocity is at least comparable with the Alfven velocity. In numerical computations of MHD turbulence (including the MHD fields that we are using), the energy of the irregular component of the magnetic field tends to be similar to the kinetic energy of the turbulence (typically 30 to 60\\%; V\\'azquez-Semadeni et al. 2000). For Alfven waves, these two components of the energy are equal. From this, it also follows that the turbulent and Alfven velocities should be comparable when $B_{\\rm rms}= \\left|\\left\\langle B_{\\rm f}^{\\rm obs}\\right\\rangle\\right|$. Padoan \\& Nordlund(1999) have emphasized that MHD computations in which the turbulent velocities are super-Alfvenic lead to predicted features for molecular clouds that are different in important ways from computations with lower turbulent velocities. In view of the uncertainties in our present analysis for $B_{\\rm rms}$, we are unable to say whether the turbulent velocities are somewhat larger or somewhat smaller than the Alfven velocity. Alternatively, the ratio of the turbulent and Alfven velocities could be obtained from the spectral linebreadth in Figure 3 and the average field strength in equation (14) if the mass density were known reliably. However, as discussed in Section 3, the mass density is poorly known. A mass density of about $10^4$ H-atoms cm$^{-3}$ is required to yield an Alfven velocity that is equal to the turbulent velocity. As discussed in Section 3, the Orion A cloud probably is ``sheet-like''. At least along one line of sight, its molecular fraction is small. It may not, thus,  be representative of star-forming, molecular clouds. We note that in other molecular clouds where the matter density is better known, the velocities of the internal motions have been interpreted as equal to the Alfven velocities (Crutcher 1999)."
    },
    "0010/astro-ph0010089_arXiv.txt": {
        "abstract": "We have investigated how varying several parameters affects the results of a collision between an extragalactic jet and a dense, intergalactic cloud, through a series of hydrodynamic simulations.  We have produced synthetic radio images for comparison with observations. These show that a variety of structures may be produced from simple jet-cloud collisions. Moderate Mach numbers and density contrasts are needed to produce observable bends. We investigate the effect of viewing from various angles on the appearance of such sources. ",
        "introduction": "\\label{numsims}  Much of the understanding of the fluid dynamics of extragalactic jets, and of the shape and structures found in observations, has come from numerical simulations \\cite{W91}. Fully three-dimensional simulations have only become possible fairly recently \\cite{N93}, but many important results have been obtained from axisymmetric calculations. Several models for the production of complex and distorted structures in radio jets and lobes have been explored through numerical simulations including: variations in the direction of the jet at its source \\cite{WG85,S82}; cross-winds \\cite{leahy84}; the source axis is not parallel to the axis of a spheroidal gas distribution, or the source galaxy moves through the cluster medium \\cite{LW84}; helical instabilities \\cite{S97}; oblique magnetic fields in the intra-cluster medium \\cite{K96}. Cloud collisions are particularly applicable in cases where these bends are very sharp. Loken {\\it et al.} \\cite{L95} show that the necessary gas velocities % can arise in cluster mergers, as can shocks that will bend the jet. These simulations still do not explain the sharpness of the bend.     The first investigation of the effect of off-axis jet-cloud collisions was by De Young \\cite{DY91} using the `beam scheme' \\cite{SP74}. De Young observed that the jet was decelerated by the cloud. A similar interaction was investigated at higher resolution by Balsara and Norman using their RIEMANN code \\cite{N93}. They argued % that a De Laval nozzle was formed which re-accelerated the jet in a new direction after impact. They did not present any results at later time to show the formation of a deflected flow pattern.  More recently Raga and Canto \\cite{RC96} have published analytical calculations and two-dimensional simulations showing bending by clouds. They conclude that slower jets will be bent more, and clouds will be eroded as jets bore through them. ",
        "conclusions": ""
    },
    "0010/astro-ph0010330_arXiv.txt": {
        "abstract": "=cmr8 \\font\\keywords=cmr8 \\font\\caption=cmr8 \\font\\references=cmr8 \\font\\text=cmr10 \\font\\affiliation=cmssi10 \\font\\author=cmss10 \\font\\mc=cmss8 \\font\\title=cmssbx10 scaled\\magstep2 \\font\\alcit=cmti7 scaled\\magstephalf \\font\\alcin=cmr6 \\font\\ita=cmti8 \\font\\mma=cmr8 \\def\\ref{\\par\\noindent\\hangindent 15pt} \\null   \\title{\\ni Tests of Gaussianity of CMB maps}  \\bsk \\bsk \\author{\\ni Enrique Mart\\'\\i nez-Gonz\\'alez$^1$, R. Bel\\'en Barreiro$^{1,2}$, Jos\\'e M. Diego$^{1,2}$, Jos\\'e Luis Sanz$^1$, Laura Cay\\'on$^1$, Joseph Silk$^{3}$, Silvia Mollerach$^4$ \\& Vicent J. Mart\\'\\i nez$^5$}  \\bsk \\affiliation{$^1$ Instituto de F\\'\\i sica de Cantabria, Consejo Superior de Investigaciones Cient\\'\\i ficas-Universidad de Cantabria, Avda. Los Castros s/n, 39005 Santander, Spain\\\\ \\indent $^2$ Departamento de F\\'\\i sica Moderna, Universidad de Cantabria, Avda. Los Castros s/n, 39005 Santander, Spain\\\\ \\indent $^3$ Center for Particle Astrophysics, Department of Astronomy and Physics, University of California, Berkeley CA 94720-7304, USA\\\\ \\indent $^4$ Departamento de F\\'\\i sica, Universidad Nacional de La Plata, cc 67, La Plata 1900, Argentina\\\\ \\indent $^5$ Departamento de Astronom\\'\\i a y Astrof\\'\\i sica, Universidad de Valencia, E-46100 Burjasot, Valencia, Spain } \\bsk \\baselineskip = 12pt  \\abstract{ABSTRACT \\ni We study two different methods to test Gaussianity in CMB maps. One of them is based on the partition function and the other on the morphology of hot and cold spots. The partition function contains information on all the moments and scales, being a useful quantity to compress the large data sets expected from future space missions like Planck. In particular, it contains much richer information than the one available through the radiation power spectrum. The second method utilizes morphological properties of hot and cold spots such as the eccentricity and number of spots in CMB maps. We study the performance of both methods in detecting non-Gaussian features in small scale CMB simulated maps as those which will be provided by the Planck mission.  } \\bsk \\baselineskip = 12pt ",
        "introduction": " ",
        "conclusions": ""
    },
    "0010/hep-ph0010269_arXiv.txt": {
        "abstract": "s{ We analyse                 % a new mechanism for the cosmological QCD first-order             % phase transition: inhomogeneous nucleation. The primordial temperature fluctuations are larger than the tiny temperature interval, in which bubbles would form in the standard picture of homogeneous nucleation. Thus the bubbles nucleate at cold spots. We find the typical distance between bubble centers to be a few meters. This exceeds the estimates from homogeneous nucleation by two orders of magnitude. The resulting baryon inhomogeneities may affect primordial nucleosynthesis. }  \\clearpage  \\pagestyle{plain}      First-order phase transitions normally proceed via nucleation of bubbles of the new phase. When the temperature is spatially uniform and no impurities are present, the mechanism is homogeneous nucleation. We denote by $v_{\\rm heat}$ the effective speed by which released latent heat propagates in sufficient amounts to shut down nucleations. The mean distance between nucleation centers, measured immediately after the transition completed, is \\begin{equation} d_{\\rm nuc,hom} = 2 v_{\\rm heat} \\Delta t_{\\rm nuc} \\label{dnuceq}, \\end{equation} where $\\Delta t_{\\rm nuc}$ is the duration of the nucleation period   % \\cite{IS,Ignatius}.                                                   %    Surface tension and latent heat are provided by lattice simulations with quenched QCD only, giving the values $\\sigma = 0.015 T_c^3$, $l = 1.4 T_c^4$~\\cite{Iwasaki}. Scaling the latent heat for the physical QCD leads us to take $l = 3 T_c^4$. Using              % these values the amount of supercooling is $\\Delta_{\\rm sc} = 2.3 \\times 10^{-4}$, with large error bars.     % Due to rapid change of energy density near $T_c$, the microscopic sound  % speed in the quark phase has a small value,                              % $3 c_{\\rm s}^2(T_{\\rm c}) < 1$                                           % (for references see \\cite{IS}).                                          %  In the real Universe the local temperature of the radiation fluid fluctuates. The temperature contrast is denoted by $\\Delta \\equiv \\delta T/ \\bar{T}$. On subhorizon scales in the radiation dominated epoch, each Fourier coefficient $\\Delta(t,k)$ oscillates with constant amplitude $\\Delta_T(k)$. Inflation predicts a Gaussian distribution, \\begin{equation} \\label{pd} p(\\Delta){\\rm d}\\Delta = {1\\over \\sqrt{2\\pi}\\Delta_T^{\\rm rms}} \\exp\\left( - \\frac12 {\\Delta^2\\over (\\Delta_T^{\\rm rms})^2}\\right) {\\rm d}\\Delta \\ . \\end{equation} We find \\cite{norm} for the COBE normalized \\cite{Bennett} rms temperature fluctuation of the radiation fluid (not of cold dark matter) $\\Delta_T^{\\rm rms} = 1.0 \\times 10^{-4}$ for a primordial Harrison-Zel'dovich spectrum. The change of the equation of state prior to the QCD transition modifies the temperature-energy density relation, $\\Delta = c_s^2 \\delta\\varepsilon /(\\varepsilon + p)$. We may neglect the pressure $p$ near the critical temperature since $p \\ll \\varepsilon_{\\rm q}$ at $T_{\\rm c}$. On the other hand the drop of the sound speed enhances the amplitude of the density fluctuations proportional to $c_s^{-1/2}$ \\cite{SSW}.             % Putting all those effects together and allowing for a tilt in the power spectrum, the COBE normalized rms temperature fluctuation reads \\begin{equation} \\Delta_T^{\\rm rms} \\approx 10^{-4} (3 c_s^2)^{3/4} \\left({k\\over k_0}\\right)^{(n-1)/2} \\ . \\end{equation} For a Harrison-Zel'dovich spectrum and $3 c_s^2 = 0.1$, we find $\\Delta_T^{\\rm rms} \\approx 2 \\times 10^{-5}$.  A small scale cut-off in the spectrum of primordial temperature fluctuations comes from collisional damping by neutrinos \\cite{SSW}.                      % For $l_{\\rm smooth}$, the length over which temperature is constant,    % we take the value $10^{-4} d_{\\rm H}$ \\cite{IS}.                        % The temperature fluctuations are frozen with respect to the time scale of nucleations.  Let us now investigate bubble nucleation in a Universe with spatially inhomogeneous temperature distribution. Nucleation effectively takes place while the temperature drops by the tiny amount $\\Delta_{\\rm nuc}$. To determine the mechanism of nucleation, we compare $\\Delta_{\\rm nuc}$ with the rms temperature fluctuation $\\Delta_T^{\\rm rms}$:  1. If $\\Delta_T^{\\rm rms} < \\Delta_{\\rm nuc}$, the probability to nucleate a bubble at a given time is {\\em homogeneous} in space. This is the case of homogeneous nucleation.  2. If $\\Delta_T^{\\rm rms} > \\Delta_{\\rm nuc}$, the probability to nucleate a bubble at a given time is {\\em inhomogeneous} in space. We call this inhomogeneous nucleation.  \\begin{figure}[t] \\begin{center} \\epsfig{figure=itfig1.eps,width=0.325\\linewidth}          % \\epsfig{figure=itfig2.eps,width=0.325\\linewidth}          % \\epsfig{figure=itfig3.eps,width=0.325\\linewidth}          % \\end{center} \\caption{\\label{fig1} Sketch of a first-order QCD transition in the inhomogeneous Universe. At $t_1$ the first hadronic bubbles (H) nucleate at the coldest spots (light gray), while most of the Universe remains in the quark phase (Q). At $t_2$ the bubbles inside the cold spots have merged and have grown to bubbles as large as the temperature fluctuation scale. At $t_3$ the transition is almost finished. The last quark droplets are found in the hottest spots (dark gray).} \\end{figure} The quenched lattice QCD data and a COBE normalized flat spectrum lead to the values $\\Delta_{\\rm nuc} \\sim 10^{-6}$ and $\\Delta_T^{\\rm rms} \\sim 10^{-5}$. We conclude that the cosmological QCD transition may proceed via inhomogeneous nucleation. A sketch of inhomogeneous nucleation is shown in Fig.~\\ref{fig1}. The basic idea is that temperature inhomogeneities determine the location of bubble nucleation. Bubbles nucleate first in the cold regions.  For the fastest fluctuations, with angular frequency $c_s/l_{\\rm smooth}$, we find \\begin{equation} \\label{coolingrate} {{\\rm d}T(t,{\\bf x})\\over {\\rm d} t} = {\\bar T\\over t_{\\rm H}} \\left[- 3 c_s^2 + {\\cal O}\\left(\\Delta_T {t_{\\rm H}\\over \\delta t}\\right)\\right]\\ . \\end{equation} The Hubble expansion is the dominant contribution, as typical values are $3 c_s^2 = 0.1$ from quenched lattice QCD and $\\Delta_T^{\\rm rms} t_{\\rm H}/\\delta t \\approx 0.01$. This means that the local temperature does never increase, except by the released latent heat during bubble growth.  To gain some insight in the physics of inhomogeneous nucleation, let us first inspect a simplified case. We have some randomly distributed cold spheres of diameter $l_{\\rm smooth}$ with equal and uniform temperature, which is by the amount $\\Delta_T^{\\rm rms}T_c$ smaller than the again uniform temperature in the rest of the Universe. When the temperature in the cold spots has dropped to $T_{\\rm f}$, homogeneous nucleation takes place in them. Due to the Hubble expansion the rest of the Universe would need the time $\\Delta t_{\\rm cool} = t_H \\Delta^{\\rm rms}_T  / 3 c_s^2$          % to cool down to $T_{\\rm f}$. The cold spots have fully been transformed into the hadron phase while the rest of the Universe still is in the quark phase. The latent heat released in a cold spot propagates in all directions, which provides the length scale \\begin{equation} \\label{lheat} l_{\\rm heat} \\equiv 2 v_{\\rm heat} \\Delta t_{\\rm cool}. \\end{equation} If the typical distance from the boundary of a cold spot to the boundary of a neighboring cold spot is less than $l_{\\rm heat}$, then no hadronic bubbles can nucleate in the intervening space.  The real Universe consists of smooth patches of typical linear size $l_{\\rm smooth}$, their temperatures given by the distribution~(\\ref{pd}). At time $t$ heat, coming from a cold spot which was transformed into hadron phase at time $t'$, occupies the volume $V(t,t') = (4\\pi/3) [l_{\\rm smooth}/2 + v_{\\rm heat}(t-t')]^3$. The fraction of space that is not reheated by the released latent heat (and not transformed to hadron phase), is given at time $t$ by \\begin{equation} \\label{fihn} f(t) \\approx 1 - \\int_0^{t}\\Gamma_{\\rm ihn}(t') V(t,t') {\\rm d}t', \\end{equation} where we neglect overlap and merging of heat fronts. $\\Gamma_{\\rm ihn}$ is the volume fraction converted into the new phase, per physical time and volume as a function of the mean temperature $T=\\bar{T}(t)$. We find                         % \\begin{equation} \\label{gammai} \\Gamma_{\\rm ihn} = 3 c_s^2{T_f\\over T}{1\\over t_{\\rm H} {\\cal V}_{\\rm smooth}} p(\\Delta = \\frac{T_{\\rm f}}{T} - 1) , \\end{equation} where the relevant physical volume is ${\\cal V}_{\\rm smooth} = (4\\pi/3) (l_{\\rm smooth}/2)^3.$  The end of the nucleation period, $t_{\\rm ihn}$, is defined through the condition $f(t_{\\rm ihn}) = 0$. We introduce the variables $N \\equiv (1 - T_{\\rm f}/T)/\\Delta_T^{\\rm rms}$ and ${\\cal N} \\equiv N(t_{\\rm ihn})$. Putting everything together we determine ${\\cal N}$ from \\begin{equation} {l_{\\rm heat}^3\\over l_{\\rm smooth}^3} \\int_{\\cal N}^\\infty {\\rm d} N \\frac{e^{-\\frac12 N^2}}{\\sqrt{2\\pi}} \\left(\\frac{l_{\\rm smooth}}{l_{\\rm heat}} + N - {\\cal N}\\right)^3 = 1 . \\end{equation} For $l_{\\rm heat}/l_{\\rm smooth} = 1, 2, 5, 10$ we find ${\\cal N} \\approx 0.8, 1.4, 2.1, 2.6$, respectively.  The effective nucleation distance in inhomogeneous nucleation is defined from the number density of those cold spots that acted as nucleation centers, $d_{\\rm nuc,ihn} \\equiv n^{- 1/3}$. We find \\begin{eqnarray} d_{\\rm nuc,ihn} & \\approx & \\left[\\int_0^{t_{\\rm ihn}} \\Gamma_{\\rm ihn}(t) {\\rm d} t\\right]^{-1/3} \\\\ & = & [\\frac{3}{\\pi} (1 - {\\rm erf}({\\cal N}/\\sqrt{2})]^{-1/3} \\, l_{\\rm smooth}. \\end{eqnarray} With the above values $l_{\\rm heat}/l_{\\rm smooth} = 1, 2, 5, 10$ we get $ d_{\\rm nuc,ihn} = 1.4, 1.8, 3.0, $ $            % 4.8 \\times l_{\\rm smooth}$, where $l_{\\rm smooth} \\approx 1 \\mbox{\\ m}$.  For a COBE normalized spectrum without any tilt and with a tilt of $n-1 = 0.2$, together with $3c_s^2 = 0.1$ and $v_{\\rm heat} = 0.1$, we find the estimate $l_{\\rm heat}/l_{\\rm smooth} \\approx 0.4$ and 9, correspondingly. Notice that the values of $v_{\\rm heat}$ and $3 c_s^2$ are in principle unknown. Anyway, we can conclude that the case $\\l_{\\rm heat} > l_{\\rm smooth}$ is a realistic possibility.  With $2 v_{\\rm heat}(3c_s^2)^{-1/4} (10^{-4}d_{\\rm H}/l_{\\rm smooth}) < 1$ and without positive tilt we are in the region $l_{\\rm heat} < l_{\\rm smooth}$, where the geometry is more complicated and the above quantitative analysis does not apply. In this situation nucleations take place in the most common cold spots (${\\cal N} \\sim 1$), which are very close to each other. We expect a structure of interconnected baryon-depleted and baryon-enriched layers with typical surface $l_{\\rm smooth}^2$ and thickness $l_{\\rm def} \\equiv v_{\\rm def}\\Delta t_{\\rm cool}$.  We emphasize that inhomogeneous and heterogeneous nucleation~\\cite{Christiansen} are genuinely different mechanisms, although they give the same typical scale of a few meters by chance. If latent heat and surface tension of QCD would turn out to reduce $\\Delta_{\\rm sc}$ to, e.g., $10^{-6}$, instead of $10^{-4}$, the maximal heterogeneous nucleation distance would fall to the centimeter scale, whereas on the distance in inhomogeneous nucleation this would have no effect.  We have shown that inhomogeneous nucleation during the QCD transition can give rise to an inhomogeneity scale exceeding the proton diffusion scale (2~m at 150~MeV).                  % The resulting baryon inhomogeneities could provide inhomogeneous initial conditions for nucleosynthesis. Observable deviations from the element abundances predicted by homogeneous nucleosynthesis seem to be possible in that case~\\cite{MathewsKainulainen}.  In conclusion, we found that inhomogeneous nucleation leads to nucleation distances that exceed by two orders of magnitude estimates based on homogeneous nucleation. We point out         % that this new effect appears for the (today) most probable range of cosmological and QCD parameters. ",
        "introduction": " ",
        "conclusions": ""
    },
    "0010/astro-ph0010366_arXiv.txt": {
        "abstract": "We investigate to what extent future microwave background experiments might be able to detect a suppression of fluctuation power on large scales in flat and open universe models.  Such suppression would arise if fluctuations are generated by causal processes, and a measurement of a small suppression scale would be problematic for inflation models, but consistent with many defect models.  More speculatively, a measurement of a suppression scale of the order of the present Hubble radius could provide independent evidence for a fine-tuned inflation model leading to a low-density universe.  We find that, depending on the primordial power spectrum, a suppression scale modestly larger than the visible Horizon can be detected, but that the detectability drops very rapidly with increasing scale. { }For models with two periods of inflation, there is essentially no possibility of detecting a causal suppression scale.  PACS No.: 98.80.Es, 98.70.Vc, 98.80.Cq ",
        "introduction": "\\label{intro}  If we assume that fluctuations in the Universe were created by causal processes, then an unambiguous feature of the power spectrum is the maximum scale over which such fluctuations are correlated.  In principle this scale is a characteristic of all density fields in the Universe, matter, cosmic microwave background radiation (CMB) scalar and tensor fluctuations, primordial magnetic fields etc.  However, the effects of this scale are generally difficult to detect, as they inevitably occur on a very large scale. Indeed, in inflationary models, the scale may exceed the Hubble length $1/H$, and we will refer to this scale as the super-Hubble suppression scale.  Efforts to understand the super-Hubble scale primarily have focused on the CMB scalar perturbation power spectrum.  {}For scalar perturbations, as well as any matter perturbations, if they are created by local causal processes, it is well known that causality requires the power spectrum to diminish at least as fast as $k^4$ at small $k$.  The argument for this was first put forth by \\scite{zel65} and over the years it has been refined \\cite{Pee74,CarrSilk83,at86,rw96}.  The advent of inflationary cosmology has placed specific focus on the size of the super-Hubble scale, in particular relative to the Hubble distance $1/H$ (we set $c=1$ throughout).  Whereas the Hubble distance characterizes the maximum distance over which causal interactions can occur during one expansion time, in distinction, the super-Hubble length scale $\\lambda_H$ characterizes the absolute largest scale over the lifetime of the Universe, on which causal interactions can seed density perturbations.  In particular the super-Hubble scale accounts for the growth of all local perturbations since their conception up to the present due to the expansion of the universe.  {}For example, in a flat universe, a perturbation of the maximum causal physical length created at time $t_i$ expands to a physical scale at the present time $t_0$ \\begin{equation} \\lambda(t_0; t_i) = \\frac{R(t_0)}{R(t_i)} \\int_{0}^{t_i} \\frac{dt'}{R(t')}, \\end{equation} where $R(t)$ is the cosmic scale factor.  Depending on the behaviour of the cosmological scale factor $R(t)$, this can exceed the present Hubble radius $1/H_0$.  In the regime of Standard Hot Big Bang cosmology $R(t) \\sim t^{1/2}$ and $t^{2/3}$ in the radiation- and matter-dominated regimes respectively.  On the other hand, inflation is a regime in which the Hubble parameter $H \\equiv {\\dot R(t)}/{R(t)}$ is approximately constant and the scale factor is accelerating, ${\\ddot R(t)} > 0$, with generically a quasi-exponential growth.  {}For example, if at $t_i$ during inflation a local perturbation is created at scale $1/H$, by the end of inflation at $t_f$ it will imply a super-Hubble scale of $\\lambda(t_f) \\approx (1/H) \\exp[H(t_f-t_i)]$. Thus immediately after even a few $e-$foldings of inflation, $\\lambda \\gg 1/H$.  Outside the inflation regime, the causal horizon in the standard Big-Bang regime grows in physical units as $\\lambda \\sim t$.  In principle this horizon eventually will grow bigger than the one set by inflation, which in physical units grows as $\\lambda R(t)/R(t_f) \\sim t^a \\lambda$ where $a\\sim 1/2$ during radiation-domination and $a\\sim 2/3$ during matter-domination. However the generic prediction of most inflation models is that $\\lambda(t_0) \\gg \\eta_0 \\sim 2/H_0$, where $\\eta_0$ is the physical distance from last scattering or equivalently the causal horizon for events generated since last scattering.  Thus, the generic expectation for density perturbations produced solely within non-inflationary regimes is $\\lambda \\sim 1/H_0$, whereas isentropic inflation models \\cite{Guth81,AS82,Linde82,Linde83,Olive90} generally predict $\\lambda \\gg 1/H_0$ and non-isentropic inflation models \\cite{Berera95,Berera97} are fairly impartial to any particular scale between these two limits. As such, the super-Hubble scale, in principle, is an ideal parameter for discriminating between inflationary and non-inflationary models of density perturbations.  However the practical limitation is that the super-Hubble scale can be measured accurately only if it no greater than about $1/H_0$.  Thus at best one hopes to distinguish from the data whether $\\lambda \\stackrel{<}{\\sim} 1/H_0$ or $\\lambda \\stackrel{>}{\\sim} 1/H_0$.  If such a measurement could be reliably made, it would be far-reaching: evidence for a maximum causal scale which was smaller or of order the Hubble radius would not be expected in inflation, but could be produced by, say, topological defect models generating fluctuations causally after the Big Bang.  A more speculative possibility is that inflation produced only just enough e-foldings to solve the horizon problem, but not enough to drive the present Universe to flatness.  The fine-tuning required for this is problematical, but measurement of a super-Hubble scale modestly larger than $1/H_0$ could provide evidence, independent of a measurement of $\\Omega_0$, for such non-flat inflation.  {}For the Cosmic Background Explorer experiment (COBE), a likelihood fit to the four-year data with respect to the index, amplitude and super-Hubble suppression scale was done in \\scite{bfh97}.  Their results interestingly preferred a super-Hubble suppression scale $\\lambda \\sim 4/H_0$ (or $k_{min}\\eta_0 \\sim 3.5$). Since the preference was within one-sigma of an infinite suppression scale, the conclusion of their analysis was that no super-Hubble suppression scale $\\lambda > 1/H_0$ was excluded.  However, the fact that this analysis did not give a clear-cut measurement of an infinite suppression scale leaves some margin of doubt.  Due to the importance that detection of a small suppression scale would have in discriminating cosmological models, it is advantageous at this stage to explore the possibilities for measuring this scale.  What are the possibilities of improving on this measurement with new experiments? The first point is that since the signal for super-Hubble suppression is confined to low-order multipoles in the CMB signal, instrumental noise is not the dominant error, and the higher signal-to-noise of MAP and Planck will give little advantage over COBE.  However, there are still two advantages. One is that the CMB power spectrum at low multipoles is determined not just by the suppression scale, but other parameters as well: the newer experiments will determine the other parameters rather precisely, thus reducing the marginal error on the suppression scale. Second, the better frequency coverage of the later experiments (especially Planck) should allow better control of systematic errors, especially through effective removal of foreground sources.  For the quadrupole, the successful removal of as much foreground emission from the Milky Way galaxy itself will be essential; the systematic uncertainty in the COBE quadrupole is about 70\\% of its measured value \\cite{K96}.  This very large uncertainty is serious for measurement of the super-Hubble suppression scale. On the one hand, this systematic uncertainty could fully account for the suppressed quadrupole found by COBE and thus cast doubt on the claimed measurement in \\cite{bfh97}.  On the other hand, this large uncertainty implies the size of the super-Hubble suppression scale remains an open question and conceivably it may be small enough to be detected.  In this paper, we take an optimistic view that foreground subtraction and parameter estimation (other than of the suppression scale) can be performed precisely, and calculate the error with which the suppression scale could be determined from CMB measurements.  We do this via a likelihood analysis (Section \\ref{Like}).  Section \\ref{Open} explores how an open inflation model, leading to $\\Omega_0<1$, can lead to a measurably small super-Hubble scale, which might provide independent support for such a model.  In Section \\ref{ThCon} we explore the theoretical consequences of a super-Hubble scale suppression, and draw our conclusions in section \\ref{Concs}. ",
        "conclusions": "\\label{Concs}  We have investigated the extent to which a large-scale suppression scale in primordial fluctuation power might be detectable in flat and open universe models.  We have concentrated on observations of the microwave background radiation, as direct measurement of the suppression scale from the galaxy power spectrum is likely to be impossible, since the scale is likely to be at least a substantial fraction of the Hubble radius.  The suppression itself is expected if the fluctuations arise from causal processes.  The limit of detectability is dependent a little on cosmology, and on the primordial power spectrum, but it is no surprise that it is set roughly by the size of the visible horizon today.  In flat models, a positive cosmological constant makes the suppression scale easier to detect, as the visible horizon is larger.  Defect models will be distinguished from inflationary models most easily by their high-$\\ell$ power spectrum or by non-gaussian tests (e.g. \\pcite{HS99}).  Should inflation models be ruled out, then the suppression scale may become a useful discriminant between defect models, as it should appear on scales smaller than the visible horizon, and be readily detectable.  Clearly, it would also act as a useful cross-check.  {}For single-inflation models, the suppression scale depends on both the final value of the matter density parameter, and also on the radiation density parameter immediately prior to inflation. {}For low final densities, there is the possibility of detecting a moderately small suppression scale, which would provide some sort of independent evidence on what would be a fine-tuned model. {}For open inflation models involving two stages of inflation, there appears to be no possibility of detection of the causal suppression scale set by the first stage of old inflation.  However, as mentioned in Sect. \\ref{ThCon}, we can infer from the analysis in Sect. \\ref{Open} that any significant effect that is slightly super-Hubble scale could be detectable.  In particular, if the bubble wall created in the second stage of inflation is just beyond the Hubble radius, the type of measurements we have been considering could be applied to detect it. However, we have not made any detailed analysis of this possibility.  We have considered only scalar modes here; in principle it is possible to include tensor modes to increase the detectability, but in fact the transition between a scale which is easily detectable to one which is not is so sharp that such additional information would change the results very little.  We expect the suppression scale signal to be contained in the low-order multipoles, which were measured by COBE.  {}Future experiments such as MAP and Planck will improve on COBE not principally because of better signal-to-noise, but via improved foreground subtraction through their better frequency coverage.  In addition, their better resolution will allow better determination of cosmological parameters.  The reduced uncertainty in other parameters will decrease the marginal error on the suppression scale.  This is quantifiable in principle, but the number of cosmological parameters is large (up to $\\sim 20$ \\scite{BET97}).  It is in the better foreground subtraction that MAP and Planck should gain: the systematic uncertainty in the COBE quadrupole is comparable to the measured signal \\cite{K96}, large enough that the apparent suppression of the quadrupole in the COBE data \\cite{bfh97} could be entirely due to this rather than super-Hubble suppression.  Provided that foregrounds can be adequately removed and other cosmological parameters are determined accurately from the high-order power spectrum, a factor of 3 improvement in the quadrupole noise is expected, and this should allow the super-Hubble scale suppression to be estimated to the accuracy presented in this paper.  If such a scale is indeed measured, then, as with other measurements, polarization measurements can be used as a consistency check."
    },
    "0010/astro-ph0010150_arXiv.txt": {
        "abstract": "}[2]{{\\footnotesize\\begin{center}ABSTRACT\\end{center} \\vspace{1mm}\\par#1\\par \\noindent {~}{\\it #2}}}  \\newcommand{\\TabCap}[2]{\\begin{center}\\parbox[t]{#1}{\\begin{center} \\small {\\spaceskip 2pt plus 1pt minus 1pt T a b l e} \\refstepcounter{table}\\thetable \\\\[2mm] \\footnotesize #2 \\end{center}}\\end{center}}  \\newcommand{\\TableSep}[2]{\\begin{table}[p]\\vspace{#1} \\TabCap{#2}\\end{table}}  \\newcommand{\\FigCap}[1]{\\footnotesize\\par\\noindent Fig.\\  % \\refstepcounter{figure}\\thefigure. #1\\par}  \\newcommand{\\TableFont}{\\footnotesize} \\newcommand{\\TableFontIt}{\\ttit} \\newcommand{\\SetTableFont}[1]{\\renewcommand{\\TableFont}{#1}}  \\newcommand{\\MakeTable}[4]{\\begin{table}[htb]\\TabCap{#2}{#3} \\begin{center} \\TableFont \\begin{tabular}{#1} #4 \\end{tabular}\\end{center}\\end{table}}  \\newcommand{\\MakeTableSep}[4]{\\begin{table}[p]\\TabCap{#2}{#3} \\begin{center} \\TableFont \\begin{tabular}{#1} #4 \\end{tabular}\\end{center}\\end{table}}  \\newenvironment{references}% { \\footnotesize \\frenchspacing \\renewcommand{\\thesection}{} \\renewcommand{\\in}{{\\rm in }} \\renewcommand{\\AA}{Astron.\\ Astrophys.} \\newcommand{\\AAS}{Astron.~Astrophys.~Suppl.~Ser.} \\newcommand{\\ApJ}{Astrophys.\\ J.} \\newcommand{\\ApJS}{Astrophys.\\ J.~Suppl.~Ser.} \\newcommand{\\ApJL}{Astrophys.\\ J.~Letters} \\newcommand{\\AJ}{Astron.\\ J.} \\newcommand{\\IBVS}{IBVS} \\newcommand{\\PASP}{P.A.S.P.} \\newcommand{\\Acta}{Acta Astron.} \\newcommand{\\MNRAS}{MNRAS} \\renewcommand{\\and}{{\\rm and }} {We present the {\\it BVI} photometric maps of the Large Magellanic Cloud. They contain {\\it BVI} photometry and astrometry of more than 7 million stars from the central parts of the LMC. The data were collected during the second phase of the OGLE microlensing project. We discuss the accuracy of the data and present color-magnitude diagrams of all 26 fields observed by OGLE in the LMC.  The {\\it BVI} maps of the LMC are accessible electronically for the astronomical community from the OGLE Internet archive.} ",
        "introduction": "The Large Magellanic Cloud (LMC) is one of the most important astrophysical objects. This nearby galaxy hosts large variety of very important stellar populations and is an ideal laboratory for testing our understanding of stellar structure and evolution and for calibrating the most important standard candles. The LMC is also very important object for extragalactic astrophysics -- the extragalactic distance scale is calibrated by the distance to the LMC. Due to relatively small distance from the Galaxy, stars and other objects from the LMC can easily be resolved and measured. Approximately the same distance to all objects from that galaxy makes it especially attractive for determination and comparison of basic parameters of different groups of stars.  Unfortunately, the LMC was relatively poorly observed, in particular with precise CCD techniques. Photometry of only selected fields in lines of sight usually toward star clusters in the halo of this galaxy can be found in literature. The situation considerably changed when the large microlensing surveys began regular monitoring of the Magellanic Clouds for microlensing events. The LMC is one of the most promising targets of the microlensing hunt, and therefore it was selected as a target of MACHO (Alcock \\etal 2000) and EROS (Lasserre \\etal 2000) microlensing projects. Large areas in the LMC were also observed by Zaritsky, Harris and Thompson (1997) as a part of a photometric survey of the Magellanic Clouds.  The LMC was included as one of the targets of the Optical Gravitational Lensing Experiment (OGLE) at the beginning of the second phase of the project (Udalski, Kubiak and Szyma{\\'n}ski 1997). Since January 1997 the large part of the central regions of the galaxy has been regularly observed practically on every clear night. As a result huge databases containing hundreds of photometric measurements of millions of stars were created.  The OGLE photometric data are particularly attractive for many projects requiring precise photometry. Among the other datasets collected during the microlensing searches, only the OGLE photometry was obtained with the standard {\\it BVI} filters what makes it very well suited for projects not related to microlensing. Also, very good astronomical site where the OGLE project is conducted -- the Las Campanas Observatory in Chile, made it possible to achieve the best resolution what is crucial in the very dense stellar fields of the LMC bar.  In this paper, which is the second of the series, we present \"The OGLE {\\it BVI} Maps of the LMC.\" The maps contain the mean {\\it BVI} photometry and astrometry of more than 7 million stars from the field of about 5.7~square degrees in the central part of the LMC. With the maps of the Small Magellanic Cloud presented in the first paper of this series (Udalski \\etal 1998b) they constitute a unique photometric database of objects from the Magellanic Clouds.  Because of potentially great impact on many astrophysical fields, the OGLE policy is to make the photometric data available to the astronomical community. Similarly to the SMC maps, the maps of the LMC are also available from the OGLE Internet archive. Details are provided in the last Section of this paper. ",
        "conclusions": ""
    },
    "0010/astro-ph0010199_arXiv.txt": {
        "abstract": "This paper describes some interesting properties of waves in, and oscillations of, the interstellar medium in the direction normal to the plane of the Galaxy. Our purpose is to examine possible reasons for four observed phenomena: the falling sky in the northern hemisphere; the apparent presence of clouds in absorption spectra when a sightline is occupied primarily only by warm intercloud gas; the peculiar structuring of spiral arms involving clumps, spurs, and feathering; and the existence of an abundance of high stage ions far off the plane of the Galaxy.  We explored the reaction of the interstellar medium - in the vertical direction only - to large imposed disturbances (initial displacements, expansive velocities, and compressions), and to the introduction of small amplitude waves via oscillation of the midplane. Our findings included: 1) the anticipated growth in amplitude of high frequency waves with height; 2) the four lowest normal modes for the oscillation of the atmosphere as a whole, as functions of the height of the outer boundary; 3) the time for material to `bounce' from one unusually dense state to the next as a function of height; and 4) the tendency for the disk to develop a hot outer halo, either after the passage of a single shock from a large event, or in response to a continuous stream of small amplitude waves.  We discovered that three of the four observed phenomena targeted are likely to be closely connected. Following a large expansion, material near the plane falls back first, with material at higher $z$ then falling in upon it. This provides precisely the sort of velocity segregation observed in the northern sky, at about 50 Myr after the event.  In addition, this bounce time (and/or the period of the subsequent smaller oscillations, which is about twice the bounce time) may be linked to structure in the spiral arms, with vertical oscillations having been provoked by initial compressions in the arms. Oscillations of the fundamental symmetric (breathing) mode of the ISM also produce substantial disturbances in the outer atmosphere. This can result in the production of an extensive layer of hot gas overlying the cooler disk material, i.e. a hot galactic halo with a significant population of high stage ions. Hence, three of the four phenomena may be natural results of the simple existence of strong local compressions at the spiral arms, and the associated vertical motions in a thick galactic disk.  Finally, the somewhat mysterious appearance of clouds in some absorption spectra can be produced by small amplitude waves in the ISM. Under the right conditions, clouds will seem to appear through ``velocity crowding,'' when in fact there are no density concentrations in space. ",
        "introduction": "In this paper, we discuss the results of a range of experiments performed with a 1-dimensional hydrocode to determine the characteristics of vertical oscillations of the interstellar medium. We were interested in the dynamics of the ISM in the direction normal to the plane of the Galaxy (i.e. the vertical direction).  Because our model treats the distribution of the ISM as plane parallel, we sometimes refer to this as the ``galactic atmosphere problem.'' Our experiments were aimed toward shedding some light on four observed aspects of the interstellar medium.  The first of these aspects is the fact that in roughly half of the northern sky, 21 cm observations indicate that there are two primary components to the material distribution, a low velocity component and another, comparable in column density, which is falling toward the plane with a velocity on the order of 50 km s$^{-1}$ (e.g. Stark et al. 1992; see Figure \\ref{fig20} for a typical profile). This infalling component has been studied extensively (e.g. Danly et al. 1992) and named the IV (Intermediate Velocity) Arch. Our experiments explored dynamical situations that could lead to such velocity segregation, a circumstance that we refer to loosely as the ``falling sky.''  The second aspect is revealed by the UV absorption line spectrum of HD 93521 (Spitzer \\& Fitzpatrick 1993), a star which has a $z$ distance of about 1.5 kpc and lies behind the IV Arch. The H~I spectrum in that direction shows the usual two components, with considerably more than thermal broadening, and one rather narrower component at roughly -10 km s$^{-1}$\\ heliocentric velocity. On the other hand, the UV spectrum (with the reduced thermal broadening of the lines of the heavier elements) shows a total of about 10 components, which have been interpreted as localized regions of enhanced density. However, it is well known that velocity crowding (caustics in the velocity distribution) can produce spectral features in an otherwise uniform medium. Hence we performed experiments to explore whether the features observed toward this star might in part derive from typical interstellar velocity irregularities rather than density concentrations in space.  The third aspect involves the structure of spiral arms in galaxies. Such arms are generally not entirely smooth and continuous, having localized regions of enhanced contrast, in a pattern that could be called feathering. One gets the impression that (as in the wake of a boat) there are wavelets that are not traveling normal to the locus of the arms. In addition, there are larger irregularities - pieces of arms or spurs - that branch off of the smooth locus of the more regular pattern. Martos \\& Cox (1998) began to explore the vertical structure anticipated for a spiral arm, including the thick disk and the significant nonthermal pressure. They used only a two-dimensional model, but speculated on some of the potential three-dimensional characteristics, including the possibility that arm substructure might be related to resonances in the induced vertical motions. In this paper we have performed two sets of experiments useful in exploring the relevance of this idea, determining the normal modes of vertical oscillations and the timescale for interstellar material to return to the galactic plane after being set into motion away from it.  The fourth aspect is related to the substantial quantities of high stage ions (in particular C~IV, N~V, and O~VI) found well off the plane of the Galaxy, with apparent scale heights up to several kpc (Savage \\& Massa 1987). A variety of possible sources for these ions have been suggested. Here we are exploring only one, a possibility related to the suggestion by Sciama (1972) that the outer boundary of the cool interstellar medium might be chromospheric, with a layer of hot gas beyond that. Our final set of experiments was aimed at examining the structure of the outer edge of the disk resulting from mechanical disturbances within it, to learn whether the transition to a hot outer layer is a natural consequence. ",
        "conclusions": "We found several potentially important properties of the interstellar medium in our experiments. Probably the most intriguing of these is that segregation of infalling material into two dominant components occurs naturally after any large expansive event in the ISM as a whole. The material closest to the midplane returns first and stops. After roughly 50 Myr a substantial low velocity column density has collected, with outer material continuing to fall in upon it. With a sufficiently large velocity differential, the two components are separated by a nearly stationary shock. The large structure known as the IV Arch, covering about half of the northern sky, could well be the consequence of such a process.  The concentration of material in velocity space (rather than in real space) is a natural occurrence in the ISM, and can easily lead to apparent structures in the diffuse gas which mimic low-density clouds. Experiments in which we added small velocity perturbations (via oscillation of the galactic midplane) to cases in which a velocity segregated infall was already present (from a general expansion, as above) provided velocity spectra closely resembling those of HD 93521, even though there were no density concentrations, i.e. clouds, whatsoever. We suggest that perhaps only the one colder component in the spectrum of that star is actually a localized density feature.  We found the periods of the four lowest normal modes for our model ISM as functions of the height of the outer boundary, both through Fourier analysis of the oscillations in the aftermath of a major perturbation, and via linear analysis, with the two methods giving remarkably similar results. Our hydrodynamics experiments further revealed two curious tendencies: first that the oscillating structure appears to concentrate energy into two modes, which two depending on the nature of the initial perturbation; and second that the relative amplitudes of the two modes are comparable at the outer boundary, which therefore shows substantial amplitude modulation in the beat pattern.  We also explored a timescale that we refer to as the bounce time. Starting with a large compression of material near the midplane, the bounce time at a given height is the length of time for the material to return to its next minimum height. For material in the general vicinity of the plane, the bounce time tended to lie between the half period of the lowest frequency breathing mode and the half period of the first overtone, depending on the magnitude of the initial compression, and therefore perhaps on the relative amplitudes of the excited modes. In general, the bounce time for material near the plane lay in the range 40 to 60 Myr, while material beyond 1.5 kpc took substantially longer to return. It is this characteristic difference that leads to the velocity segregation discussed above. It is our impression that the approximate equality between the bounce time and the half period of the most strongly excited breathing mode derives from the fact that the amplitude is very nonlinear in the initial and secondary compressions, with the result that very little time is spent in the compressed state. Most of the bounce time is spent in the expanded state, which lasts roughly half the period of the oscillation itself.  For atmospheric boundary heights, $z_{max}$, in the range 2 to 4 kpc, we found the periods of the fundamental breathing mode to be between 100 and 120 Myr, and bounce times to be about half those values. Following Martos \\& Cox (1998), we should then expect to find dense structures within or along spiral arms, separated in space by distances proportional to relative velocity and timescales of 50 to 60 Myr, as the disturbed structure bounces. With a relative velocity within the arm of 20 km s$^{-1}$, the spacing would be roughly 1 kpc. With an average relative speed between the pattern and rotation of 100 km s$^{-1}$, the scale would be 5-6 kpc. If the first overtone in the breathing mode is excited, the half periods drop to the range of 30 to 40 Myr, and the characteristic feature spacings for the assumed velocities reduce to 0.6-0.8 kpc and 3-4 kpc (for relative speeds of 20 km s$^{-1}$ and 100 km s$^{-1}$, respectively). For comparison, the circumference at the solar circle is roughly 50 kpc, and the timescale for material to circumnavigate it, relative to the pattern, is of order 500 Myr. With a bounce timescale of roughly one tenth that, one might expect to encounter roughly ten unusually dense regions around the circle, some spaced closer than 1 kpc (feathers?) with others several kpc apart along the circle (arms and spurs?). If the full oscillations of the normal modes (rather than strong bounces) are responsible for the structuring, the feature separations will be roughly twice these, their number on the solar circle reduced to roughly four or five, possibly reinforcing a four arm spiral.  Whether these expectations are borne out in some form by 3-dimensional modeling of the arm pattern remains to be seen. Some alteration is expected as waves which are not entirely vertical will have somewhat different periods and finite wavelengths parallel to the plane. The basic idea, however, is that the spiral arm ``shock'' is fragmented, feeding energy preferentially into over dense regions resulting from waves provoked by other spiral arm fragments upstream (and slightly off to the side). That energy input drives further oscillations to provide dense regions further downstream yet, where the next shock fragments occur. As these shock fragments are not oriented parallel to the arm locus, significantly stronger interactions can occur than if their strength were determined only by the component of the entry velocity normal to that more oblique locus.  An individual large disturbance in the disk leads eventually to a shock wave propagating up into the low-density outer parts of the atmosphere. This shock strengthens as it proceeds, finally reaching such low density that it is no longer radiative on the timescale for propagating through the rest of the atmosphere. A hot low-density layer is produced, on top of the cool disk material. With no further disturbance, this layer slowly cools and rejoins the disk, but the timescale is extraordinarily long. If the time between such disturbances is longer than this cooling timescale, then the hot halo may become a more or less permanent feature of the atmosphere. Similarly, when low amplitude waves are generated continuously within the disk and propagate into its outer reaches, they too can heat an outer layer to high temperatures, eventually reaching a new equilibrium with the unreflected portion of the wave power. We found that there is a strong resonance at which the temperature and size of the outer layer are greatly enhanced, leading to temperatures in excess of 10$^6$\\ K and a thickness of many kiloparsecs. In our experiments, the period of the resonance was that of the fundamental breathing mode for the slightly altered disk, essentially the same oscillation invoked to provide structuring in the spiral arm pattern. It appears that the two situations are linked. If the spiral waves are sufficiently large in amplitude to cause substantial vertical motion in rebounding from their compressions, vertical bounces and subsequent oscillations will be produced that can both structure the details of the arms themselves and lead to substantial heating of material at the outer disk. The latter effect produces a large and very hot gas layer, potentially contributing to the populations of high stage ions (O~VI in particular) and to the soft X-ray background. The details of the hot halo depend on the flux of driving mechanical energy, but quite an impressive halo was formed in our simulation with a very small power, on the order of 4$\\times $10$^{-9}$\\ erg cm$^{-2}$ s$^{-1}$, which is less that 2\\% of that available from spiral wave disturbances.  Our experiments on hot halo formation were all performed with a nominal disk boundary height of 5 kpc. The resulting haloes have very low densities, with thermal pressures more than two orders of magnitude below the total midplane pressure. We estimate that their O~VI column densities would be of order 10$ ^{13}$\\ cm$^{-2}$. With a lower boundary height, the halo pressure and O~VI column densities could be significantly higher, though much more power would be required to sustain them.  Surely the details of these results will change when the studies are repeated in two and three dimensions, when MHD effects are included explicitly rather than just through an effective pressure term, when self gravity and spiral wave behavior are included, and perhaps again when the driving terms associated with correlated star formation are incorporated. But the essential effects are likely to persist. What goes up must come down. Material nearer the plane returns sooner, hence producing the ``falling sky.'' Velocities associated with small disturbances in the intercloud medium will produce spectral features, appearing to be clouds which in reality do not exist. Spiral arm shocks will provoke vertical motions with periods that will lead to structure within the spiral waves themselves, and at periods that will tend to promote the formation of a hot halo overlying the cool disk. It is conceivable that the spiral waves might also, on occasion, produce such large upflows that the falling sky is an associated phenomenon. We refer the reader once again to Martos \\& Cox (1998) for further discussion.  \\begin{appendix} \\begin{center} \\bf \\appendixname{: Linear Analysis} \\rm \\end{center}"
    },
    "0010/astro-ph0010400_arXiv.txt": {
        "abstract": "The absence of radio pulsars with long periods has lead to the popular notion of a high \\teq{P} ``death line.''  In the standard picture, beyond this boundary, pulsars with low spin rates cannot accelerate particles above the stellar surface to high enough energies to initiate pair cascades, and the pair creation needed for radio emission is strongly suppressed.  In this paper we explore the possibility of another pulsar ``death line'' in the context of polar cap models, corresponding to high magnetic fields B in the upper portion of the period-period derivative diagram, a domain where few radio pulsars are observed.  The origin of this high B boundary, which may occur when B becomes comparable to or exceeds $B_{\\rm cr} = 4.4 \\times 10^{13}$ Gauss, is also due to the suppression of magnetic pair creation, but primarily because of ineffective competition with magnetic photon splitting.  Threshold pair creation also plays a prominent role in the suppression of cascades.  We present Monte Carlo calculations of the pair yields in photon splitting/pair cascades which show that, in the absence of scattering effects, pair production is effectively suppressed, but only if all three modes of photon splitting allowed by QED are operating in high fields.  This paper describes the probable shape and position of the new ``death line,'' above which pulsars are expected to be radio quiet, but perhaps still X-ray and gamma-ray bright.  The hypothesized existence of radio-quiet sources finds dramatic support in the recent discovery of ultra-strong fields in Soft Gamma-ray Repeaters and Anomalous X-ray Pulsars.  Guidelines for moderate to high B pulsar searches at radio wavelengths and also in the soft and hard gamma-ray bands are presented. ",
        "introduction": "\\label{sec:intro}  Most current theories for the generation of coherent radio emission in pulsar magnetospheres (see summations in Michel 1991; Melrose 1993; Lyutikov et al. 1999) require formation of an electron-positron pair plasma.  Such a pair plasma has been shown to develop via electromagnetic cascades along the open magnetic field lines as a result of particle acceleration to TeV energies, followed by curvature, synchrotron radiation and one-photon pair production in the strong magnetic field, \\teq{\\gamma\\to e^+e^-}, near the neutron star surface (e.g. Sturrock 1971; Ruderman \\& Sutherland 1975, Arons \\& Scharlemann 1979).  Numerical simulations of pair cascades above a pulsar polar cap show that \\teq{10^3-10^4} pairs are produced for each accelerated primary electron (Daugherty \\& Harding 1982) in a typical young pulsar.  But if any of the necessary conditions for development of a pair cascade, {\\it i.e.} production of $\\gamma$-ray photons (requiring particle acceleration to high energy) or production of sufficient pairs, is absent the pulsar should not, according to theory, emit detectable radio emission.  The existence of an observed ``death line'' in the \\teq{P-\\dot P} distribution of known radio pulsars (see Figure~\\ref{fig:quiescence} below), to the right of which no pulsars were detected, until the recent discovery (Young, Manchester \\& Johnston 1999) of the 8.5 second pulsar PSR J2144-3933, provided some circumstantial evidence that pairs are required for radio emission. The slope of this putative boundary fits a line of constant voltage across the open field lines, \\teq{V \\propto P^{-3/2} \\dot P^{1/2}}, suggesting that radio emission ceases when the particle acceleration is not high enough to sustain pair production.  The radio emission from some long period pulsars like PSR J2144-3933 can be explained through production of pairs by photons from inverse-Compton scattering (Arons 2000, Zhang, Harding \\& Muslimov 2000), a process which does not require as high a voltage as does curvature radiation.  In this paper, we study another area of pulsar phase space, that of very high magnetic fields, where the character of pair cascades is significantly altered by the process of photon splitting and other effects such as ground state pair creation and positronium formation.  Magnetic photon splitting, \\teq{\\gamma\\to\\gamma\\gamma}, a QED process in which a single photon splits into two lower-energy photons (Adler 1971, Baring \\& Harding 1997), operates efficiently and competes effectively with pulsar pair production only in magnetic fields above \\teq{\\sim 10^{13}}Gauss (Harding, Baring \\& Gonthier 1997, hereafter HBG97). This region of high magnetic field strength lies in the upper-right part of the \\teq{P-\\dot P} diagram.  We will explore whether photon splitting may be the explanation for why there are no classical radio pulsars detected with derived magnetic fields above \\teq{\\sim 10^{14}}Gauss. Our initial results (Baring \\& Harding 1998) have suggested that photon splitting could indeed explain the absence of very high-field radio pulsars.  This paper considers in more detail the underlying assumptions made in that calculation, explores the physics of high-field pair cascades and makes predictions for high-energy searches for high-field pulsars.  Although pulsars in this region of phase space may be radio-quiet, those that are to the left of the death line are still prolifically accelerating particles to high energies. Their pulsed emission may therefore be detectable in the X-ray and gamma-ray wavebands, making this relatively-unexplored region of pulsar phase space extremely interesting.  There has recently been growing evidence for another class of pulsars with ultra-strong magnetic fields.  Observations at X-ray energies have yielded detections of both long periods and high period derivatives in two types of sources, anomalous X-ray pulsars (AXPs) and soft $\\gamma$-ray repeaters (SGRs), which suggest dipole spin-down fields in the range \\teq{10^{14} - 10^{15}} Gauss.  The AXPs are a group of seven or eight pulsating X-ray sources with periods in the range 6--12 seconds, and are continuously spinning down (Vasisht \\& Gotthelf 1997).  The SGRs are a type of $\\gamma$-ray transient source that undergoes repeated bursts; four (possibly now five: see Cline et al. 2000 for SGR 1801-23) are currently known to exist.  A detailed discussion of these sources and recent discoveries pertaining to them is presented in Section~\\ref{sec:magnetars}.  While conventional radio pulsars tap their rotational energy to power their emission, if such a new class of ultra-magnetized neutron stars or ``{\\it magnetars}'' does exist, their long periods indicate that their emission cannot be rotationally-powered: they may instead be driven by their magnetic energy.   With the possible exception of the recent report (Shitov 1999; Shitov, Pugachev \\& Kutuzov 2000) of a low frequency pulsed detection of a counterpart (PSR J1907+0919 at 111 MHz) to the soft gamma repeater SGR 1900+14, which has not been confirmed by pulsation searches in Aricebo data at higher frequencies (Lorimer \\& Xilouris 2000) none of these sources has detectable pulsed radio emission.  In several previous papers, we have computed the photon splitting attenuation lengths of photons emitted parallel, or at small angles, to the magnetic field near the surface of strongly magnetized neutron stars and compared them to the attenuation lengths for one-photon pair production (Harding, Baring \\& Gonthier 1996; 1997).  Both processes depend on the photon energy \\teq{\\omega} and the angle \\teq{\\theta_{\\rm kB}} of photon propagation to the local field.  However, since photon splitting has no threshold, it may occur before the photon reaches the threshold for pair production at \\teq{\\omega \\sin\\theta_{\\rm kB} = 2m_ec^2}.  We found that the photon splitting attenuation lengths are lower than those for pair production in fields higher than \\teq{\\sim 3 \\times 10^{13}}Gauss, for emission in the open field region near the neutron star surface.  One must therefore incorporate photon splitting in cascade models of highly-magnetized pulsars.  This has been done in the case of PSR1509-58 (Harding, Baring \\& Gonthier 1997), which has a field of \\teq{3\\times 10^{13}}Gauss as estimated using Equation~(\\ref{eq:spindown}).  We considered both the case where all {\\it linear} polarization modes of photon splitting allowed by CP invariance operate, and the case where only the \\teq{\\perp \\rightarrow \\parallel\\parallel} mode (our polarization convention is defined near the beginning of Section~\\ref{sec:suppression}) allowed by selection rules (Adler 1971) in the weakly dispersive limit operates.  The latter case, we argued, was most appropriate for pulsars.  In both cases, we found that photon splitting attenuation is a plausible explanation for the very low spectral cutoff between 10 and 30 MeV in PSR1509-58 (Kuiper et al. 2000).  Since the rates of these attenuation mechanisms are very sensitive to the polarization states of the incoming photons, it is evident that the polarization properties of the radiated photons are very important in modeling both the output spectrum and pair suppression in high-field pulsars.  In ultra-strong magnetic fields, the vacuum dispersion increases to the point where the selection rules derived for weak linear dispersion (appropriate for \\teq{B\\lesssim B_{\\rm cr}}) may no longer be valid, and higher order non-linear contributions to such dispersion may be significant.  We presently do not know which modes of photon splitting operate in supercritical fields and thus do not know which photon polarization modes are allowed to split.  We will therefore explore several possibilities for photon splitting mode behavior in our study of high-magnetic field pair cascades.  In Section~\\ref{sec:suppsplit}, we discuss the physics of pair suppression in high magnetic fields, both by photon splitting and by pair production in low-lying Landau states, as well as the possible alternative mechanism of pair suppression by bound-state pair creation, discussed by Usov \\& Melrose (1996).  In Section~\\ref{sec:noradio}, we present results of detailed simulations of splitting/pair cascades which explore the conditions for suppression of pair creation in a pulsar magnetosphere.  We compute the boundary in the pulsar \\teq{P-\\dot P} distribution where the escape energies for photon splitting and magnetic pair production for photons of \\teq{\\perp} polarization are equal.  This condition then defines a putative radio-quiescence boundary, posited in Baring \\& Harding (1998), above which pulsars may not produce the dense pair plasmas required for radio emission.  For reasonable assumptions about the location of particle acceleration and the angles of the emitted photons in high-field pulsars, the computed radio-quiescence boundary lies at \\teq{\\dot P} above those in the known radio pulsar population, potentially explaining the absence of radio pulsars with fields above \\teq{\\sim 10^{14}}Gauss.  Our cascade simulations indicate that such a boundary is viable only if photons of both polarizations can split; otherwise pair creation is generally postponed a generation rather than suppressed.  In the latter case, significant reductions in pair production can ensue only if the maximum energy of primary photons is below or not much above the pair creation escape energy.  If pair suppression is rife at these high \\teq{\\dot P}, it suggests the possible existence of a new class of radio-quiet, ultra-magnetized pulsars that may be related to the emerging class of magnetars. ",
        "conclusions": "\\label{sec:conclusion}  We have presented a detailed study of the comparative attenuation of photons by magnetic pair production and photon splitting, of pair cascades and of conditions for pair suppression in very highly magnetized pulsars.  While ground-state pair creation and positronium formation can act at \\teq{B_0\\gtrsim 0.1B_{\\rm cr}} to reduce the number of {\\it free} pairs, only photon splitting has the potential capability of dramatically inhibiting the creation of pairs, bound or free.  Our quantitative study of cascade pair yields at different field strengths confirms the location of the approximate radio quiescence boundary in the \\teq{P}-\\teq{\\dot P} diagram proposed by Baring \\& Harding (1998), assuming that copious pair production is a requirement for pulsar radio emission.  However, it also shows that the existence of such a boundary requires the rate of photon splitting to be non-zero for both photon polarization modes.  This is not believed to be true in the low-field, weakly dispersive limit, where kinematic selection rules prohibit splitting of photons of \\teq{\\parallel} polarization.  Whether photon splitting can significantly suppress pairs in highly magnetized radio pulsars depends on presently unexplored physics.  It therefore has become critically important to study the photon splitting process in the high-field, highly dispersive regime.  Understanding the behavior of splitting in this regime will not only resolve the radio quiescence question, but is crucial to modeling acceleration and radiation in high-field pulsars and magnetars.  The circumstantial evidence, that the most highly magnetized pulsars (AXPs and SGRs) tend to be radio quiet, argues for some pair suppression at high field strengths.  The radio pulsar and magnetar populations are no longer cleanly separated in \\teq{P}-\\teq{\\dot P} space, which indicates that another dimension of phase space is necessary to divide these two source populations.  Delineation of these two source groups is complicated by the likelihood of alternate spin-down mechanisms and evolution issues for magnetars.  More systematic surveys of the overlap region of \\teq{P}-\\teq{\\dot P} space at X-ray and $\\gamma$-ray energies, correlated with radio pulsar surveys are needed.  We have given some guidelines for such high-energy searches based on the expected physics of particle acceleration and photon attenuation, in anticipation of the next generation of medium and high-energy gamma-ray experiments such as GLAST.  Spectral and polarization observations may be crucial in identifying signatures of photon splitting and even in resolving the issue of which splitting modes are operating in ultra-magnetized sources, an attractive possibility for both astrophysicists and physicists."
    },
    "0010/astro-ph0010546_arXiv.txt": {
        "abstract": "We study the optical counterpart of the black-hole X-ray binary LMC X$-$3, by using {\\it XMM-Newton/OM} observations carried out during a low-hard X-ray state. We derive a better constraint for the temperature, mass and radius of the companion star, and we show that the star is likely to be a $\\sim$ B5 subgiant filling its Roche lobe. Taking into account the effect of X-ray irradiation, we suggest a value $f_{\\rm M} \\approx 1.5 M_{\\odot}$ for the mass function in this system, lower than previously thought; we provide a more accurate lower limit to the mass of the compact object. ",
        "introduction": "An important unsolved problem for the black-hole candidate (BHC) LMC X$-$3 is the process of mass transfer. With an estimated mass of the companion star $4 M_{\\odot} \\simlt M_2 \\simlt 8 M_{\\odot}$ (Cowley et al.\\ 1983), the system appears to be intermediate between high-mass black-hole binaries like Cyg X$-$1 (mass of the companion star $\\approx 33 M_{\\odot}$, see Giles \\& Bolton 1986), and low-mass black-hole binaries like A0620$-$00 (mass of the companion star $\\approx 0.7 M_{\\odot}$). In the former class of systems, mass transfer occurs mainly via a stellar wind, and the donor star is more massive than the primary; in the latter, the donor star is usually a late-type star filling its Roche lobe.  The optical counterpart (Warren and Penfold 1975) of LMC X$-$3 shows ellipsoidal modulations with a total amplitude of $\\simlt 0.2$ mag and a period corresponding to the 1.705d binary period (van der Klis, Tjemkes and van Paradijs 1983; van der Klis et al 1985). It also shows long-term brightness variations in the range $16.7 \\simlt V \\simlt 17.5$ (e.g., van Paradijs et al.\\ 1987). Its optical spectrum was found to be consistent with that of a B3 main sequence star (Cowley et al.\\ 1983). From the radial velocity shifts of the optical absorption lines, Cowley et al.\\ (1983) derived a mass function $f_{\\rm M} = 2.3 M_{\\odot}$, thus establishing this system as a strong BHC. However, the spectral identification of the companion star remains uncertain (Mazeh et al.\\ 1986) because of the effect of X-ray irradiation on the secondary, and the possible contribution to the optical flux by an accretion disk. In fact, the long-term variations in both the optical colours and the $V$ brightness are found to be associated with changes in the soft X-ray flux (Cowley et al.\\ 1991).  Moreover, Nowak et al.\\ (2000) and Wilms et al.\\ (2000) suggest that spectral transitions between a high-soft and a low-hard state could be associated with changes in the accretion rate.  A more correct identification of the companion star can provide a better understanding of the physics of mass transfer and state transitions. ",
        "conclusions": "We have observed the BHC LMC X$-$3 with the Optical Monitor on board {\\it{XMM-Newton}} during an X-ray low-hard state. The $2$--$10$ keV flux at that epoch was the lowest ever observed by RXTE/ASM, consistent with zero. The brightness and colours of the optical counterpart inferred from our observations are likely to be a good approximation of the intrinsic values of the companion star. This allows us to constrain the mass and radius of the secondary. We have found that an evolved subgiant of mass $M_2 \\approx 4.7 M_{\\odot}$ and temperature $T_{\\rm eff} \\approx 16500$ K (spectral type $\\sim$ B5\\,IV) is consistent with the observed colours and luminosity. We have also shown that such a star would be filling its Roche lobe, thus explaining why the source is mostly observed in a high-soft state, dominated by an accretion disk. No significant wind is expected from such star, in agreement with the low column density inferred from the {\\it XMM-Newton/EPIC} and {\\it RGS} X-ray data. The evolutionary time-scale of a B5\\,IV star is also consistent with the observed X-ray luminosity powered by nuclear-evolution driven mass transfer.  The companion star was previously thought to be a main-sequence B3 star. Although it would have the same brightness, a B3\\,V star would fill only about one half of its Roche lobe, ruling out mass transfer via Roche-lobe overflow.  A B5\\,IV companion implies a lower limit to the mass of the primary $M_{\\rm X} > 7.3 \\pm 0.6 M_{\\odot}$, if we assume the $K$-velocity and mass function determined by Cowley et al.\\ (1983). However, the spectroscopic observations of Cowley et al.\\ (1983) were carried out during an X-ray active state, when the effect of soft X-ray irradiation on the surface of the secondary star is significant. The true radial velocity amplitude of the centre of mass of the secondary is probably $\\approx 12$--$15$ per cent less than the value observed by Cowley et al.\\ (1983). This would reduce the mass function to $f_{\\rm M} = 1.5 \\pm 0.3 M_{\\sun}$, and set a lower limit $M_{\\rm X} > 5.8 \\pm 0.6 M_{\\odot}$ for the mass of the compact object."
    },
    "0010/astro-ph0010287_arXiv.txt": {
        "abstract": "We examine the light-cone effect on the two-point correlation functions using numerical simulations for the first time. Specifically, we generate several sets of dark matter particle distributions on the light-cone up to $z=0.4$ and $z=2$ over the field-of-view of $\\pi$ degree$^2$ from cosmological N-body simulations. Then we apply the selection function to the dark matter distribution according to the galaxy and QSO luminosity functions. Finally we compute the two-point correlation functions on the light-cone both in real and in redshift spaces using the pair-count estimator and compare with the theoretical predictions.  We find that the previous theoretical modeling for nonlinear gravitational evolution, linear and nonlinear redshift-distortion, and the light-cone effect including the selection function is in good agreement with our numerical results, and thus is an accurate and reliable description of the clustering in the universe on the light-cone. ",
        "introduction": "In the proper understanding of on-going redshift surveys of galaxies and quasars, in particular the Two-degree Field (2dF) and the Sloan Digital Sky Survey (SDSS), it is essential to establish a theory of cosmological statistics on the light cone. This project has been undertaken in a series of our previous papers (Matsubara, Suto, \\& Szapudi 1997; Yamamoto \\& Suto 1999; Nishioka \\& Yamamoto 1999; Suto et~al. 1999; Yamamoto, Nishioka, \\& Suto 2000; Suto, Magira \\& Yamamoto 2000). Those papers have formulated the light-cone statistics in a rigorous manner, described approximations to model the clustering evolution in the redshift space, and presented various predictions in canonical cold dark matter (CDM) universes.  Their predictions, however, have not yet been tested quantitatively, for instance, against numerical simulations. This is not surprising since it is fairly a demanding task to construct a reliable sample extending over the light-cone from the conventional simulation outputs at a specified redshift, $z$.  In the present paper, we examine, for the first time, the validity and limitation of the above theoretical framework to describe the cosmological light-cone effect against the mock catalogues on the light-cone. Such catalogues from cosmological $N$-body simulations have been originally constructed for the study of the weak lensing statistics (Hamana et al. 2000, in preparation). Applying the same technique (\\S\\ref{subsec:lcfromnbody}), we generate a number of different realizations for the light-cone samples up to $z=0.4$ and $z=2$, evaluate the two-point correlation functions directly, and compare with the theoretical predictions. ",
        "conclusions": "We have presented detailed comparison between the theoretical modeling and the direct numerical results of the two-point correlation functions on the light-cone. In short, we have quantitatively shown that the previous theoretical models by Yamamoto \\& Suto (1999) and Yamamoto, Nishioka \\& Suto (1999) are quite accurate on scales $1 h^{-1} {\\rm Mpc} < x < 20 h^{-1} {\\rm Mpc}$ where the numerical simulations are reliable. It is also encouraging that this conclusion remains true even for the particle number of around $10^4$.  In fact, the error-bars in our estimates of the two-point correlation functions are dominated by the sample-to-sample variance due to the limited angular-size ($\\pi$-degree$^2$) and thus the limited volume.  In order for the more realistic evaluation of the statistical and systematic uncertainties, one needs mock light-cone data samples with a much wider sky coverage.  More importantly such datasets enable one to access the effect of biasing on the two-point correlation functions on the light-cone. Since our present study indicated that all the physical effects except for the biasing are well described by the existing theoretical models, it is very interesting to examine in detail how to extract the effect of the galaxy/QSO biasing from the upcoming redshift survey on the basis of the above mock samples.  We plan to come back to these issues with larger simulation datasets in near future."
    },
    "0010/astro-ph0010414_arXiv.txt": {
        "abstract": "The first positive detection of the X-ray background fluctuations at small angular scales is reported. {\\it ROSAT} PSPC archive pointed observations are used to measure fluctuations at scales of $0\\fdg03 - 0\\fdg4$. The pointings have been selected from an area free from galactic contamination. At separations below $\\sim 0\\fdg1$ clusters of galaxies become a substantial source of the background fluctuations. The autocorrelation function of the fluctuations in the power law approximation has a slope of $\\sim 1$ for all the data but is substantially flatter (with slope of $\\sim 0.7$) when pointings containing bright clusters are removed. At separations $0\\fdg3-0\\fdg4$ where the ACF estimates based on the ROSAT pointings and All-Sky Survey are available, both data sets give consistent results. ",
        "introduction": "Fluctuations of the X-ray background (XRB) provide a unique method to investigate statistical properties of the distribution of X-ray sources. At angular scales above $\\sim 0\\fdg3$ the {\\it ROSAT} All-Sky Survey (RASS) revealed distinct variations of the soft XRB (Soltan et al.\\ 1996). Subsequent analysis (Soltan et al.\\ 1999) showed that clustering of active galactic nuclei (which are the main contributor to the XRB) potentially could explain the large amplitude of the XRB autocorrelation function (ACF). However, the amplitude of X-ray source spatial clustering required to reproduce the observed ACF is higher than that derived from direct investigation of the distribution of sources (e.g.\\ Carrera et al.\\ 1998). Thus, it is likely that fluctuations of the soft XRB are produced also by some other effects. In particular, hydrodynamical computations by Cen \\& Ostriker (1999) show that the process of accumulation of primordial gas in the potential wells created by galaxies and clusters of galaxies produced large scale concentrations of hot plasma. This material is expected to emit thermal bremsstrahlung in the soft X-ray domain. A clumpy distribution of emitting plasma would contribute substantially to the total fluctuations of the XRB. To investigate the importance of different mechanisms producing the XRB anisotropy one needs to measure the amplitude of the XRB fluctuations over a wide range of angular separations and at different energies. In the present work we have determined the autocorrelation function of the soft XRB using a large number of {\\it ROSAT} pointings. Below we describe the observational material used in the analysis, the computational details and compare the present results with our earlier measurements of the ACF at large separations using the {\\it ROSAT} All-Sky Survey. ",
        "conclusions": ""
    },
    "0010/astro-ph0010308_arXiv.txt": {
        "abstract": "We report multiband observations of the Optical Transient (OT) associated with GRB000301c carried out between 2--4 March 2000 using the 2.34-m Vainu Bappu Telescope (VBT) at Kavalur, India. When combined with other reported data, the initial decline in the R-band magnitude with log ($t-t_0$), the time since the burst is fit with a slope $\\alpha_{1}$ = -0.70 $\\pm$ 0.07 which steepens after about 6.0 days to a slope of $\\alpha_{2}$ = -2.44 $\\pm$ 0.29. This change in slope does not occur smoothly but there is an indication for a bimodal distribution. The available measurements of the evolution of (B--R) color do not show any discernible evolution in the first 12 days. ",
        "introduction": "Ever since in a pioneering effort, Costa (1997) and Van Paradijs (1997) detected the first ever counterpart, associated with GRB970228, identification of the optical counterparts of Gamma Ray Bursts (GRB) and rapid follow-up observations in other wavelength bands have given impetus to GRB astronomy during the last 3 years. The fading counterpart of a GRB, also known as 'Afterglow', is generally monitored down to the faintest detection limit of a telescope to study its brightness variation as a function of time since the occurence of the burst event, to derive the decay law from the light curve and to try to locate precisely the burst counterpart with respect to the host galaxy, if any has been identified. A study of all the afterglows so far detected shows a diversity of light curve properties and it is necessary to accumulate a statistically significant set of such observations for the classification of the bursts and for subsequent theoretical modelling. Two excellent reviews by Kulkarni et al. (2000) and by Piran (1999) provide a clear picture of the current status of this rapidly evolving field of astronomy.  GRB000301c was detected by All Sky Monitor (ASM) on board {\\it Rossi X-Ray Timing Explorer (RXTE) } on 2000, March 1.4108UT and also by two other spacecrafts {\\it Ulysses} and {\\it NEAR}. The burst had a single peak with slow decay structure lasting 10 seconds and was localized by an error box of area 50 sq. arcmin centred at (J2000.) RA=$16^{h}20^{m}21^{s}.5$, DEC=+29$^{\\circ}24^{\\prime}56^{\\prime \\prime}.37$ (Smith, Hurley \\& Cline 2000). The Optical counterpart was first reported by Fynbo et al.(2000a) and the redshift has been measured to be z=1.95 $\\pm$ 0.1 by Smette et al.(2000) using HST and to be z=2.03 $\\pm$ 0.003 by Castro et al. (2000) using the Keck-II telescope. The OT was visible at R=27.9$\\pm$0.15 from the late-time HST imaging on 19 Apr 2000 (Fruchter et al. 2000a), but there was no evidence for a host galaxy underlying the GRB to a magnitude $\\sim$ 28.5.  In the next section we present the details of our observations, data analysis and results. In section 3 we discuss briefly our conclusions. ",
        "conclusions": "In the preceeding sections we presented our observations and analysis of GRB000301c afterglow observations using Vainu Bappu Telescope, Kavalur, India. In order to place our observations in the context of other observations in the R-band we have collected the data available through GCN circulars and those given by Jensen et al. (2000) in Table.~2. who reported the most extensive coverage of the event. Following the standard practice in combining the photometric data obtained by different groups and different instruments, we renormalize all measurements to that of Jensen et al. (2000). The corrections applied to various data sets is shown in Table.~ 3. Notice the correction for our observations at VBT is $<$ 0.01m and falls within the uncertainty in the observations.   We attempt to fit the data on intensities in R-band to a time evolution defined by Eq.~1.  \\begin{eqnarray} I(t)&=&I(t_1) (t/t_1)^{\\alpha_1} \\quad \\quad \\hbox{for} \\> \\,t<t_1 \\nonumber \\\\ I(t)&=&gI(t_1)(t/t_1)^{\\alpha_1}+(1-g)I(t_1)(t/t_1)^{\\alpha_2}\\,\\,\\hbox{for} \\,\\> t_{1}\\le t <t_{2}\\nonumber \\\\ I(t)&=&\\{g(t_2/t_1)^{\\alpha_1}+(1-g)(t_2/t_1)^{\\alpha_2}\\} I(t_1)(t/t_2) ^{\\alpha_2}\\,\\, \\hbox{for}\\> \\, t\\ge t_2 \\end{eqnarray} This form is motivated by the feature observed in the R-band light curve at $t-t_{0} \\simeq 4.5 \\, \\rm days$. This  might have been generated by a major burst followed after a short interval by a minor burst, each being represented by a power-law form with a slope $\\alpha_1$ which at later times steepens to a slope $\\alpha_2$. The function given in Eq.~(1) is designed to test this hypothesis. Notice that this function has 6 parameters $I(t_1)$, g, $t_1$, $t_2$, $\\alpha_1$ and $\\alpha_{2}$, two more than the functional form adopted by Jensen et al. (2000) The values of the parameters and their $1\\sigma$ errors which provide the best fit to the data along with the normalized $\\chi^{2}$ are computed using 'Levenberg-Marquardt method' (Press et al. 1992) and are listed in Table.~4. Notice that our fit for very early and very late times coincides quite closely with the fit obtained by Jensen et al. (2000) and has a $\\chi^2_{65} =1.80$. Even though Jensen et al. (2000) obtain with a single break an excellent fit to their own data, the single-break form fails to reproduce the compilation of world data which have a slightly closer coverage around the break point at 4--5 days after the burst; the $\\chi^2$ they quote for the fit to the world data is 3.687 for $\\nu = 88$. After applying the corrections given in Table.~3 to the data in Table~2. we fit the single-break form in Eq.~1 of Jensen et al. (2000) and find that the $\\chi^{2}_{64}$ reduces to 1.95 from its value of 2.85 for the same data points when no magnitude corrections are applied. Thus with suitable corrections for normalizing different photometric data the $\\chi^{2}$ reduces to 1.8 ($\\nu=65$) for two-break form and to 1.95 ($\\nu =64$) for single-break form. The improvement on the $\\chi^{2}$ with double-break is significant at 38\\% level as seen from F-distribution test, although we can not rule out the fact that the high $\\chi^2$ may be due to a genuine variations due to inhomogeneities in the medium surrounding the GRB host.   The motivation for the function given in Eq.~(1) is clearly seen in Fig.~1. The OT is fit with two events each of the type where a power-law decay sharply steepens into another power-law several days after the burst. The light curve in the K-band reported by Rhoads \\& Fruchter (2000) agrees broadly with the fit we have obtained here for the first break (Eq~1.). If their curve is simply extrapolated and compared with the data obtained by Jensen et al. (2000) in the later and fainter epochs then one may find an apparent color evolution. More accurate photometry with better coverage is needed to confirm and characterize the possible existence of secondary events which may be expected in some supernova-shock models (Meszaros, Rees and Weijers 1998). In case of GRB980228 the re-brightening of the afterglow $\\sim$3 weeks after the burst event and the reddening of the spectrum have been hypothesized (Bloom et al. 1999) to be due to an underlying SN explosion  which triggered from the energy released by a 'Collapsar' (MacFadyen \\& Woosely 1999) that gave rise to the initial GRB event. Reichart(1999) also explained the color evolution in case of GRB980228 using the supernova hypothesis.   To check if there are any such color changes during the evolution of GRB000301c we display in Fig.~2. the behaviour of the light curve both in the B and in the R bands. It is seen that the evolution of R and B magnitudes are nearly the same within the photometric uncertainties. This result has been reported by Jensen et al. (2000) and Masetti et al. (2000) with a smaller sample, although Rhoads \\& Fruchter (2000) finds shift in R--K' color towards blue. Running a F-test (Press et al. 1992) gives $90\\%$ probability that the two distributions are equal. The color B--R can be fit with a function $ (B-R)=a(t-t_0)^{n}$ where, $a$ and $n$ are the two fitted parameters. Our best fit values are: $a =0.793\\pm0.073$ and the slope $n=0.073 \\pm0.074$ with a reduced $\\chi^{2} =1.3$. This is consistent with the achromatic behaviour of the B--R evolution. Hence chromatic behaviour is not established on the basis of B--R data and the suggestions of color variations are based on comparison of R-evolution with the K data by Rhoads \\& Fruchter (2000).  As a closing remark to this brief paper we would like to add that Vainu Bappu Observatory and the Uttar Pradesh State Observatory being located at a longitude of $\\sim 78^{\\circ}$E in India could make the earliest set of observations of the afterglow of GRB000301c and that excellent observational opportunities have opened up at longitude $78^{\\circ}57^{\\prime}51^{\\prime\\prime}$E, latitude $32^{\\circ}46^{\\prime}46^{\\prime\\prime}$N  with the commissioning of a 2-m telescope at Hanle ($\\sim$15000ft) in September 2000."
    },
    "0010/astro-ph0010622_arXiv.txt": {
        "abstract": "s{ I critically review a methodology of using clusters of galaxies as cosmological probes.  The understanding of the abundances and spatial correlations of dark matter halos has been significantly advanced especially for a last few years. Nevertheless such dark matter halos are not necessarily identical to clusters of galaxies in the universe. This is a quite obvious fact, but it seems that the resulting systematic errors, on the cosmological parameter estimates for instance, are often neglected in the literature.} ",
        "introduction": " ",
        "conclusions": "Here I presented a brief review of cosmological implications of galaxy clusters, specifically considering the Sunyaev-Zel'dovich effect and the cluster abundance in some detail.  I believe that the clusters are quite important and useful probes of cosmology and in fact they already proved to be successful in many respects. If one would like to go further and to extract more stringent constraints on the cosmological models, however, the one-to-one correspondence between virialized halos and observed clusters, whatever they mean, should be critically examined.  This assumption is a reasonable working hypothesis, but we need more quantitative justification or modification in order to improve the {\\it cosmology with clusters}."
    },
    "0010/astro-ph0010144_arXiv.txt": {
        "abstract": "In interacting and merging galaxies, gas is subject to direct hydrodynamic effects as well as tidal forces. One consequence of interactions is the rapid inflows of gas which may fuel starbursts and AGN. But gas dynamics is not limited to inflows; a small survey of equal-mass and unequal-mass encounters produces a wide variety of features, including plumes between galaxies, extended disks formed by infall of tidal debris, and counterrotating nuclear disks.  An even richer spectrum of behavior awaits better thermodynamic models for gas in merging galaxies. ",
        "introduction": "It has long been recognized that interstellar material plays an important role in galaxy interactions.  Spitzer \\& Baade (1951) described fast collisions of galaxies as essentially hydrodynamic affairs.  While emphasizing that stellar dynamics governs the outcome of galaxy encounters, Toomre \\& Toomre (1972) anticipated more than a decade of numerical work in suggesting that interactions `bring {\\it deep\\/} into a galaxy a fairly {\\it sudden\\/} supply of fresh fuel'.  The mechanics of this fueling process seem fairly clear.  Negroponte \\& White (1983) found central concentrations of gas in simulated mergers of gas-rich disk galaxies, but the low resolution of their models left the physics uncertain.  Noguchi (1987, 1988) and Combes, Dupraz, \\& Gerin (1990) showed that tidal encounters trigger bar formation in stellar disks and that {\\it gravitational\\/} torques exerted by these bars extract angular momentum from the gas. Hernquist (1989) reported rapid gas inflows in minor mergers where a disk galaxy swallows a small companion; in these experiments the gravitational torques driving the gas inward were generated by both the stellar disk and the companion.  Barnes \\& Hernquist (1991, 1996) modeled major mergers between disk galaxies.  This work extended the gravitational inflow picture to show how orbital decay could deliver $\\sim 5 \\times 10^9 {\\rm\\,M_\\odot}$ of gas within the central $100 {\\rm\\,pc}$ of a merger remnant; such massive gas clouds are common in the central regions of infrared-selected galaxies, many of which are merger remnants (Sanders \\& Mirabel 1996).  Hernquist \\& Barnes (1991) simulated the formation of a counterrotating gas disk in the aftermath of a major merger.  Misaligned or counterrotating disks are found in both merger remnants (Schweizer 1982) and elliptical galaxies (e.g.~Bender 1990), strengthening the hypotheses that merging disk galaxies form elliptical galaxies (Toomre \\& Toomre 1972) and that the centralized starbursts which power luminous infrared galaxies represent the formation of the cores of elliptical galaxies (e.g.~Kormendy \\& Sanders 1992).  Existing simulations are still incomplete in one key respect: they don't really capture the {\\it thermo\\/}dynamics of the gas.  Early studies used `sticky particles' to mimic effects of dissipation. Hernquist and co-workers adopted `smoothed particle hydrodynamics' (SPH) techniques which can include radiative heating and cooling.  But in practice, radiative cooling is usually cut off below $10^4 {\\rm\\,K}$ to prevent catastrophic instabilities, and most of the gas remains near this cut-off temperature throughout the simulation. Consequently, simulations including radiative processes are all but indistinguishable from those using an isothermal equation of state (Barnes \\& Hernquist 1996).  At this point there are two options.  One is to adopt the isothermal gas model and survey a range of encounters.  The other is to try to develop models which represent the thermodynamics of a multi-phase ISM.  Below I discuss both. ",
        "conclusions": ""
    },
    "0010/astro-ph0010372_arXiv.txt": {
        "abstract": "We present an investigation of the dynamical state of the cluster A262.  Existing optical line of sight velocities for select cluster galaxies have been augmented by new data obtained with the Automated Multi-Object Spectrograph at Lick Observatory.  We find evidence for a virialized early-type population distinct from a late-type population infalling from the Pisces-Perseus supercluster ridge. We also report on a tertiary population of low luminosity galaxies whose velocity dispersion distinguishes them from both the early and late-type galaxies.  We supplement our investigation with an analysis of archival X-ray data.  A temperature is determined using {\\it ASCA} GIS data and a gas profile is derived from {\\it ROSAT} HRI data.  The increased statistics of our sample results in a picture of A262 with significant differences from earlier work.  A previously proposed solution to the ``$\\beta$--problem'' in A262 in which the gas temperature is significantly higher than the galaxy temperature is shown to result from using too low a velocity dispersion for the early-type galaxies. Our data present a consistent picture of A262 in which there is no ``$\\beta$--problem'', and the gas and galaxy temperature are roughly comparable. There is no longer any requirement for extensive galaxy-gas feedback to drastically overheat the gas with respect to the galaxies. We also demonstrate that entropy-floor models can explain the recent discovery that the $\\beta$ values determined by cluster gas and the cluster core radii are correlated. ",
        "introduction": "\\label{intro}  Clusters of galaxies provide an opportunity to study the formation of massive structures through gravitational accretion. The cluster dark matter potential well is an environment in which infalling and virialized galaxies interact with intra-cluster gas in a complex, time-dependent fashion involving dynamical friction with dark matter as well as galactic winds and ram-pressure stripping.  Measurements of cluster galaxies and gas temperatures as a function of radius, element abundances, luminosity-temperature relations and the evolution of these quantities with redshift all provide data to test theories of cluster formation and evolution.  Clusters also provide a probe of many cosmological parameters.  The density of clusters forming as a function of redshift can be used to measure quantities such as \\omega0 and the spectrum of density perturbations in the early universe.  One problem which arises in exploring the topics described above is the $\\beta$--problem. Assuming that gas and galaxies are both tracers of the dark matter potential well, and that no source of energy other than gravity is operative, there is a relation between the ratio of the internal energy of the galaxies and the gas temperature (determined using spectroscopic data and generally termed \\bspec) to a quantity related to the gas and galaxy density profiles (determined by fitting isophotes to the X-ray gas and optical emission and generally termed \\bfit).  The failure of these two quantities to agree in most clusters was noted some time ago \\citep{sar86}. Numerous solutions have been offered over the years.  These include adjustments of \\bfit to properly account for the galaxy distribution as indicated by galaxy-galaxy correlation measurements \\citep{bah94}; galaxy velocity distribution asymmetries \\citep{men86}; incorrect determinations of \\bspec due to contamination of velocity dispersions by cluster substructure \\citep{edg91b}; and presence of significant galaxy-gas feedback or dynamical friction with dark matter \\citep{bir95}. \\citet{dav96} have proposed that the $\\beta$--problem is solved by excluding late-type galaxies when estimating the cluster velocity dispersion since such galaxies are still infalling and therefore not virialized.  The $\\beta$--problem is significant for several reasons.  The measurement of $\\beta$ from X-ray and optical data provides information about the nature of galaxy-gas feedback. It is complementary to other methods used to probe the galaxy-gas interaction such as heavy element abundance gradients. The $\\beta$--problem will become increasingly important as more accurate cluster temperatures become available from {\\it Chandra} and {\\it XMM}.  With improved temperature measurements, $\\beta$ uncertainty, which appears in the conversion from gas temperature to mass in the Press-Schechter formula \\citep{pre74}, can contribute significantly to the mass uncertainty. It is essential to confirm that the gas dynamic-dark matter simulations are correct, since they are utilized to obtain $\\beta$.  In fact, the comparison of $\\beta$ values derived from modeling with those obtained from optical observations indicate marginal agreement given the observational and theoretical uncertainties \\citep{hen97}.  Many problems can lead to a cluster violating the assumptions required for \\bspec to equal \\bfit.  These problems include use of non-virialized galaxy populations, use of galaxies associated with infall or other peculiarities in the velocity field, local sub-clustering, presence of cooling flows or magnetic fields etc..  There are several approaches to evaluating the possibilities.  The first approach is to observe many clusters, with the number of spectra collected per cluster generally limited to $\\about20$ or less, and to statistically analyze the resultant data.  With adequate temperatures and substructure analyses the calculation of ``average'' cluster properties may then represent a ``typical'' member of the sample population, with any scatter being random.  As a complement to this statistical approach, an analysis of individual clusters through use of extensive X-ray and optical data permits correction for many effects and thus provides an alternative testbed for comparison with both theory and the results of the statistical approach. Such detailed analyses are rare because they require much high quality optical and X-ray data.  We are obtaining large numbers of redshifts for several clusters for eventual comparison with {\\it XMM} and {\\it Chandra} data.  In this paper we describe a detailed investigation of A262.  We augmented existing galaxy redshift data with a sample of new redshifts obtained with the Automated Multi-Object Spectrograph at Lick Observatory. We have also analyzed previously unpublished {\\it ROSAT} and {\\it ASCA} data.  The complete data set allows us to address the problems discussed above. ",
        "conclusions": "\\label{conclusions}  1) The early-- and late-type galaxy populations in A262 both have isothermal VDPs.  There is no evidence that the early-type galaxies have a much lower velocity dispersion than the late-type galaxies.  The velocity dispersion of the early populations is 569 \\kms, much higher than the 330 \\kms\\  used in a previous solution to the $\\beta$--problem, and based on a sample of only 9 elliptical galaxies.  The main population of late-type galaxies do differ from the early-type galaxies in having a non-Gaussian velocity dispersion, a different heliocentric velocity and a spatially extended profile.  2) The"
    },
    "0010/astro-ph0010658_arXiv.txt": {
        "abstract": "{\tWe present detailed spatially-resolved spectroscopy results of the observation of Abell 1835 using the European Photon Imaging Cameras (EPIC) and the Reflection Grating Spectrometers (RGS) on the XMM-Newton observatory. Abell 1835 is a luminous ($10^{46}$ ergs $\\mbox{s}^{-1}$), medium redshift ($z=0.2523$), X-ray emitting cluster of galaxies.  The observations support the interpretation that large amounts of cool gas are present in a multi-phase medium surrounded by a hot ($kT_{e}$=8.2 keV) outer envelope.  We detect O VIII Ly$\\alpha$ and two Fe XXIV complexes in the RGS spectrum.  The emission measure of the cool gas below $kT_{e}$=2.7 keV is much lower than expected from standard cooling-flow models, suggesting either a more complicated cooling process than simple isobaric radiative cooling or differential cold absorption of the cooler gas. ",
        "introduction": "\\begin{figure} \\resizebox{8.5cm}{!}{\\rotatebox{0}{\\includegraphics{smoothimage.ps}}} \\caption{Smoothed logarithmic intensity EPIC-MOS image of Abell 1835. The image is approximately 23 by 13 arcminutes.} \\label{fig:f1} \\end{figure}   Previous soft X-ray imaging and medium spectral resolution studies have implied the presence of cool gas near the centers of clusters (See e.g. Sarazin \\cite{Sarazin88}, Fabian \\cite{Fabian94a}).  For the Centaurus, Perseus, Virgo, and Coma clusters there is direct spectroscopic evidence for cool gas from the {\\it Einstein} Focal Plane Crystal Spectrometer (Canizares et al. \\cite{Can79}, \\cite{Can82}, Mushotzky \\& Szymkowiak \\cite{Mushot88}).  Based on simple, robust estimates for the electron densities and temperatures, it is estimated that in the large majority of clusters, gas in the central volume should cool radiatively to very low temperatures on a timescale short compared to the characteristic age of the cluster. Due to the consequent lack of central pressure support, gas has to continuously flow inward into the cluster potential.  This picture of cluster cooling flows is incomplete in that the fate of the cooling gas is unknown. Material that has cooled to below X-ray emitting temperatures generally fails to be detected by emission at any other wavelengths, although more recently there are indications that cold gas may be distributed throughout the cooling flow. This gas betrays its presence through X-ray absorption of the cooling flow emission (but see below for evidence that such absorption is not confirmed at higher spectral resolution).  A simple single phase model for the cooling gas based on the conservation laws for the global flow predicts a sharply peaked density distribution that is not consistent with the observed distributions, and this fact has given rise to the idea of a diffuse, multi-phase cooling flow (Johnstone et al.\\cite{Johnstone92}).  Abell 1835 is a prototypical strong cooling-flow cluster with an inferred mass deposition rate of $1750 \\pm 550$ (Allen et al. \\cite{Allen96}).  The central cooling time (the time it would take the gas to radiate away all of its kinetic energy) is less than 1 Gyr, whereas the cluster could remain relatively undisturbed by mergers for $\\sim$ 10 Gyr.  So roughly the inner 350 kpc of the cluster should have had sufficient time to cool below X-ray temperatures. Allen et al. (\\cite{Allen96}) found Abell 1835 to be one of the strongest cooling-flows from the deprojection of ROSAT HRI images.  The ASCA spectrum was consistent with a $kT_e=8.5~\\mbox{keV}$ collisionally ionized plasma, but cooling-flow spectral models were unconstrained (Allen et al. \\cite{Allen96}). Allen and Fabian (\\cite{Allen98}) also carried out a joint ROSAT-ASCA study and found evidence for a massive cooling flow, as well as a relatively high metallicity.  Until this observation, cooling-flow properties have largely been inferred by relatively short cooling times.  Here we present spectroscopic results on the properties of the cool gas.  As part of the XMM-Newton (Jansen et al \\cite{Jansen00}) performance verification program, Abell 1835 was observed for 60 ks. Data were obtained simultaneously with the two Reflection Grating Spectrometers (RGS) and the three European Photon Imaging Cameras (EPIC).  The RGS is capable of achieving high spectral resolution for the entire soft wavelength band (5 to 38 $\\mbox{\\mbox{\\AA}}$) for moderately extended X-ray sources.  The spectral resolution for sources of characteristic angular size, $\\theta$, is roughly, $\\Delta \\lambda \\approx 0.1 ~\\mbox{\\AA}~(\\theta / 1^{\\prime}) $. The high throughput of the XMM mirrors allows the EPIC-pn and EPIC-MOS detectors to perform detailed spectral-spatial studies of clusters since the spatial extent of most clusters is resolved. Both sets of instruments can be used together to produce a spectral-spatial model of the X-ray emission.  Throughout this paper we assume $H_0= 70~\\mbox{km}~\\mbox{s}^{-1}~\\mbox{Mpc}^{-1}$, $\\Omega_m = 0.3$, and $\\Omega_{\\Lambda}=0.7$.  This implies an angular distance of 1280 Mpc and a luminosity distance of 1600 Mpc for Abell 1835.  One arcsecond corresponds to 6.2 kpc. Errors are quoted at the $90\\%$ confidence level and systematic errors are added in quadrature whenever they can be estimated. ",
        "conclusions": "We describe the  spectral-spatial results of the observation of Abell 1835 using the European Photon Imaging Cameras and the Reflection Grating Spectrometers on the XMM-Newton observatory. The observations are consistent with large amounts of cool gas surrounded by a $\\mbox{kT}_e$=8.2 $\\mbox{keV}$ component outer envelope.  We find no evidence for excess uniform cold absorption above the galactic value of $2.32\\times10^{20} \\mbox{cm}^{-2}$.  The emission measure of cool gas below $\\mbox{kT}_e$=2.7 $\\mbox{keV}$ is much lower than expected in simple isobaric cooling flow models, and it indicates either differential cold absorption applied to the cool gas or that a more complicated description of the cooling process is required."
    },
    "0010/astro-ph0010002_arXiv.txt": {
        "abstract": "We present an in-depth weak lensing analysis of the cluster MS1008${}.{}$1$-$1224 based on deep multicolor imaging obtained during the Science Verification of FORS1 at the VLT. The image quality (half arcsec seeing) and depth of the VLT images allow the shear signal to be mapped with high signal-to-noise and to be traced out to $1.2 \\, h_{50}^{-1} \\mbox{ Mpc}$, near the edge of the $6\\farcm 8 \\times 6\\farcm 8$ field of view.  Using BVRI color information, as well as 81 redshifts in the field from the CNOC survey, background galaxies can be effectively separated from cluster and foreground objects. PSF distorsions are found to be moderate across the FORS images and thus easily removed. Due to the small statistical errors in the mass reconstruction, this dataset provides a testing ground where several systematic effects (e.g.\\ mass-sheet degeneracy, redshift distribution of the background sources, cluster galaxy contamination), which are involved in the weak lensing analysis, can be quantified. Several methods are used to remove the mass-sheet degeneracy which is found to dominate the systematic error budget. We measure a lower limit to the mass of $2.6 \\times 10^{14} \\, h_{50}^{-1} \\mbox{ M}_\\odot$ within $1 \\, h_{50}^{-1} \\mbox{ Mpc}$ and a ``total'' mass of $5.3\\times 10^{14} \\, h_{50}^{-1} \\mbox{ M}_\\odot$ by fitting a softened isothermal sphere.  We find the mass distribution fairly uniform, with no significant substructures, in agreement with the virial analysis.  The availability of the CNOC redshift data and X-ray observations on this cluster allow a comparison of different determinations of the mass radial profile. We find the lensing and X-ray measurements in excellent agreement, while the mass derived from the virial analysis is marginally ($1$--$2 \\sigma$) in agreement at radii where both methods are reliable.  This analysis underscores the importance of systematics in the mass determination of clusters, particularly when such a high quality dataset is not available or in similar studies at higher redshifts. ",
        "introduction": "The study of the weak lensing distortion of background galaxies is a powerful tool for measuring the mass distribution of galaxy clusters. It has long been recognized (Tyson et al.\\ 1984) that the tidal field of a cluster acts by slightly modifying the images of distant, background galaxies.  By measuring these small distortions the projected mass distribution of the lensing cluster can be derived in a model independent fashion (Kaiser \\& Squires 1993), in contrast with other mass estimators (X-ray measurements of intra-cluster gas density and temperature, galaxy dynamics).  To date, there are approximately 20 clusters which have been well studied with lensing techniques (see, e.g., Mellier 1999 for a review). In several cases the resulting masses are claimed to be larger (by a factor of two) than those obtained using different methods. At present, it is not known if this discrepancy is due to biases and inaccuracies in the X-ray and dynamical estimates, or to the lensing data reduction method. As a matter of fact, it is not even clear whether there is a real discrepancy (see Wu et al.\\ 1998; see also Allen 1998).  For example, Boehringer et al.\\ (1998) have shown that different mass estimators are in excellent agreement when very good quality X-ray and optical data are used.  In this paper, we study the mass distribution of the cluster MS1008${}.{}$1$-$1224 (in the following, for simplicity, MS1008) using high quality imaging data obtained during the Science Verification of FORS1 at the VLT. The depth and seeing quality of these observations in four different bands (BVRI), make this dataset ideal for weak lensing mass reconstruction.  The large surface density of background galaxies in these images, as well as the sharpness of the PSF and its small variations across the field, significantly improve the accuracy in measuring the distortion field from object ellipticities.  As a result, high signal-to-noise shear maps can be obtained. This allows us to better investigate the effect of systematics involved in the mass inversion procedure, which in this case exceed statistical errors. Moreover, the availability of radial velocities for $65$ galaxies in MS1008 from the CNOC survey (Carlberg et al. 1996), as well as X-ray spectroscopic observations, provide independent estimates of the total mass.  The paper is organized as follows. In Sect.~2 we introduce the basic lensing relations and the notation used throughout the paper. The observations and the imaging dataset are briefly described in Sect.~3. Section~4 is devoted to the weak lensing analysis of the cluster, where we describe in detail the mass reconstruction method and the results obtained. A virial analysis based on the CNOC redshift survey of MS1008 is provided in Sect.~5. A comparison between the light and the mass distribution of MS1008, and in particular its mass to light ratio, are discussed in Sect.~6. Finally, we draw the conclusions of this study in Sect.~7.  An independent lensing study using the same dataset has recently been presented by Athreya et al.\\ (2000); moreover, the depletion effect on this cluster has been recently investigated by Mayen \\& Soucail (2000, preprint). We briefly compare our results with these studies in the conclusions.  Throughout the paper we adopt $\\Omega = 0.3$, $\\Omega_\\Lambda = 0.7$, $H_0 = 50\\; h_{50} \\mbox{ km s}^{-1} \\mbox{ Mpc}^{-1}$. ",
        "conclusions": "We have studied the weak lensing mass distribution of the cluster MS1008 using multicolor imaging data obtained during the Science Verification of FORS1 on the VLT. The depth, angular resolution, and image quality across the entire field of view make these images a rare and ideal dataset for weak lensing analysis. In addition, the information on radial velocities of a large set of cluster galaxies provided by the CNOC survey allows the mass of this cluster to be derived with dynamical methods and compared with that reconstructed from the weak lensing. These combined datasets allowed us to conduct a detailed study of the systematic effects involved in cluster mass estimates.  We have found PSF distortions of FORS1 to be moderate and hence easily removed with a low order polynomial fit. The B through I multicolor information, as well as the redshift information from the CNOC catalog, have allowed us to efficiently separate the background galaxies from the cluster and foreground populations. Approximately 8000 objects have been detected, corresponding to $200 \\mbox{ sources arcmin}^{-2}$. Further selection lead to approximatively $1700$ background galaxies with high signal-to-noise ratio which have been used for the determination of lensing shear field. We have detected a weak lensing signal out to $r \\simeq 1.2 h_{50}^{-1}\\mbox{ Mpc}$.  The projected mass within $1 h^{-1}_{50} \\mbox{ Mpc}$ is measured in the range $(2.5\\div 3.2)10^{14} h_{50}^{-1}\\mbox{ M}_\\odot$ in the four filters, with typical statistical errors (deriving from uncertainties in galaxy ellipticities) of 5\\% .  We have discussed the impact of systematics in the weak lensing reconstruction of the mass distribution in physical units, which can be summarized as follows. The removal of the mass-sheet degeneracy is inevitably model-dependent, inducing mass variations of $\\sim\\!  15\\%$ at $r \\approx 1 h^{-1}_{50} \\mbox{ Mpc}$. [Note that the estimate of this error is difficult and somewhat arbitrary since the error clearly depend on the ``range'' of possible models allowed. As a result, the estimate of the total error, given below, is also bound to be inaccurate.  We stress, however, that a similar problem exists for X-ray or virial mass estimate (cf., e.g., the different predictions using King-like or NFW profiles if Fig.~\\ref{fig:mass_vir}).]\\@ By polluting the background galaxy population with a significant fraction of cluster galaxies, the total mass is biased low by $\\mincir 5\\%$. However, different selections of background galaxies, as well as edge effects due to masked portions of the images, do affect the morphology of the shear maps.  We have not found any evidence for a substructure of the cluster core; this in part is due to lack of resolution in the mass reconstruction, set by the smoothing scale used.  However, we have found the central region of the cluster to be elongated NS, with some evidence of a displacement of the projected mass peak respect to the cD galaxy, similarly to the X-ray emission.  Cosmic variance is expected to affect the assumed redshift distribution of the background field galaxies, and hence the derived cluster mass.  From a comparison between the Hubble Deep Field North and South, cosmic variance would introduce a systematic error of approximatively $10\\%$ in the mass. {\\mdf This can also be taken as a rough estimate of the error due to cosmic variance.  Note also the main contribution to this error is from large scale structures rather than from Poisson noise in the HDF-N and in our field.  In other words, a simple error estimate which assumes that galaxies are uncorrelated would lead to an underestimate of the error.} Altogether, the cluster light distribution traces the mass distribution remarkably well. We have measured a mass-to-light ratio of $M/L_\\mathrm{V} = (95\\pm 22) \\, h_{50} \\mbox{ M}_\\odot / \\mbox{L}_\\odot$ at radii $\\ge 0.5 h_{50}^{-1}$ Mpc.  We have also compared the weak lensing mass profile with that derived from a virial analysis of the CNOC redshift data, as well as with the X-ray mass.  The mass derived from X-ray observations (Lewis et al.\\ 1999) is found in excellent agreement with our weak lensing determination at $r \\simeq 0.8 h^{-1}_{50} \\mbox{ Mpc}$.  Different approaches to estimate the cluster mass at the virialization scale produce very similar results.  At scales $0.1 \\,\\mincir h_{50} r \\,\\mincir 1$ Mpc, lensing and virial mass determinations can be compared.  The virial mass reconstruction at these small radii critically depends on the assumed density profile.  Using a King-like profile describing the galaxy distribution of local clusters (G98; cf. also Eq.~\\eqref{eq:king}), we have obtained a mass which is more than $2 \\sigma$ away from the X-ray and weak-lensing mass. Assuming instead a NFW profile, as fitted by Carlberg et al.\\ (1997) to CNOC clusters, a much better agreement is obtained over the entire overlapping region.  Note that the discrepancy at small radii ($r < 1'$) between lensing and virial estimate can be explained by recalling that the lensing determination is affected by a Gaussian smoothing of $30''$, as well as a departure from the weak lensing approximation. The former effect leads to underestimate the mass in the core, while the latter will generally bias the central mass high. Both effects clearly vanish at large radii. In addition, we notice that the weak lensing data alone cannot discriminate between different mass profiles (e.g.\\ NFW versus isothermal model).  This analysis shows that the combination of depth and good imaging quality provided by VLT/FORS allows us to obtain high $S/N$ mass maps via optimized weak lensing reconstruction methods. On the other hand, these data have the virtue of revealing important systematic errors in weak lensing analyses. We find that the mass-sheet degeneracy dominates the budget of systematic errors and ultimately makes mass measurements via weak lensing model dependent.  The results obtained here are, at a first look, in good agreement with the mass reconstruction performed by Athreya et al.\\ (2000). However, a detailed analysis of their study reveals that a different recipe to remove the mass-sheet degeneracy was used. Specifically, they set to zero the mass at large radii ($3\\farcm2$--$3\\farcm6$). Had they used our (model-dependent) method to break the mass-sheet degeneracy, their estimate would have been a factor of two higher. Therefore, there is a discrepancy of the same factor between the two analysis which, at present, we are unable to explain."
    },
    "0010/astro-ph0010234_arXiv.txt": {
        "abstract": "We have measured the spectral properties of five galaxies of low to intermediate L$\\rm _x/L_b$ ratios with \\bsax\\ and ASCA.  A hard component (kT $\\sim 4-10$ keV) is observed in all galaxies.  In NGC~1553 the \\bsax\\ data show that this component is extended, and suggest an origin for the emission in the evolved stellar population. In NGC~3115, a point-like source appears embedded in an extended component morphologically similar to the stellar body,  suggesting an almost equal contribution from the nuclear region and the binary population.  A large central mass concentration and low level optical activity in NGC~3379 argue for a contribution from the nucleus also in this object.  However, for both NGC~3115 and NGC~3379, the nuclear emission is at a level well below that observed in other galaxies who host similar nuclear black holes. A second soft (kT $\\sim$0.3- 0.7 keV) spectral component is needed to fit the data of NGC~1407, NGC~1553 and NGC~4125 over the entire energy range probed  (\\about 0.2-10 keV), best represented by a thermal component with line emission. We discuss possible interpretations of the origin of this component, which however will be better defined only with higher quality data. ",
        "introduction": "\\label{Intro}   One of the observational evidences from X--ray data of early type galaxies is the large scatter in the correlation between the X--ray and optical luminosities (Fabbiano, Kim and Trinchieri 1992): at a given optical luminosity, the X--ray to optical flux ratios L$\\rm _x/L_b$ may range from those observed in the bulge of M31 and in spiral galaxies, to values $\\sim$50--100 times higher.  The presence of at least two components, hot gas dominating at high X--ray luminosities, and the evolved stellar population at faint X--ray luminosities, can explain in part this observational evidence.  The spectral characteristics of the emission also reflect the relative importance of these components: already from the limited $Einstein$ spectral data, the average emission temperature is larger in galaxies with lower L$\\rm _x/L_b$ (Kim et al. 1992), in agreement with the idea of an increasing contribution of hard, individual X--ray sources relative to the hot gas component as the L$\\rm _x/L_b$ ratios decrease. An additional very soft component was also found in the lowest L$\\rm _x/L_b$ class, possibly due to a $\\sim$0.2--0.4~keV interstellar medium or to the collective emission of stellar sources (Pellegrini \\& Fabbiano 1994).  \\begin{table*} \\caption[] { General galaxy properties} \\begin{flushleft} \\begin{tabular}{ l l  l l  l  l l  l l  l } \\noalign{\\smallskip} \\hline \\noalign{\\smallskip} Name & Type$^a$ & $B_{\\rm T}^0\\, ^a$ &Size $^a$  &  d$\\, ^b$ & log$L_{\\rm B}\\, ^c$ & $\\sigma \\, ^d$ &  group$^e$        & group$^e$  \\\\ &          &  (mag)             &(arcmins)  &  (Mpc)    & ($L_{\\odot}$)       & (km s$^{-1}$)  &  ($Einstein$) & ($ROSAT$)  \\\\ \\noalign{\\smallskip} \\hline \\noalign{\\smallskip} \\hline \\noalign{\\smallskip} NGC1407 & E0 & 10.71 & $4.6\\times 4.3$ &34.7 & 10.99 &  272 & 3 & 3  \\\\ NGC1553 & S0 & 10.26 & $4.5\\times 2.8$ &21.5 & 10.75 &  200 & 3 & 2  \\\\ NGC3115 & S0 &  9.74 & $7.2\\times 2.5$ &10.8 & 10.36 &  264 & 1 & 1  \\\\ NGC3379 & E1 & 10.18 & $5.4\\times 4.8$ &13.0 & 10.35 &  209 & 1 & --  \\\\ NGC4125 & E6 & 10.67 & $5.8\\times 3.2$ &38.0 & 11.08 &  232 & 1-2 & 2  \\\\ \\noalign{\\smallskip} \\hline \\end{tabular} \\end{flushleft} \\bigskip  $^a$ from de Vaucouleurs et al. 1991. $B_{\\rm T}^0$ is the total B magnitude, corrected for Galactic and internal extinction; the size gives the apparent major and minor axes diameters at the surface brightness level of 25 mag/square arcseconds.  $^b$ distance from Fabbiano et al. (1992), who adopt a Hubble constant of 50 km s$^{-1}$ Mpc$^{-1}$.  $^c$ total B-band luminosity $L_B$, derived using the indicated distance and $B_{\\rm T}^0$.  $^d$ central stellar velocity dispersion from McElroy (1995).  $^e$ Group into which the galaxy has been classified on the basis of its L(0.2--4 keV)/L$_{\\rm B}$ ratio (for $Einstein$ data, Kim et al. 1992), or its L(0.5--2.0 keV)/L$_{\\rm B}$ ratio (for $ROSAT$ data, Irwin \\& Sarazin 1998). The X-ray faintest galaxies belong to group 1 while the X-ray brightest to group 4. The group boundaries differ in the two papers; in both works they have been chosen so that each group contains roughly the same number of galaxies of the same general spectral characteristics.   \\label{gen} \\end{table*}  The ROSAT PSPC and ASCA data confirm the \\ein\\ results: a soft (kT$\\sim$0.5-1~keV) optically thin emission due to hot gas dominates in X-ray bright objects, and an harder component, with kT $\\ge 3-4$ keV, most likely associated with the evolved stellar population (Matsushita et al. 1994; Matsumoto et al. 1997; Buote 1999), or to nuclear activity (Allen, Di Matteo \\& Fabian 2000; Matsumoto et al.  1997) and present in all galaxies, dominates in X-ray faint objects. In the lowest L$\\rm _x/L_b$ class, a very soft component is also measured, with a temperature now  well constrained to $\\sim$0.2-0.3 keV (Fabbiano et al 1994; Pellegrini 1994; Fabbiano \\& Schweizer 1995; Kim et al 1996).  However, the origin of the very soft component has not been properly understood yet: it could be placed in stellar sources, in X-ray binaries (Irwin \\& Sarazin 1998), or could be a cooler phase of the ISM: as detailed hydrodynamical simulations show (Pellegrini \\& Fabbiano 1994), hot gas with the required emission temperature and luminosity can be retained by the galaxies.  However, the hot gas temperature is comparable to that of the very soft component only in galaxies with quite shallow potential wells, i.e., with central stellar velocity dispersion $\\sigma \\leq 200$ km s$^{-1}$.    More galaxies with intermediate to low L$\\rm _x/L_b$ ratios still need to be investigated. We report here the results on the \\bsax\\ and ASCA observations of 5 such galaxies.  Their general optical properties are summarized in Table~\\ref{gen} together with the $\\rm L_X / L_B$ group in which they fall according to their soft band X--ray fluxes. The wide energy band and the good spectral resolution of both satellites are suitable to detect and measure separately the amount and spectral parameters of the different emission components.  Moreover, the spatial resolution of \\bsax\\ at high energies allows us to study the  spatial characteristics of the hard component, to determine its extent. We discuss the observations and data analysis in \\S 2-6, we briefly compare the spectral results with those reported in the literature from previous missions and/or different authors in \\S 7,  and we discuss the results in \\S 9.   In \\S 8 we further discuss a detection at very high energy of a source in the field of NGC~1553, but most likely unrelated to this galaxy.   \\begin{table} \\caption[]{Log of the Observations} \\bigskip  \\begin{flushleft} \\begin{tabular}{llrr} \\hline Object&Instrum  & Obs.  Dates & On Time    \\\\ && Beginning - End &\\multicolumn{1}{c}{(s)} \\\\ \\hline NGC 1407&GIS2&23/08/95 - 24/08/95&34759  \\\\ &GIS3&23/08/95 - 24/08/95 &34755  \\\\ &SIS0&23/08/95 - 24/08/95 &33801\\\\ &SIS1&23/08/95 - 23/08/95 &33747\\\\ \\hline \\hline NGC 1553& MECS1& 16/01/97 - 17/01/97& 25880 \\\\ &MECS2&16/01/97 - 17/01/97& 25849 \\\\ &&16/11/97 - 17/11/97& 30019 \\\\ &MECS3& 16/01/97 - 17/01/97& 25664 \\\\ &&16/11/97 - 17/11/97&  29962 \\\\ &LECS & 16/01/97 - 17/01/97& 14262 \\\\ &     & 16/11/97 - 17/11/97&18763 \\\\ \\hline \\hline NGC 3115&MECS2 & 15/05/98 - 16/05/98 & 35986 \\\\ &MECS3 & 15/05/98 -  16/05/98 & 35952 \\\\ &MECS2 &    26/05/98 -   27/05/98 &36594 \\\\ &MECS3 &    26/05/98 -   27/05/98 &36543\\\\ &LECS &15/05/98 - 16/05/98 &14997 \\\\ &LECS &26/05/98 - 27/05/98&13177 \\\\ \\hline \\hline NGC 3379 &MECS2 & 14/12/98 - 17/12/98 & 99122 \\\\ &MECS3 &    14/12/98 - 17/12/98 &98749 \\\\ &LECS & 14/12/98 - 17/12/98 &34594 \\\\ \\hline \\hline NGC 4125& MECS1& 26/04/97 - 27/04/97& 57128 \\\\ &MECS2 &  26/04/97 - 27/04/97& 57148\\\\ & MECS3 &  26/04/97 - 27/04/97& 56971 \\\\ & LECS &  26/04/97 - 27/04/97& 22189 \\\\ \\hline &GIS2&05/04/95 - 05/04/95 &38634  \\\\ &GIS3&05/04/95 - 05/04/95&38634  \\\\ &SIS0&05/04/95 - 05/04/95&30073\\\\ &SIS1&05/04/95 - 05/04/95 &21505 \\\\ \\end{tabular} \\end{flushleft} \\label{log} \\end{table} ",
        "conclusions": "The spatial analysis on the MECS data indicates that the distribution of the hard emission in the four galaxies observed with \\bsax\\ is different in different cases:  NGC~1553 is clearly extended, while NGC~3379 is most likely dominated by a nuclear source.  A significant contribution from a point source cannot be excluded by the data of NGC~3115, although a more extended component is also observed.  The data of NGC~4125 are not conclusive.  For three sources (NGC~1407, NGC~1553 and NGC~4125) the data require a two component spectrum, with a soft and a hard component,  although neither can be  determined with high accuracy. The soft component is well parameterized by a thin plasma spectrum, with kT$\\sim 0.3- 0.9$ keV, and in all three cases its contribution is comparable or higher than that of the hard component in the soft energy band.  At hard energies, the extended nature of the emission in NGC~1553 ensures that a nuclear source, if present, does not contribute significantly to the observed luminosity. Both NGC~3115 and NGC~3379 data instead could be parameterized by a power law spectrum, consistent with a significant contribution from a nuclear source.  An intriguing result is provided by the lack of a very soft component in NGC~3115 and NGC~3379, commonly found in the lowest $\\rm L_x/L_b$ galaxies  (see discussion in \\S~\\ref{Intro}). While the poor statistics does not allow us to speculate on whether this is a ``significant\" result, in the sense of singling out these two galaxies for their lack of a very soft component, the existence of a nuclear component probably further complicates matters.  In fact, the soft band ROSAT data of either galaxies do not show a significant contribution from extended components: the galaxies appear rather compact both in the PSPC image of NGC 3115 and in the HRI image of NGC~3379 (cf. Robert \\& Warwick 2000), suggesting that the nuclear source dominates at all energies.  We have estimated the contribution from a soft component allowed by the present data.  As discussed before, one component is adequate, and in fact the addition of a second component to either set of data does not give any additional information: the second component either mimics the parameters of the first component or has a zero normalization.  To estimate a possible contribution from a soft component, we have therefore fixed the power law parameters at the best fit values and added a $raymond$ with kT = 0.3 keV.  We find that in either cases a soft component is allowed, and would have an unabsorbed flux $f\\rm _x(0.1-2 keV) \\sim 1-3 \\times 10^{-13}$ \\ergcms (NGC~3115 and NGC~3379 respectively).  We will discuss this further in \\S~\\ref{discuss}."
    },
    "0010/astro-ph0010090_arXiv.txt": {
        "abstract": "For a simple spherically symmetric mean-field dynamo model we investigate the possibility of determining the radial dependence of the coefficient $\\alpha$. Growth rates for different magnetic field modes are assumed to  be known by measurement. An evolutionary strategy (ES) is used for the solution of the inverse problem. Numerically, we find quite different $\\alpha$-profiles giving nearly the same eigenvalues. The ES is also applied to find functions $\\alpha(r)$ yielding zero growth rates for the lowest four magnetic field modes. Additionally, a slight modification of the ES is utilized for an ''energetic'' optimization of $\\alpha^2$-dynamos. The consequences of our findings for inverse dynamo theory and for the design of future dynamo experiments are discussed. ",
        "introduction": "At the very beginning, the question of the origin of cosmic magnetic fields was formulated as an inverse problem. ''How could a rotating body such as the sun  become a magnet?'' was the question Larmor put in 1919 (Larmor 1919), i.e. he asked for the unknown cause of an evident effect. In the last decades, dynamo theory saw an enormous progress. Testing a large number of kinematic models in the sense of a forward problem, the inverse character of the original problem was partly  pushed into the background. It seems to be only a few percent of  papers on the dynamo topic which are dealing with an inverse problem in the strict sense.  In geophysics, there is a long tradition to reconstruct the tangential components of the velocity field at the core-mantle boundary from geomagnetic secular variations (see, e.g, Bloxham (1989), and references therein). But the value of this type of inverse frozen-flux modeling was called into question recently (Love 1999). The main argument of the criticism was that the removal of the diffusion term from the induction equation changes its differential order resulting possibly in a drastic change of the nature of the solutions.  Another approach to the inverse dynamo problem was formulated by Love and  Gubbins (1996). In this paper, the eigenvalue equation was complemented by some regularizing functionals in order to optimize dynamo models with respect to energetic demands and smoothness properties of the arising magnetic field. However, geomagnetic observations were taken into account only in the sense that some smoothness properties of the model magnetic field were compared with those of the observed field and not in the sense of fitting the model to the actual data.  With the advent of the first successful hydromagnetic dynamo experiments (Gailitis et al. 2000; M\\\"uller and Stieglitz 2000) quite new perspectives for inverse dynamo theory come into play. Compared to cosmic bodies, much more measurement techniques and measurable quantities can be taken into account. One additional measurable quantity is the electric potential at the boundary of the dynamo module which can provide complementary information  in addition to the magnetic field on the outside. Further on, in contrast to the case of cosmic bodies it is possible to apply various magnetic fields from outside and to determine the response of the dynamo on these primary fields. Those date can be collected in the regime below the critical magnetic Reynolds number ($Rm$) in order to predict the dynamo behaviour at larger $Rm$.  In connection with this, but up to now restricted to small $Rm$, there are some new developments concerning the determination of the velocity field from externally measured induced magnetic fields and electric potentials when primary magnetic fields are applied (Stefani and Gerbeth 1999, 2000a, 2000b). For a static homogeneous primary field and spherical geometry it was shown (Stefani and Gerbeth 2000a) that the defining scalars of the velocity can be determined except for an ambiguity of the radial dependence of their spherical harmonics expansion coefficients. Although one can try to get a reasonable guess of the unknown radial dependence by some regularization techniques, in the strict sense this dependence remains unobservable. A generalization of this method to higher $Rm$ would amount to a combination of an eigenvalue solver (similar to the philosophy of Love and Gubbins) with a data-fitting procedure. However this approach might look like in detail, we guess that the radial dependence will remain undetectable. Roughly speaking, two two-dimensional measured quantities (radial component of magnetic field plus electric potential) allow only to determine two two-dimensional desired quantities. Actually, this radial ambiguity was the first reason for the following investigation. At least for the small $Rm$ regime it seems clear that the necessary third dimension to determine the radial dependence can only be the time (or frequency).  Whereas a general inverse dynamo theory is still missing, it is worth to have a sidelong glance at the quite similar inverse problem in quantum mechanics. There is a huge literature on inverse scattering theory and inverse spectral theory (for an overview, see Chadan and Sabatier (1977) and Chadan et al. (1997)), and evidently many methods  from quantum mechanics (Gelfand-Levitan equation, WKB methods, operator methods, etc.) remain to be adapted and utilized  in inverse dynamo theory.  In the present paper, we will pursue a less ambitious program restricting ourselves to a pragmatic treatment of a simple inverse spectral dynamo problem. It is a simple problem as it concerns spherically symmetric $\\alpha^2$-dynamos in which, as a slight modification of the original  model of Krause and Steenbeck (1967), the radial dependence of $\\alpha$ is assumed to be unknown. The treatment will be pragmatic in the sense that we suppose eigenvalues of only a few magnetic field modes to be \"measured\" and that we will try to infer the radial dependence of $\\alpha$ from this restricted set of information.  Basically,  it is our aim to get some acquaintance with the peculiarities if one tries to get information on the radial distribution of the source of dynamo action. Reminding quantum theory might be useful  in the following respect: The existence of isospectral potentials  is a well established fact in quantum theory. For example, it is possible to deform {\\it{continuously}} the harmonic oscillator potential in such a way that all the eigenvalues are the same as for the original quadratic potential (De Lange and Raab 1991). Therefore, we should be aware of the possibility that any inversion procedure of any set (even of an infinite number) of measured eigenvalues might lead to a variety of radial $\\alpha$-distributions which cannot be discriminated using only their eigenvalues. For the corresponding quantum inverse problem it is known that a potential can be uniquely determined if the spectra {\\it{for two different boundary conditions}} were known. A similar behaviour should also be expected for the inverse dynamo problem, but this seems to be irrelevant for any realistic measurement strategy.   Our method might have application for cosmic dynamos as well as for laboratory dynamos. Of course, a spherically symmetric $\\alpha^2$-dynamo model is surely  too simple to explain cosmic dynamos but the general method is easily extendable to more realistic  models.  The paper is structured as follows: in the second section, the used eigenvalue solver is presented and five paradigmatic types of radial dependence of $\\alpha$ are treated as a forward problem. In the third section, the evolutionary strategy  to invert the ''measured'' growth rates for a number of magnetic field modes into a radial dependence of $\\alpha$ is presented. Then, we try to reconstruct five models from section 3. Numerically, we find quite different $\\alpha$-profiles giving nearly the same eigenvalues. Additionally,  we treat two other problems which can be, at least in principle, of interest for future dynamo experiments. At first we try to find such a function $\\alpha(r)$ which yields zero growth rate for the lowest four field modes. Such a profile for $\\alpha$ might be a candidate for a (hypothetical) dynamo experiment showing such interesting effects like mode-switching. Further on, we use our code for an ''energetic'' optimization of $\\alpha^2$-dynamos. The paper closes with some  remarks on possible  future developments. ",
        "conclusions": "In this paper we have tried to reconstruct the radial distribution of $\\alpha$ from growth rates of a few $l$-modes of the magnetic field.  It was shown that some paradigmatic functions $\\alpha(r)$ can indeed be recovered by the used evolutionary strategy if a careful regularization of the trial functions is used. With decreasing smoothing effect of the regularization, the ES erratically jumps into various local minima and the resulting functions $\\alpha(r)$ can differ drastically from the original one.  Although by our simple numerical means we are not able to test the isospectrality of different $\\alpha$-profiles in a strict sense, the obvious existence of different  minima with nearly the same quality functions seems to be more than an accident. A more precise identification of such isospectral $\\alpha$-profiles is highly desirable and we would like to suggest further analytical and numerical investigation in this direction.  On the other side, for any practical application (e.g. with measurement errors and only a finite number of $l$-modes) it is quite clear that the function $\\alpha(r)$ can be determined only roughly.  A unique determination of $\\alpha(r)$ would probably require knowledge of the spectra for two different boundary conditions which is, even for laboratory dynamos, not very realistic.  An obvious future task is to generalize the presented method to the oscillatory case, possibly in connection with varying conductivity (for possible astrophysical applications of this case, see Schubert and Zhang 2000). In principle, the developed ES code can be combined with any forward solver for more complicated dynamo models in order to apply the method to real cosmic or laboratory dynamos. For other than spherical geometry it might also be useful to combine the method with an forward solver based on the integral equation approach (Stefani, Gerbeth, and R\\\"adler 2000)."
    },
    "0010/astro-ph0010573_arXiv.txt": {
        "abstract": "New optical and infrared observations along the sight-line toward the quasar OI$\\;$363 (0738+313) are presented and discussed.  Excluding systems which lack confirming UV spectroscopic observations of the actual Ly$\\alpha$ line, this sight-line presently contains the two lowest-redshift classical damped Ly$\\alpha$ (\\DLya) quasar absorption line systems known (i.e. with $N_{HI} \\ge 2 \\times 10^{20}$ atoms cm$^{-2}$), one at $z_{abs}=0.0912$ and the other at $z_{abs}=0.2212$. Our new observations suggest identifications for the \\DLya galaxy counterparts of these absorption-line systems. The $z=0.09$ \\DLya galaxy appears to be an extended low surface brightness galaxy which is easily visible only in infrared images and shows rich morphological structure.  Assuming there is no contribution from the quasar host galaxy, we place an upper limit on the K-band luminosity of the $z=0.09$ \\DLya galaxy of $L_K \\le 0.13L^*_K$ (assuming a cosmology with $H_0=65$ km s$^{-1}$ Mpc$^{-1}$, $\\Omega=1$, and $\\Lambda=0$).  More realistically, a subtraction of the quasar nuclear and host light yields $L_K \\approx 0.08L^*_K$.  The impact parameter between the galaxy and quasar sight-line is very small, $b<3.6$ kpc ($<2$ arcsec), which makes measurements difficult.  The $z=0.22$ \\DLya galaxy is an early-type dwarf with a K-band luminosity of L$_K \\approx 0.1L^*_K$ at impact parameter $b=20$ kpc.  Its colors are neutral and consistent with star formation models suggesting its formation epoch was less than a few Gyr ago (i.e. $z_f\\approx0.3-0.9$).  Thus, it is conceivable that its progenitor originated from the population of ``faint blue galaxies'' seen at moderate redshifts. In general, these results serve to support mounting evidence that \\DLya galaxies are drawn from a wide variety of gas-rich galaxy types. ",
        "introduction": "The $z_{abs}=0.0912$ and $z_{abs}=0.2212$ damped Ly$\\alpha$ (DLA) absorption-line systems along the sight-line toward the quasar OI$\\;$363 (0738+313) have been the subject of several recent papers by members of our group. Rao \\& Turnshek (1998, hereafter RT98) announced the discovery of these systems during the course of an HST-FOS survey for \\DLya absorption in QSO MgII systems.\\footnotemark \\footnotetext{The results of the full spectroscopic survey for low-redshift \\DLya systems are discussed in Rao \\& Turnshek (2000).} In this paper they discussed the results of a preliminary optical imaging search for galactic light from these two absorbers, and they concluded that the galactic counterparts are definitely not luminous spirals.  Lane et al. (1998a,b) and Lane, Briggs \\& Smette (2000a) presented the results of 21 cm observations along the sight-line toward OI$\\;$363. They found that both of the damped systems exhibit 21 cm absorption. However,  an initial search for 21 cm emission from the lower-redshift system failed to detect a signal (Lane et al. 2000b).  The existence of these two systems is noteworthy because they currently represent the two lowest-redshift classical quasar \\DLya absorption-line systems which are confirmed, i.e., actual UV spectroscopic observations of their \\DLya profiles exist and analyses of the spectra demonstrate that the systems' neutral hydrogen column densities are above the classical survey limit of $N_{HI} \\ge 2 \\times 10^{20}$ atoms cm$^{-2}$. In several other reported cases of low-redshift \\DLya systems, either confirming UV spectroscopy is lacking (Miller, Knezek \\& Bregman 1999), the \\ion H1 column density is somewhat below the classical limit (Lanzetta et al. 1997), or both. Thus, although detection of foreground galaxies near a bright background QSO remains difficult, the OI$\\;$363 sight-line offers a most ideal opportunity for studying any emission from  stellar and gaseous components of objects responsible for classical \\DLya absorbers. We refer to such luminous objects as \\DLya galaxies.  Here we report on the results of new observations of galaxies along the sight-line toward OI$\\;$363.  The data consist of optical and infrared imaging and also optical spectroscopy in seeing conditions sometimes as good as $\\approx 0.55$ arcsec. Throughout the analysis we assume $H_o = 65$ km s$^{-1}$ Mpc$^{-1}$, $\\Omega=1$, and $\\Lambda=0$; we also adopt the naming convention used in RT98. In \\S2 we start by summarizing the results from our earlier studies.  In \\S3 the new observations and analyses of these data are presented.  In brief, the \\DLya galaxy associated with the $z=0.0912$ absorber appears to be an extended low surface brightness dwarf galaxy at impact parameter $b<3$ kpc, with rich morphological structure and a K-band luminosity of $L_K\\le0.08-0.13L^*_K$, where the measured light is an upper limit depending on the amount of contamination from the quasar host galaxy at redshift $z \\approx 0.63$ ($L_K\\le0.13L^*_K$ assumes no host contribution). The \\DLya galaxy associated with the $z=0.2212$ absorber is considerably different in character; it is an early-type dwarf galaxy which is relatively compact at an impact parameter of $b=20$ kpc, with a K-band luminosity of $L_K\\approx0.1L^*_K$. In \\S4 we discuss the implications of these findings. ",
        "conclusions": "Although there are still open questions which better data could address, the findings discussed here represent some of the first results on the {\\it detailed properties} of \\DLya galaxies. Perhaps fittingly, these results were determined from the two lowest redshift ($z=0.09$ and $0.22$) confirmed classical \\DLya systems known.  The results of our analysis indicate the following. The $z=0.22$ \\DLya galaxy is a faint (B$=22.7$, K$=17.8$) neutral-colored (B$-$K$=4.9$) dwarf galaxy with K- and B-band luminosities of $L_K = 0.10L_K^*$ and $L_B = 0.10L_B^*$, respectively.  Its spectrum is that of an early type galaxy (S0 or even E).  Analysis of the 7-color (UBRIJHK) data suggests that G1's spectral energy distribution is well-fitted by an instantaneous single-burst model with the burst taking place 0.36 (A$_V = 1.3$) to  3.5 (A$_V = 0.1$) Gyrs ago, or a 3.5 Gyr old model with an exponentially decreasing star formation rate ($\\tau=1$ Gyr) and A$_V =0.5$.  Its current rate of star formation is low, $< 0.3$ M$_{\\odot}$ yr$^{-1}$. Examination of the evolutionary history of these models leads us to conclude that the progenitor population of G1, if seen near the population's formation epoch at $z_f\\approx0.3-0.9$, would have properties similar to the population of ``faint blue galaxies'' presently observed at moderate redshift.  Analysis of the extended radial light profile from the WIYN R-band image of G1 indicates that its outer isophotes are better fit by an exponential than an r$^{1/4}$ law, which offers some evidence for an early-type dwarf S0 or spiral interpretation. The putative $z=0.09$ \\DLya galaxy is likely to be all or part of the resolved light surrounding the quasar. It is also a dwarf galaxy, and is inferred to have $L_K \\le (0.08-0.13)L^*_K$, with the higher value holding for the case where the quasar host galaxy at $z=0.63$ does not contribute to the detected resolved light.  There is considerable morphological structure in the light surrounding the quasar, including what we have called an arm and a jet. The arm is red in color and if it is an {\\it actual} arm, this suggests that the \\DLya galaxy might be a low surface brightness dwarf spiral, but the presence of the other structure (i.e. a blue colored jet with apparent structure) leaves open the possibility that the \\DLya galaxy is an irregular or an interacting system. In any case, the $z=0.22$ \\DLya galaxy is more compact, while the inferred \\DLya galaxy at $z=0.09$ has low surface brightness. For these galaxies the \\ion H1 gas masses deduced from the observed impact parameters and \\ion H1 column densities lie in the range $M_{HI} \\approx (6-23)\\times10^{9}$ M$_{\\odot}$. These estimates are crude because we do not know if the observed \\ion H1 is distributed like a face-on circular slab around the center of the identified \\DLya galaxy (like our calculation of the \\ion H1 mass assumed) or whether the \\ion H1 gas is exponentially distributed or clumped in giant clouds associated with the identified galaxy.  In any case, it would clearly have been much more difficult to deduce these properties from similar observations if these same objects were at moderate to high redshift.  However, a collection of such detailed results, used in conjunction with statistical results on the incidence, cosmological \\ion H1 gas mass density and column density distribution of \\DLya absorption systems (Rao \\& Turnshek 2000), will eventually be needed to accurately reveal the evolution of the bulk of the neutral gas in the Universe (as tracked by \\DLya quasar absorption systems) and the corresponding evolutionary relationship to \\DLya galaxies (i.e. the associated stellar light).  One qualitative question presently of interest is to determine if the \\DLya population is slowly evolving from $z\\approx4.0$ to $z\\approx0.5$, as has been recently suggested by Pettini et al. (1999) and Rao \\& Turnshek (2000) on the basis of their metallicities and statistics, respectively.  It would seem that there has to be some significant evolution in the \\DLya population from high to low redshift if the results of Rao \\& Briggs (1993) at $z=0$ are correct. The Rao \\& Briggs (1993) study places almost all of the local \\ion H1 gas in luminous spirals. Present evidence suggests that luminous galaxies were approximately in place by $z\\approx1$, with less luminous galaxies showing more evolution at lower redshift (e.g. Im et al. 1996, 1999).  Therefore, one could envisage a scenario where luminous galaxies contribute little to the \\DLya detections, except locally, while an evolving population of dwarfs and low surface brightness galaxies cause the bulk of the detections at $z>0$. Indeed, the present searches for low-redshift DLA galaxies (Le Brun et al 1997; Turnshek et al. 2000) suggest that a significant fraction of them are dwarf and low surface brightness galaxies, and the possible connection to the evolving moderate redshift ``faint blue galaxy'' population mentioned earlier is intriguing.  Clearly the statistical survey results and the low-redshift follow-up studies of DLA systems need to be extended and improved to address these evolutionary issues properly. It is also of interest to consider selection effects more carefully as well as design experiments which will reach to lower \\ion H1 masses in searches for large quantities of \\ion H1 gas not associated with luminous spirals. The current 21 cm emission surveys are sensitive to $M_{HI} \\approx 10^7$ M$_{\\odot}$ locally.  Regardless of the outcome of such new programs, the results reported here are consistent with our earlier findings (RT98) and serve to emphasize that \\DLya absorbing galaxies are not exclusively luminous spirals.  They are drawn from a wide variety of types of gas-rich galaxies.  \\medskip  \\centerline{\\bf Acknowledgements}  \\smallskip  We wish to thank Dr. Frank Briggs for collaboration on parts of this program and Drs. Bolzonella, Pello, and Miralles for making their spectral synthesis code {\\it hyperz} publicly available. We wish to also thank the support astronomers at the IRTF, Mayall, MDM, and WIYN telescopes for their assistance. This work was supported in part by an NSF grant and a NASA-LTSA grant.  \\newpage"
    },
    "0010/astro-ph0010059_arXiv.txt": {
        "abstract": "s{ I review some of the current efforts to create unbiased samples of galaxy clusters. Readers are referred elsewhere for general wide area sky surveys and redshift surveys, and for Sunyaev-Zeldovich, radio, infrared and submm surveys, some of which were not designed to search primarily for clusters. My focus will be on optical and X-ray samples and on high redshift clusters. } ",
        "introduction": " ",
        "conclusions": ""
    },
    "0010/astro-ph0010603_arXiv.txt": {
        "abstract": "In this paper we revisit the  radio pulsar death line problem within the framework of curvature radiation and/or inverse compton scattering induced vacuum gap model above neutron star polar caps.  Our special interest is in the recently detected pulsar PSR J2144$-$3933 with extremal period 8.5 seconds, which lies far  beyond conventional death lines. We argue, that  formation of vacuum gaps    requires a complicated multipolar surface magnetic field, with a strenght $B_s$ much higher than the surface dipolar  component $B_d$, and radii of curvature ${\\cal R}$ much smaller than the neutron star radius $R=10^6$~cm.  Such a multipolar surface field is also consistent with death lines including the extremal pulsar PSR J2144$-$3933. Since vacuum gap  models produce sparks, our paper naturally supports the spark related models of subpulse drift phenomenon as well  as to the spark associated models of coherent pulsar  radio emission. ",
        "introduction": "Radio pulsars are believed to turn off when they can no longer produce electron-positron pairs in strong and curved magnetic fields within a so called inner accelerator region just above the polar cap. The limiting rotational period $P$ at which this occurs depends on the magnitude and configuration of the surface magnetic field $B_{s}$. Unfortunately, only the dipolar component $B_{d}$ of this field can be deduced from the observed spin-down rate. The line on the $B_{d}-P$ plane or $\\dot{P}-P$ plane corresponding to this critical period is called a death line. No radio pulsar should be observed to the right or above of this line i.e. with a period longer than the critical one. Recently, \\citet{ymj99} reported on the existence of PSR J2144$-$3933 with a period of 8.5 second which is by far the longest known.  This pulsar, which is located far beyond all conventional death lines, seem to challenge existing emission models. As \\citet{ymj99} concluded themselves, under the usual assumptions, this slowly rotating pulsar should not be emitting a radio beam.  In this paper we attempt to examine this problem assuming that PSR J2144$-$3933 generates radio emission just the same way as other pulsars. Assuming that pulsar's inner accelerator is the Vacuum Gap (VG henceforth) driven by Curvature Radiation (CR henceforth) for pair production in strong magnetic fields, we find that its radio detection is consistent with a strong and complicated surface magnetic field in pulsars. When such fields with a magnitude much greater than the measurable dipole component and radius of curvature much smaller than the neutron star radius $R=10^6$~cm is assumed, then one can easily find general death lines which include PSR J2144$-$3933 on their left side (see Fig.~1). This is consistent with \\citet{as79}, who noted that if pair creation is essential for coherent radio emission, then pulsars with very long periods ($\\sim 5$ seconds) require more complex surface field than a pure dipole (see also Arons 2000), where similar conclusions were drawn in a more complete treatment, including the Inverse Compton Scattering (ICS henceforth) as a source of pairs and frame-dragging effect modifying the electric field near the polar cap surface). We demonstrate in this paper that formation of vacuum gap above pulsar polar cap is possible, provided that the surface magnetic field is extremely non-dipolar (sunspot-like).  \\citet{ymj99} in their discovery paper already noticed that extremely curved and strong surface field would solve the problem of death line in PSR J2144$-$3933. They were however reluctant to accept that such extreme configuration with $B_s\\gg B_d$ and ${\\cal R}\\ll R$ can exist. \\citet{zhm00} argued that if the pulsar's inner accelerator is the so-called Space Charge Limited Flow (SCLF henceforth) with the ICS as the dominating mechanism of pair production then a pure dipolar magnetic field configuration is sufficient to produce death lines which can explain the presence of this pulsar. We however, for reasons explained below, will assume that the surface magnetic field configuration required by the curvature radiation induced vacuum gap (VG-CR) and/or inverse compton scattering vacuum gap (VG-ICS) is just what Nature creates in pulsars, which perhaps is not the simplest explanation, but nevertheless a possibility, which we intend to explore in this paper.  The only strong condition to define pulsar deathlines is the condition for pair production, regardless of the detailed mechanisms to produce these pairs. Pulsar deathlines are therefore model dependent. In history, both the VG model and the SCLF model have been developed and the corresponding death lines of both models were investigated by Ruderman \\& Sutherland (1975; RS75 henceforth) and \\citet{as79}. The original versions of both models assume CR to be the dominating mechanism for pair production and neglect the influence of the general relativistic frame-dragging effect and other sources of pairs. Both models were later improved to include the frame-dragging effect \\citep{mt92} and the ICS \\citep{zql97a,zql97b} as an alternative mechanism for pair production. \\citet{zhm00} fully investigated death lines of all different kinds of the models. In \\S~2 we argue that both the observational data and theory of coherent radio emission seem to favour the VG models of inner pulsar accelerator. We find that the group of pulsars with drifting subpulses can be naturally explained by the VG-CR model, but to explain all the normal pulsars one has to also invoke the VG-ICS model. ",
        "conclusions": "In this paper we have examined an influence of extremely strong and curved surface magnetic field on the long standing binding energy problem \\citep[for review see][]{xqz99}. We have demonstrated within CR and/or ICS driven pair production model that formation of a vacuum gap above the pulsar polar cap is in principle possible, provided that $B_s\\gg B_d$ and ${\\cal R}\\ll R$. We have also addresed in this paper a very interesting and currently topical problem of why so many radio pulsars are beyond the conventional death lines, where theoretically they should not be converting high energy photons into electron-positron pairs. PSR J2144$-$3933 is the pulsar of our special interest with $P=8.5$~s and $\\dot{P}=4.75\\times 10^{-16}$ (thus $B_d=4\\times 10^{12}$~G), which is located extremely far beyond conventional death lines (circle in Fig.~1).  Assuming strong multipolar surface magnetic field suggested by the binding energy problem, we can explain this extremal object without invoking any different radiation mechanism than that for ordinary pulsars. A value of $B_s \\stg 10^{13}$~G and ${\\cal R}\\stl 10^5$~cm seems to fit the constraint imposed by the 8.5~s period well. \\citet{ymj99} have noticed that extremely twisted field could marginally explain the pulsar location in the $B_d-P$ diagram, but they consider such fields unlikely. However, they did not consider a sunspot like configuration that we use in this paper.  There is a growing evidence that quite a large number of conal-type pulsars \\citep{r86} with drifting subpulses (grazing cuts of the line-of-sight) or periodic intensity modulation (central cuts of the line-of-sight) have the RS75-type polar gap accelerators with curvature radiation dominated magnetic pair production \\citep{rs75,dr99,vj99,gs00,xqz99}. We argue in this paper that in such cases the surface magnetic field penetrating the polar gap should be dominated by a strong multipolar (presumably sunspot-like magnetic field with $B_d\\ll B\\stl B_q$ and ${\\cal R}\\ll R$) components, reconnecting with a global dipole field well before the radio emission region. \\citet{zhm00} concluded in their recent paper that it is not necessary to postulate ad hoc multipolar field configuration to explain PSR J2144$-$3933. In fact, they demonstrated that the ICS induced SCLF accelerator produces death lines that include PSR J2144$-$3993. However, we would like to strongly emphasize that the SCLF model is unable to explain the subpulse drift phenomenon, which seems to be a common phenomenon, at least among the conal-type profile pulsars \\citep{r86,gs00}, although there is no direct evidence for conal type emission in PSR J2144$-$3933.  As can be seen from Fig.~1, the VG formation is in principle possible for all normal pulsars (excluding binary and millisecond ones), provided that the surface magnetic field is extremaly strong and curved. Pulsars above the line (1) can form the VG-CR inner accelerator even with relatively low surface field $B_s\\stl 5\\cdot B_d$. However, pulsars below this line require much higher fields $B_s\\gg B_d$ to form the VG-CR ($b\\sim 100$ at line (2) but it has to decrease towards the line (1), as the actual field should not exceed the critical field $B_c\\sim 5\\cdot 10^{13}$~G). Below the line (2) one cannot form the VG-CR accelerator unless $B_s\\gg 100B_d$. However, it is possible to form a VG-ICS inner accelerator with $b\\sim 100$ and ${\\cal R}_6\\sim 0.1$ on the line (3). Since the dipolar field $B_d\\sim 10^{11}$~G around this line, one can use $b$ even larger than 100, and therefore all normal pulsars lie above critical lines for the VG formation. The general conslusion is that the VG formation, either CR or ICS induced, require strong and curved surface magnetic field with radius of curvature ${\\cal R}\\sim 10^5$~cm and the field strenght $B_s\\gg B_d$, where $B_c$ is the photon splitting level. This implies, that radio pulsars (at least non-milisecond ones) are neutron stars with surface magnetic field strenght 10-100 times the surface dipolar component derived from $P$ and $\\dot{P}$ values. Such NS can form VG-CR or VG-ICS inner accelerators, which discharges via a number of localised sparks. Both VG-CR and VG-ICS produce sparks, importance of which for the mechanism of coherent pulsar radio emission was recently emphasised by \\citet{xzq00}.  The sparking discharge within a VG inner accelerator is also a natural explanation of the subpulse drift/periodic modulation. The group of pulsars with clearly drifting subpulses seem to favour the VG-CR model, although the VG-ICS model is not excluded. Few well know drifters lying below the line (2) in Fig.~1 seem to require the ICS contribution or even domination of the gap discharge. It is important to emphasize that pulsars with signatures of periodic subpulse modulations seem to be distributed all over the bulk of the $P-\\dot{P}$ plane \\citep{r86}. Therefore, it seems unlikely that pulsars with drifting subpulses represent the VG accelerator while others are driven by the SCLF accelerator. We suggest that normal radio pulsars are those neutron stars which can develope the VG inner accelerator above their polar caps. This interpretation is consistent with the current status of theory of the coherent pulsar radio emission, which also seem to favour the sparking VG model, at least in normal pulsars \\citep{gs00,mgp00,qlzh00}. The millisecond pulsars are probably driven by the SCLF inner accelerator. This problem will be examined in the forthcoming paper.  We would like to note that a sunspot-like configuration conjectured in this paper for VG formation and supported by the case of PSR J2144$-$3933 is also suggested by the spin-down index of PSR B0540$-$69 \\citep[CR93,][]{rzc97}. Also \\citet{cz99}, analyzing the X-ray emissions from regions near the polar cap of the rotation powered pulsars, argued that $B_s\\sim 10^{13}$~G and ${\\cal R}\\sim 10^5$~cm, which agrees well with our independent estimates. As already mentioned, \\citet{gs00} attempting to explain morphological differences in pulse shapes and variations in polarization properties of different profile types as well as the subpulse drift rates, concluded that the surface magnetic field at the polar cap should be dominated by sunspot-like structures. It seems that there is a growing evidence of such small scale anomalies in surface magnetic field of radio pulsars \\citep[see also][]{ps96,r91}, and our paper gives independent arguments supporting this idea.  Polarization studies in radio pulsars seems to suggest that significant part of the radio emission arises from regions in the magnetosphere where the magnetic field is largely dipolar. Also studies concerning the morphology of pulse profiles is consistent with dipolar magnetic field structure in the radio emission altitudes which is thought to be arising close to the stellar surface given by the relation $r\\sim 50\\cdot R\\cdot P^{0.3}$ \\citep{kg97,kg98}. Thus, if there are strong and complicated surface magnetic fields $B_s\\sim 100B_d$ on the stellar surface it is important to access whether complicated fields would decay fast enough with altitude such that in the emission region the magnetic field structure is almost purely dipolar as constrained by observations. On the other hand, the radius of curvature of surface field lines should be about $10^{5-6}$~cm, which means that the structure of the gap magnetic field is determined by multipoles higher than the quadrupole. Here we consider the model of the magnetic field to be sunspot like near the surface of the neutron star while the global dipolar magnetic field is that of a star centered dipole. Such small scale magnetic fields on the surface of the neutron  star has also been considered by several authors to explain the radio emission properties from pulsars \\citep[e.g.][]{kro91,vr80}. In the `crustal-model' of the neutron star where the magnetic field is generated in neutron star due to the currents in the crust, it is predicted that these models are only capable of producing small scale magnetic fields \\citep{uly86}. These small scale fields can be modelled as small current loops giving rise to dipolar fields which are oriented in different directions all over the stellar surface and superposed on the global dipole field. Following this one can express the magnetic field $B(r)$ at a distance $r$ from the stellar center as the multipolar expansion $B(r)\\approx B_d(R/r)^3+\\Sigma_{l\\geq3}B_l(R/r)^{l+1}$, where $l=2,3,4...$ correspond to dipole, quadrupole, octupole etc. and $B_l$ is the magnetic field strength of a given pole at the neutron star's surface. The number of reversals of the magnetic field across the stellar surface depends on the order of the multipole in question. For example a pure dipolar field will have 2 reversals while a pure quadrupole will have 4 reversals and in general for a multipole of order $l$ there will be $2^{l-1}$ reversals. Here we consider the multipolar field which has sun-spot like magnetic loops with typical radius of curvature ${\\cal R}\\sim 0.1R\\approx 10^5$~cm. This type of structure corresponds to multipoles of order $l>4$. The actual multipole order of a crust origin field clump can be estimated as $l\\sim\\Delta r/R$, where $\\Delta r\\sim 0.1R$ is a characteristic crust thickness. For typical pulsars with $P\\sim1$ second the radio emission arises at altitudes $r\\sim 50\\cdot R$, and the ratio $B_d/B_{l>4} > 1$ in the emission region even if $B_l/B_d\\sim 100$ at the surface. This means that radio emission arises from dipolar field lines even if there are strong, small scale multipolar anomalies on the surface of the polar cap. The field surface strenght of a local multipole is $B_s\\sim(R/\\Delta R)^mB_d$, where $m=1$ or 2 depending on relative orientation of adjacent magnetic moments \\citep[e.g.][]{a81}. This gives an estimate $B_s/B_d\\sim10\\div 100$, as inferred in this paper using the binding energy arguments.  We have shown in this paper that VG formation in typical pulsars requires the crust-origin surface multipolar field to be much stronger (10-100 times) than the dipolar surface component of the star centered global field. In order to apply to the millisecond pulsars, we would have to invoke magnification factors $10^4-10^5$. This would mean that the millisecond pulsars have surface magnetic fields of the same order as the typical (normal) pulsars. \\citet{cgz98} came to similar conclusion analysing the non-thermal X-ray emission from outer gaps of rotation powered pulsars. Whether such strong surface magnetic field can exist in millisecond pulsars is not known. We would like to mention here one constraint found in the literature. \\citet{a93} has shown that the location of the spin-up line in the $P-\\dot{P}$ diagram constraints the large scale anomlies of the magnetic field at the surface to no more than about 40\\% of the surface strenght of dipolar field. This constraint does not concern the crust-origin small scale anomalies invoked in this paper, which will not affect the dipolar structure of the magnetic field in the radio emission region. However, if such extremely strong fields are excluded in the millisecond pulsars, then their inner accelerator must be of SCLF type, which implies the maser kind of radio emission mechanism \\citep{kmmu87,kmm92}.  It is worth to emphasize that all our conclusions  presented in this paper concern vacuum gaps in neutron stars. Therefore, we claim that it may not be necessary to invoke the BPCSS conjecture proposed by \\citet{xqz99}, according to which pulsars, at least those with drifting subpulses, are strange stars with bare polar caps (for review see Xu et al. 2000a). However, we have to admit that we cannot exclude the BPCSS hypothesis.  A complicated magnetic field with radius of curvature much smaller than the star radius has to be confined to the neutron star crust. \\citet{mkb99} examined the evolution of multipole components generated by currents in the outer crust. They found that mostly low order multipoles contribute to the required small radii of curvature, and that the structure of the surface magnetic field is not expected to change significantly during the radio pulsar lifetime. The important question is if the crust can support surface magnetic fields with a magnitude approaching $10^{14}$~G. Equating the magnetic pressure of a strong multipolar field to the crustal stress one concludes that the maximum field which the neutron star can sustain is approximately about $10^{14} - 10^{15}$ G \\citep{td95}. Thus, our inferred magnitudes $B_s>10^{13}$ G are in the regime where cracking of the neutron star crust will still not occur.  We would like to point out a small weakness of our paper, namely an apparent lack of pulsars in a valley between the marginal line (4) in Fig.~1 (including the extremal pulsar PSR J2144$-$3933) and the right-hand boundary of the bulk of $P-\\dot{P}$ distribution. The lack of pulsars at the long period regime near the 8.5 second pulsar is understandable, since detectibility of long period pulsars is much smaller than that of shorter ones, as pointed out by \\citet{ymj99}. The lack of shorter period pulsars within this valley has been explained by \\citet{zhm00}, who invoked a death line below which pair production is not supported by the SCLF-ICS mechanism (line IV in their Fig.~1). Within our VG scenario, this might just mean that strong multipole fields are possible but not common in old pulsars. Another possible explanation concerns a luminosity issue. Let us notice that PSR J2144$-$3933 is a very old (281 Myr), extremaly weak (4 mJy) and close to the Earth (0.19 kpc) pulsar. It would not probably be detected if it was located just a bit further away. So perhaps the radio luminosity of pulsars drops below a detection threshold of a typical survey before their inner accelerators stop completely producing a pair plasma.  Finally, we would like to emphasize once again that vacuum gaps which we have shown to exist in pulsars, produce sparking discharges. These isolated sparks seem to be naturally involved with drifting subpulses observed in typical pulsars \\citep[e.g.][]{r86,dr99,gs00}. On the other hand, spark-associated models of the coherent pulsar radio emission \\citep{qlzh00,mgp00} critically depend on existence of non-stationary vacuum gaps, and therefore our paper supports these ideas on fundamental grounds."
    },
    "0010/astro-ph0010329_arXiv.txt": {
        "abstract": "A new generation of low threshold Imaging Atmospheric Cherenkov Telescopes (IACTs) may reach $\\gamma$-ray energies about 10 GeV with high sensitivities and very large collection areas. At these low energies cosmic electrons significantly contribute to the telescope background and are in principle indistinguishable from $\\gamma$-rays. In this paper we estimate the electron background expected for two configurations of the low energy IACT MAGIC. We discuss in particular the reduction of the background caused by the geomagnetic field at different locations on the Earth's surface. ",
        "introduction": "A number of low energy threshold Imaging Atmospheric Cherenkov Telescopes are under design or construction and will cover the $\\gamma$-ray region from 10 GeV to 10 TeV. Ground-based instruments \\cite{iacts99} provide high sensitivities and huge collections areas of the order of 10$^5$ m$^2$, i.e. orders of magnitude higher than those achievable by satellite experiments in the same energy range like GLAST \\cite{kniffen99}. \\par In this paper we shall concentrate on the IACT MAGIC \\cite{bradbury95,barrio98} since this project aims at the lowest energy threshold. The MAGIC collaboration is currently building a 17 meter diameter telescope equipped with a $\\sim$ 600 PMT pixel camera (MAGIC phase I) which will reach an energy threshold of $\\sim$ 30 GeV \\cite{barrio98} (defined as the peak in the raw differential rate, i.e., before any $\\gamma$ selection cuts). In its second phase the camera will be equipped with high quantum efficiency hybrid PMTs (HPD) that will lower the energy threshold down to $\\sim$ 10 GeV. A third phase may use a camera with Avalanche PhotoDiodes (APDs) of even higher efficiency, but we will not consider this for our studies. \\par IACTs have to deal with a background of electron- and hadron-initiated atmospheric showers. At very low energies below 50 GeV hadrons rarely generate enough Cherenkov light to trigger the telescope. According to \\cite{barrio98,gonzalez00} its contribution to the background is negligible below 30 GeV. In turn cosmic electrons will give rise to showers which are practically indistinguishable from $\\gamma$-induced ones and thus can not be eliminated using traditional $\\gamma$-hadron discrimination methods. They are isotropic and can be suppressed by means of an improved angular resolution when the telescope searches for point sources, but they constitute an irreducible background in extended source searches and diffuse gamma flux estimations. \\par The purpose of this paper is to estimate the fluxes and detection rates of cosmic electrons expected in the new generation IACT MAGIC in its phases I and II. In particular we shall carefully consider the effect of the geomagnetic field on the electron background observed at ground level. ",
        "conclusions": "We have estimated the electron background expected in both phases of the low-energy-threshold IACT MAGIC at different geomagnetic latitudes. The raw electron rate expected for MAGIC phase I is approximately 3 Hz in the zenith direction at the magnetic pole and drops to 2.5 Hz at the magnetic equator. For MAGIC phase II these rates are correspondingly 20 and 10 Hz. The reduction with decreasing magnetic latitude reflects the fact that the electron rigidity cutoff is minimal at high geomagnetic latitudes and maximal at the equator. The sensitivity is also improved when going to lower latitudes by as much as 50\\% for $\\Phi (>5~\\u{GeV})$ at the equator. The detection rate decreases for some azimuth angles reaching a minimum when the telescope points to the west and increase for other angles peaking in the east direction. The electron background is expected to be very much reduced in high zenith angle observations. \\par The conclusion can be drawn that any estimate of the $\\gamma$-ray flux of an extended source or of the isotropic $\\gamma$-ray flux will have to take into account the geomagnetic location of the IACT as well as the specific direction in which the instrument is pointing, especially for observations in which the zenith angles are less than 30\\deg."
    },
    "0010/astro-ph0010435_arXiv.txt": {
        "abstract": "We present here the first analysis of the XMM Newton EPIC-MOS data of the galaxy group around \\object{NGC4839}, which lies at a projected distance to the \\object{Coma cluster} center of 1.6\\h5-1 ~Mpc. In our analysis, which includes imaging, spectro-imaging and spectroscopy we find compelling evidence for the sub group being on its first infall onto the Coma cluster. The complex temperature structure around NGC 4839 is consistent with simulations of galaxies falling into a cluster environment. We see indications of a bow shock and of ram pressure stripping around NGC4839. Furthermore our data reveal a displacement between NGC4839 and the center of the hot gas in the group of about 300 \\h5-1 ~kpc. With a simple approximation  we can explain this displacement by the pressure force originating from the infall, which acts much stronger on the group gas than on the galaxies. ",
        "introduction": "There is now compelling evidence from studies in different wavelength bands that the Coma cluster has substructure. The search for substructure in the Coma cluster starts more than a decade ago: Fitchett \\& Webster (\\cite{fitchett}) find a double structure in the cluster center, and Mellier et al. (\\cite{mellier}) find a second peak in the large scale galaxy distribution around NGC4839, which was further confirmed in a study by Merritt \\& Tremblay (\\cite{merritt}). While these studies were based on optical data, Briel et al. (\\cite{briel1}) using X-ray data coming from the all-sky-survey of the ROSAT satellite, found substructures coinciding with NGC4911, and again, NGC4839. Another study by White et al. (\\cite{white}) based on a deeper ROSAT observation revealed even more structure. In this paper we concentrate on the XMM-Newton (Jansen et al. \\cite{jansen}) observation of the sub group around NGC4839, which has a projected distance from the Coma cluster of 1.6 \\h5-1 ~Mpc. In section 2 we review briefly the observation and data treatment of the EPIC data (see Turner et al. \\cite{turner}). This is followed by Sections dealing with  imaging and spectro-imaging. In section 5 we perform a spectroscopic analysis in different regions which is followed by discussion and conclusion. This paper is one of three papers which present XMM-Newton data on the Coma cluster (see also Arnaud et al. \\cite{arnaudm1}; Briel et al. \\cite{briel2}).  Throughout this paper, we assume $H_0=50$~ km s$^{-1}$ Mpc$^{-1}$ \\h5-1 and $q_0 = 1/2$. At a redshift of $z=0.0232$,  $1^\\prime$ corresponds to 38.9~kpc. ",
        "conclusions": "We performed a first analysis of the XMM-Newton data of the NGC4839 group, which has a projected distance of the Coma cluster of 1.6$h_{50}^{-1}$~Mpc. In our analysis we resolve for the first time a complex temperature structure around NGC4839 which can be explained by a first infall of the galaxy onto the Coma cluster.  Furthermore, we measure a displacement between NGC4839, the dominant galaxy of the group and the extended X-ray emission coming from its ICM. In a simple approximation we can explain this displacement by a ram pressure force of Coma's ICM acting on the hot gas of the sub group as it falls onto the Coma center."
    },
    "0010/astro-ph0010353_arXiv.txt": {
        "abstract": "The observed fluxes of cosmic rays and gamma rays are used to infer the maximum allowed high-energy neutrino flux allowed for Gamma Ray Bursts (GRBs), following Reference \\cite{MPR_00}. It is shown that if GRBs produce the ultrahigh-energy cosmic rays, they should contribute (a) at least 10\\% of the extragalactic gamma ray background between 3~MeV and 30~GeV, contrary to their observed energy flux which is only a minute fraction of this flux, and (b) a cumulative neutrino flux a factor of 20 below the AMANDA-$\\nu2000$ limit on isotropic neutrinos. This could have two implications, either GRBs do not produce the ultrahigh energy cosmic rays \\cite{Stecker_00,Pugliese_00} or that the GRBs are strongly beamed and emit most of their power at energies well above 100 GeV \\cite{Totani_99} implausibly increasing the energy requirements, but consistent with the marginal detections of a few low-redshift GRBs by MILAGRITO \\cite{Atkins_00}, HEGRA-AIROBICC \\cite{Padilla_98}, and the Tibet-Array \\cite{Amenomori_98}. All crucial measurements to test the models will be available in the next few years.  These are measurements of (i) high-energy neutrinos with AMANDA-ICECUBE or an enlarged ANTARES/NESTOR ocean detector, (ii) GRB redshifts from HETE-2 follow-up studies, and (iii) GRB spectra above 10 GeV with low-threshold imaging air Cherenkov telescopes such as MAGIC and the space telescopes AGILE and GLAST. ",
        "introduction": " ",
        "conclusions": ""
    },
    "0010/astro-ph0010486_arXiv.txt": {
        "abstract": "{ We present in this paper a study on the peculiar supergiant HD\\,172481. Its spectral type (F2Ia), high galactic latitude ($b$\\,=\\,$-$10\\fdg37), circumstellar dust, high radial velocity and moderate metal deficiency ([Fe/H]\\,=\\,$-$0.55) confirm the post-AGB character of this object. A detailed chemical analysis shows slight but real s-process overabundances, however no CNO-enhancement was detected. Furthermore, the spectral energy distribution and the TiO bands in the red part of the spectrum reveal a red luminous companion. The luminosity ratio of the hot F type component and this cool M type companion L$_{F}$/L$_{M}$ is derived for a reddening of E(B-V)\\,=\\,0.44 (L$_{F}$/L$_{M}$ $\\sim$\\,1.8) and indicates that the companion must also be strongly evolved and probably evolving along the AGB. Neither our photometric data-set, nor our radial velocity monitoring show evidence for orbital variability which may indicate that the period is too large for direct binary interaction. Most interestingly, a strong lithium resonance line is detected, which yields an abundance of $\\log\\epsilon({\\rm Li})$\\,=\\,3.6. Several explanations for this large lithium content are explored. ",
        "introduction": "Although the research on the late stages of stellar evolution gained enormous momentum thanks to the IRAS survey back in the beginning of the eighties, a detailed understanding of the final stellar evolution in general and of the post-AGB phase in particular, is still lacking. Since the evolutionary time for an object to cross the HR diagram from the tip of the AGB to the planetary nebula phase is short (100-10000 years depending on the core-mass; e.g. Bl\\\"ocker 1995) the objects are rare and the significance of results on individual objects is difficult to extrapolate to the whole AGB stellar evolution theory.  Moreover, in recent years it became clear that binarity can influence the final stages in many ways. This complicates the general observational picture since results on disguised binaries may lead to misinterpretations when confronted with single star evolution theory. Stars like the Red-Rectangle, which were longtime considered to be proto-typical post-AGB objects, were later discovered to be evolved interacting binaries (Waelkens et al. 1996). More recently, ISO results on the detailed infrared analysis of the circumstellar dust spectra revealed that also the dust evolution can be quite different in binaries than in single stars (e.g. Waters et al. 1998, Molster et al. 1999). There is growing observational evidence that in evolved binaries with a wide period distribution, part of the circumstellar material is trapped somewhere in the system. This has a strong impact on the chemical and dynamical evolution of the system (e.g. Van Winckel et al. 1995, 1999; Jura \\& Kahane 1999). Thanks to the trapped dust, binaries will have a prolonged IR lifetime so any sample of post-AGB objects defined on IR measurements will be biased towards binarity (Waters et al. 1997; Van Winckel 1999).  Since the stellar luminosity increases by large factors (a typical L$_{*}$ is between 3000-8000\\,L$_{\\sun}$) during stellar evolution, most of the systems with one component in a post-AGB evolutionary phase, are single-lined spectroscopic binaries (SB1). Double-lined binaries should be rare since they must consist of two, almost equally luminous and thus almost equally evolved objects. The initial mass of both stars must have been therefore extremely similar.  One possible such object was found to be AFGL\\,4106 (Molster et al. 1999, Van Loon et al. 1999) but it was recognized to be a massive double evolved system consisting of a massive F-type post-red-supergiant and an M-type red supergiant. The spectral energy distribution (SED hereafter) is complex (see Fig. 2 and 5 in Molster et al. 1999) with the F-type dominating the UV and optical wavelengths while the M-type component dominates the near-IR. In the far-IR the huge cool dust component is clearly observed. The luminosity ratio of L$_{F}$/L$_{M}$=1.8 was found with an estimated total luminosity of L$_{F}$\\,=\\,1.3\\,$\\times$\\,10$^{5}$ L$_{\\sun}$ assuming a distance of 3.3 kpc (Molster et al. 1999).  In this paper we discuss in detail another such rare double evolved SB2 system: HD\\,172481 (Table \\ref{tab:basics}). The binary consists of a luminous F-type star, probably in a post-AGB evolutionary stage, and an M-type luminous companion, probably evolving on the AGB. The paper is organized as follows: in Sect. 2 we give an overview of the divers observational data we acquired during the last years. In Sect. 3 we list all the observational evidence for the binary SB2 nature of the system; we construct and discuss the spectral energy distribution (SED) and give the spectral indications of the cool component. In Sect 4 we focus on the photospherical chemical composition of the F-type component and discuss the high Li content. The overall discussion on the system is given in Sect. 5 while the summary of the main conclusions is given in Sect. 6.  \\begin{table}\\caption{Basic parameters of HD\\,172481 (SIMBAD database).}\\label{tab:basics} \\begin{center} \\begin{tabular}{rrr} \\hline \\multicolumn{3}{c}{HD\\,172481} \\\\ \\hline \\hline Coordinates & $\\alpha_{2000}$ & 18 41 36.96 \\\\ & $\\delta_{2000}$ &$-$27 57 01.2\\phantom{0} \\\\ \\hline Galactic    & $l$ &    6.72  \\\\ coordinates & $b$ &$-$10.37  \\\\ \\hline  Mean Magnitude   &   B & 9.58 \\\\ &   V & 9.09 \\\\ \\hline Spectral type &  & F2Ia0 \\\\ \\hline \\end{tabular} \\end{center} \\end{table} ",
        "conclusions": "\\subsection{Post-AGB nature} In this section we list the arguments for the proposed post-AGB character of our object. \\begin{enumerate} \\item The high galactic latitude (Tab. \\ref{tab:basics}; $b$\\,=\\,$-$10\\fdg37), \\item the large radial velocities (Fig. \\ref{fig:fotorv}) and \\item the moderate metal deficiency ([Fe/H]=$-$0.55) \\end{enumerate} point to a population II membership. Furthermore, \\begin{enumerate} \\setcounter{enumi}{3} \\item the SED (Fig. \\ref{fig:SED}) shows the presence of dust, as a result of previous mass loss probably during the AGB phase. \\item The photometry and the H$\\alpha$ profiles show a variability very similar to other post-AGB stars (Sect. \\ref{sec:photrad}). \\item Finally, the slight s-process overabundances are probably the result of the third dredge-up, when He burning products are brought to the surface. \\end{enumerate} The last argument is, however, not straightforward. If we expect the signature of a 3rd dredge-up, also the CNO elements should be enhanced, which is clearly {\\em not} detected: the total CNO abundance is equal to the initial value.  Binary post-AGB stars are often depleted, with a dust disk as an ideal site for the gas-dust separation (see Van Winckel 1999 for a review). Depletion alters the abundances profoundly. An ideal element to trace for possible depletion is zinc (Zn) with its low condensation temperature. A high [Zn/Fe] ratio points to efficient depletion. For HD\\,172481, [Zn/Fe]=+0.12, suggesting that no depletion process has taken place.  \\subsection{Li-rich evolved stars}  Plez et al. (1993) published an abundance analysis of seven luminous AGB stars in the Small Magellanic Cloud. These moderately metal deficient stars ([Fe/H]=$-$0.5) are Li rich, C poor and show a low $^{12}$C/$^{13}$C ratio. The chemical pattern observed in these stars is explained by envelope burning, which is the activation of the H burning CN cycle at the base of the convective envelope (called ``hot bottom burning'' HBB). The Li abundance in these stars is typically $\\log \\epsilon({\\rm Li})=3.0$, which is much larger than expected from standard AGB evolution where Li has been destroyed and/or diluted due to convection. The mechanism thought to be responsible for this Li production is the $^7$Be-transport mechanism (Cameron \\& Fowler 1971) in which $^3$He is converted to $^7$Li via the reaction chain $^3$He($^4$He,$\\gamma$)$^7$Be($e,\\nu)^7$Li. Theoretical models suggest that this mechanism can only be efficient in massive AGB stars (4-7 M$_{\\odot}$; e.g. Sackmann \\& Boothroyd 1992, Mazzitelli et al. 1999). Also massive {\\em galactic} AGB stars with signs of envelope burning were found (Garc\\'{\\i}a-Lario et al. 1999), but not yet extensively chemically analysed.   On the other hand, Li is found in many evolved stars in the AGB or post-AGB evolutionary phase for which there is strong observational evidence that many of these stars are low-mass objects with M$<$2$-$3\\,M$_{\\odot}$: galactic low-mass C stars have been found with Li excesses of $\\log \\epsilon({\\rm Li})>1$ (Abia et al. 1993, Abia \\& Isern 1996, 1997, 2000); unexplained high Li abundances were also found in some metal poor population II post-AGB stars, which showed strong evidence of an efficient 3rd dredge-up: IRAS\\,22272+5435 (Za\\v{c}s et al. 1995), IRAS\\,05341+0852 (Reddy et al. 1997), IRAS\\,Z02229+6208 and IRAS\\,07430+1115 (Reddy et al. 1999). It has been suggested by Reddy et al. (1999) that these stars could be the successors of the Li-rich carbon stars studied by Abia (see references above). However, a detailed abundance analysis of 12 J-type Li-enhanced carbon stars (Abia \\& Isern 2000) showed no s-process enhancements in these stars and a solar metallicity while the $^{12}$C/$^{13}$C ratio is low. These J-type carbon stars are clearly chemically different from the 4 Li-enhanced metal-poor post-AGB stars with high $^{12}$C/$^{13}$C ratios and strong s-process overabundances.  It is generally known that also among red giants ascending the 1st giant branch high Li-abundances can be found. Wasserburg et al. (1995) developed a non-standard mixing mechanism for these RGB stars to explain their low $^{12}$C/$^{13}$C ratio (dubbed ``cool bottom processing'' CBP) and Sackmann \\& Boothroyd (1999) showed that CBP can account for the high lithium abundances observed in the RGB Li-rich K giants (de la Reza et al. 1997 and references therein). Moreover, they suggested that CBP might also occur in the TP-AGB phase (Boothroyd \\& Sackmann 1999). de La Reza et al. (1997) suggested that {\\em all} K giants become Li-rich for a short time. During this period the giants are associated with an expanding thin circumstellar shell supposedly triggered by an abrupt internal mixing mechanism resulting in a surface Li enrichment. However, Jasniewicz et al. (1999) did not find any correlation between Li abundance and infrared excess in a sample of 29 late-type RGB giants with infrared excess. There are even Li-rich giants with no circumstellar dust.  Alternative external scenarios to explain the high lithium abundances invoke the accretion of a substellar object (objects). Such scenarios gained interest by the discovery of several extra solar planets and brown dwarf candidates around solar-type stars. Planets could be even commonly present around PNe (Soker 1999). Engulfment scenarios were developed (e.g. Siess \\& Livio 1999a,b), but cannot account for Li abundances exceeding the initial one and therefore fail to explain the very Li rich stars with $\\log \\epsilon({\\rm Li})\\ge2.8$ (Siess \\& Livio 1999b). In other scenarios (Denissenkov \\& Weiss 2000) planet engulfment acts as the activating mechanism for the CBP, producing also these higher Li abundances.  We can conclude that the Li-enhancements in the low-mass C stars and the 4 metal-poor post-AGB giants emphasize our poor knowledge of the dredge-up and mixing processes or possible accretion during stellar evolution.  \\subsection{Li-enhancement in HD\\,172481}  It is clear that planet engulfment (alone) cannot explain a Li abundance of $\\log\\epsilon({\\rm Li})$\\,=\\,3.57. ``Hot bottom burning'' (HBB) is thought to be responsible for the Li enhancements in moderately metal-poor luminous AGB stars (estimated initial mass $\\sim$7\\,M$_{\\sun}$) in the Small Magellanic Cloud (Plez et al. 1993).  This scenario is theoretically thought to occur only in luminous AGB stars of initial mass 3-7\\,M$_{\\sun}$ (Sackmann \\& Boothroyd 1992). The low metallicity and the kinematics of HD\\,172481 suggest an old and hence lower mass nature. Another argument against the HBB scenario for HD\\,172481 is the absence of a nitrogen enhancement. During the HBB, C is converted into N by the CN cycle. But possibly lithium production may have occurred without burning significant amounts of carbon.  On the other hand, the CBP mechanism may be relevant for the Li enhancement in HD\\,172481. The presence of circumstellar dust is in favour of de La Reza's scenario, if applicable to low mass AGB stars. Unfortunately, we are not able to derive a $^{12}$C/$^{13}$C ratio, which could shed some light on the deep mixing involved.  The recent surprising results of the chemical content of J-type carbon stars (Abia \\& Isern 2000) are interesting in many respects. The O-rich circumstellar dust which is observed in about 5\\%-10\\% of these objects (Lloyd Evans 1991) is difficult to reconcile with standard AGB evolution and led Little-Marenin (1986) and Lloyd Evans (1990, 1991) to conclude that the material expelled by the M-type progenitor was trapped and stored in a binary system. Recent CO measurements of J-type carbon stars confirm this scenario and point to a long-lived reservoir of oxygen rich dust (Jura \\& Kahane 1999). In the scenario proposed by Yamamura et al. (2000) to explain the ISO-SWS spectra of some J-type carbon stars, these stars are wide binaries in which the oxygen rich dust was stored around the unseen companion. Note, however, that no direct evidence of the binary nature of these stars exists yet. Since no large amplitude radial velocity variations are observed (Barnbaum et al. 1991), the eventual orbits must be rather wide and direct binary interaction by Roche-Lobe overflow is unlikely to have happened during evolution on the AGB. Whether the special chemical composition of these stars is linked to the proposed binarity in wide systems is still an open question.  For HD\\,172481 there is direct observational evidence for binarity in a probably wide system. Also this object shows very peculiar abundances with a high Li abundance, only very small overabundances of s-process elements and no strong C enhancement. Here, again, the eventual role of the binary nature of HD\\,172481 with respect to the remarkable chemical composition remains an open question."
    },
    "0010/astro-ph0010215_arXiv.txt": {
        "abstract": "For a sample of 51 blazars with extensive optical polarization data, we used circular statistics to calculate the scatter among the polarization position angles for each object. We found that this scatter is correlated with the radio core dominance. We compared this relationship with the predictions of a simple transverse shock model. The result suggests that blazar jets are likely to cover a wide range of speeds consistent with those derived from observations of superluminal motion. ",
        "introduction": "Jet are very important for AGN unification. All radio loud objects have powerful jets. In the orientation Unification Scheme, blazars, with their extreme flux and polarization variability, have the jets beamed towards us.  However viewing angle unification is not the complete story. Do all the jets have similar physical properties on average, so that orientation is able to explain most of the variations of observed properties among blazars?  Optical polarization of blazars provides magnetic field information very close to the jet formation region and is therefore a very important tool for probing jet physics. We have investigated a sample of 51 blazars with many optical polarization measurements made over the past 20-30 years. In paper I, we have investigated the relationship between the optical polarization position angle and the VLBI radio structure. The results are consistent with a shock model. In this paper, we focus on the variability of optical polarization position angles. Assuming a simple shock model, we link the angle variability with jet orientation and speed to explore the jet physical parameters among blazars. ",
        "conclusions": ""
    },
    "0010/astro-ph0010509_arXiv.txt": {
        "abstract": "Strong evidence for cooling flows has been found in low resolution X-ray imaging and spectra of many clusters of galaxies. However high resolution X-ray spectra of several clusters from the Reflection Grating Spectrometer (RGS) on XMM-Newton now show a soft X-ray spectrum inconsistent with a simple cooling flow. The main problem is a lack of the emission lines expected from gas cooling below 1--2~keV. Lines from gas at about 2--3~keV are observed, even in a high temperature cluster such as A\\,1835, indicating that gas is cooling down to about 2--3~keV, but is not found at lower temperatures. Here we discuss several solutions to the problem; heating, mixing, differential absorption and inhomogeneous metallicity. Continuous or sporadic heating creates further problems, including the targetting of the heat at the cooler gas and also the high total energy required. So far there is no clear observational evidence for widespread heating, or shocks, in cluster cores, except in radio lobes which occupy only part of the volume. The implied ages of cooling flows are short, at about 1~Gyr. Mixing. or absorption, of the cooling gas are other possibilities. Alternatively, if the metals in the intracluster medium are not uniformly spread but are clumped, then little line emission is expected from the gas cooling below 1~keV. The low metallicity part cools without line emission whereas the strengths of the soft X-ray lines from the metal-rich gas depend on the mass fraction of that gas and not on the abundance, since soft X-ray line emission dominates the cooling function below 2~keV. ",
        "introduction": "X-ray spectra of the cores of clusters with cooling flows made with the Reflection Grating Spectrometer (RGS; den Herder et al 2000) on XMM-Newton show a remarkable lack of emission lines from gas at 1~keV or below. For the cluster A\\,1835, at redshift $z=0.2523$, both Chandra (Schmidt et al 2000) and XMM-Newton (Peterson et al 2000) CCD spectra show a strong temperature decrease towards the cluster centre from about 9~keV down to about 3~keV. RGS spectra (Peterson et al 2000) are well fitted by a model of gas cooling over that energy range, at a rate of about $1000\\Msunpyr,$ but not to lower temperatures. There is clear evidence for emission lines from FeXXIV, but not from lower ionization stages such as FeXX--XXII and in particular FeXVII which characterize gas cooling through temperatures of 1~keV and below, respectively. The limit on the rate of gas cooling below 1~keV is about $150\\Msunpyr$. A similar temperature decrease inward is seen in Chandra (Fabian et al 2000b) and XMM-Newton (Tamura et al 2000) CCD spectra of A\\,1795 ($z=0.063$), again with no emission lines from gas cooling through and below 1~keV apparent in the RGS spectra (Tamura et al 2000). The low temperature cooling rate limit in this case is about $140\\Msunpyr$, which is close to the total cooling rate inferred from ASCA spectra (Fabian et al 1994; Allen et al 1999).  Various explanations for this disagreement with simple cooling flow expectations (see Fabian 1994 for a review) are given by Peterson et al (2000). Just invoking a heat source is not sufficient, unless it has very specific properties, since gas does not appear to accumulate at 3~keV, it appears to cool down to that temperature then vanish. Moreover, since gas at 3~keV occupies less than 6 per cent of the volume of gas at 9~keV, any heating needs to be targetted. It is possible that the 3~keV gas mixes with hotter or much colder gas, or absorption becomes important for cooler gas. We discuss these problems and possibilities here in more detail.  Finally we advance a new possibility involving strong metallicity variations in the intracluster gas. The line radiation from intracluster gas above 3 keV is proportional to its metallicity, but if it cools to lower temperatures where line radiation dominates the cooling function, the predicted line strengths depend on the mass fraction of the gas which is metal rich. That fraction may be small. ",
        "conclusions": "We have shown that the XMM-Newton RGS soft X-ray spectra of cooling flows are difficult to interpret. There are problems with many solutions. Sporadic heating, mixing, differential absorption and inhomogeneous metallicity offer possible solutions. If the last proves correct then it has wide implications for the enrichment of both the intracluster medium, and the intergalactic medium which presumably would also have a patchy metallicity. Some issues to do with whether the metals from the first stars are mixed uniformaly into the intergalactic medium or not are discussed by Rees (1997). The chemical evolution of galaxies might also become complex, with bimodal populations\\footnote{The globular cluster population of M87 has a bimodal colour distribution which has been intrepreted as due to the later merger of a metal-rich galaxy (Kundu et al 1999). Formation from gas with bimodal metallicity is an alternative possibility.}.  One possible advantage with a heating explanation is that there is no, or little, mass deposition. This means that the problem of the fate of the cooled gas in a standard cooling flow is avoided. As already mentioned however there is strong evidence for massive star formation and cooled gas in many cooling flows.  If hot mixing or heating solutions apply to cooling flows then they may well apply to the formation of all galaxies by cooling. There has long been an assumption that the visible matter in galaxies is that which could cool on an infall time (Rees \\& Ostriker 1977; White \\& Rees 1978). Massive galaxies do however accrete large hot envelopes of gas where the cooling time exceeds the infall time but is still less than the age of the Universe; i.e. they have cooling flows (Nulsen \\& Fabian 1995). Whether such gas cools and builds galaxies further or is heated and prevented from cooling is not known. Heating by a central engine is one solution. The consequence of this is that the central engine (the growth of the central black hole) then has a profound influence on the final state of the galaxy.  Further observations of the soft X-ray phenomenon in cooling flows therefore have potential implications for our understanding of cooling and heating generally, as well as for the uniformity of metal enrichment of hot gas, the energy content of jets, and galaxy formation."
    },
    "0010/astro-ph0010023_arXiv.txt": {
        "abstract": "Radio galaxies are known to inflate lobes of hot relativistic plasmas into the intergalactic medium. Here we present hydrodynamical and magnetohydrodynamical simulations of these hot plasma bubbles in FRII objects. We focus on the later stages of their evolution after the jet has died down and after the bow shock that surrounded the lobes at earlier stages has vanished. We investigate the evolution of the plasma bubbles as they become subject to Rayleigh-Taylor instabilities. From our simulations we calculate the radio and X-ray emissivities of the bubbles and discuss their appearance in observations. Finally, we investigate the influence of large-scale magnetic fields on the evolution of the bubbles. The issues of re-acceleration and diffusion of relativistic particles are briefly discussed. ",
        "introduction": "The jets of radio galaxies and radio-loud quasars are known to inflate large lobes of radio plasma in the Inter Galactic Medium (IGM). These regions appear either as well-defined cavities in the case of Fanaroff - Riley (FR) type II objects (Fanaroff \\& Riley 1974)\\nocite{fr74} or form broad, more diffuse outflows in FRI objects. In this paper we concentrate on the evolution of the radio lobes of the more powerful FRII objects. In our view, the dynamics of FRIs are more complex and we will refer a discussion of these to a future paper.\\\\  The standard model for FRIIs explains the lobes as overpressured bubbles that expand supersonically into the surrounding intergalactic medium (IGM) (Scheuer 1974)\\nocite{ps74}. In principle, the lobes are unstable because they are surrounded by the denser, unperturbed IGM in the gravitational potential well of the host galaxy. This situation should cause the buoyant lobes to rise through the ambient medium (e.g. Gull \\& Northover 1973\\nocite{gn73}). However, as long as the bow shock, caused by their supersonic expansion, protects the radio lobes, buoyant forces are not important for the dynamics of the lobes (Scheuer 1974\\nocite{ps74}). Analyses of spectral ageing show that the jet phase may last for a few $10^8$ years (e.g. Alexander \\& Leahy 1987\\nocite{al87}) after which the jets presumably stop supplying energy to the lobes. After that, they may still be overpressured with respect to the IGM and continue to expand, driving a bow shock. This phase resembles that of a strong explosion or supernova remnant (e.g. Sedov 1959)\\nocite{ls59}.\\\\  The continued expansion of the lobes will bring them into pressure equilibrium with the IGM. At this point, the bow shock vanishes and the lobes become susceptible to Rayleigh-Taylor instabilities. Because of the absence of strong surface tension in the radio lobes, this situation is reminiscent of the later phases of strong (nuclear) explosions in the Earth's atmosphere. In both cases, the hot rarefied gas forms a torus which, in its wake, drags material up into the atmosphere. Thus the bubble develops a shape that resembles a mushroom. This was discussed in connection with the large scale radio structure of M87 by B{\\\"o}hringer et al. (1995)\\nocite{bnbf95}, Churazov et al. (2000a)\\nocite{cfjb00} and Churazov et al. (2000b)\\nocite{cbkbf00}.\\\\  In this paper we extend the hydrodynamical simulations for the specific case of M87 presented in Churazov et al. (2000b) \\nocite{cbkbf00} to more general initial configurations. Moreover, we performed MHD simulations to study the influence of a magnetic field that is initially aligned with the physical boundary of the radio lobes. We also present predictions for the radio synchrotron emission from the magnetised plasma within the lobes. Finally, we demonstrate that neither re-acceleration processes nor inhomogenous magnetic fields in the radio lobes significantly alter the emission properties of the lobes unless the diffusion of particles from weak fields to high field regions is slow. \\\\  The paper is organised as follows: We discuss our methods and assumptions in Sec. 2. The results of the simulations are presented in Sec. 3, their observational consequences in Sec. 4 and the conclusions in Sec. 5. ",
        "conclusions": "Using hydrodynamical simulations, we investigated the evolution of buoyant bubbles of different sizes and geometries in a galactic gravitational potential. Common to all simulations with a non-spherical bubble is the formation of a large mushroom-shaped lobe that rises through the ambient medium. It is interesting to note that the hot plasma remains confined within a single cloud for a considerable time without requiring surface tension or magnetic fields. This is also observed in the evolution of buoyant thermals which are formed, for example, in strong nuclear explosions in the Earth's atmosphere. In their wake the bubbles lift colder material upwards. \\\\  Differences between simulations with different initial geometries become manifest in different rise velocities and somewhat different morphologies of the buoyant bubble. We have selected a set of initial conditions to show the ranges of rise velocities and morphologies that can occur. The growth times as well as the morphologies of the instabilities are quite different for spherical and elongated bubbles. Only in the latter case a large `mushroom' forms. This `mushroom' rises upwards as a whole before it succumbs to instabilities on smaller scales. Spherical bubbles, however, evolve very differently: Due to their symmetry, Rayleigh-Taylor instabilities mix the material in the bubble with the ambient fluid {\\it before} any large-scale motions can occur. Our conclusions have to be diluted by the usual caveats that come with finite-difference simulations. Our simulations merely serve as toy models that demonstrate the underlying physics and show the global morphology of the evolving bubbles. The reader should bear in mind that the detailed morphologies on smaller scales depend on the numerical procedure, the resolution of the grid and the details of the initial model. Moreover, we only presented 2D simulations which may produce artificial features. However, we have repeated some of our simulations in 3D on a grid with much smaller resolution and did not find significantly different results.\\\\  We also performed MHD simulations where the bubbles were filled with a magnetic field that was parallel to the surface of the bubble. It was found that the magnetic field helps to protect the rising bubbles against secondary Rayleigh-Taylor and Kelvin-Helmholtz instabilities. Only a strong and ordered magnetic field, where the magnetic pressure is comparable to the thermal pressure, may suppress the formation of the mushroom structure.\\\\  Finally, we calculated maps of the radio surface brightness due to synchroton radiation. These maps were compiled for two different assumptions about the magnetic field strength: (i) a low field, where the initial field strength is roughly equal to the equivalent field strength of the CMB, and (ii) a strong field, where the energy density of the magnetic field and that of the relativistic particles are in equipartition. In the weak-field case, the bubble is visible in the radio for more than $2\\cdot 10^8$ years, whereas in the strong-field case the higher energy losses cause the radio image to fade after $1.5\\cdot 10^8$ years. We discuss various processes of re-acceleration and diffusion of relativistic particles as well as their influence on the synchrotron radiation properties of the bubbles. We find that some of these processes may prolong the phase of synchroton emission but if so, the bubbles will be exceedingly faint at radio frequencies."
    },
    "0010/astro-ph0010201_arXiv.txt": {
        "abstract": "We study the dynamical stability against bar-mode deformation of rapidly and differentially rotating stars in the first post-Newtonian approximation of general relativity.  We vary the compaction of the star $M/R$ (where $M$ is the gravitational mass and $R$ the equatorial circumferential radius) between 0.01 and 0.05 to isolate the influence of relativistic gravitation on the instability.  For compactions in this moderate range, the critical value of $\\beta \\equiv T/W$ for the onset of the dynamical instability (where $T$ is the rotational kinetic energy and $W$ the gravitational binding energy) slightly decreases from $\\sim 0.26$ to $\\sim 0.25$ with increasing compaction for our choice of the differential rotational law. Combined with our earlier findings based on simulations in full general relativity for stars with higher compaction, we conclude that relativistic gravitation {\\em enhances} the dynamical bar-mode instability, i.e.~the onset of instability sets in for smaller values of $\\beta$ in relativistic gravity than in Newtonian gravity.  We also find that once a triaxial structure forms after the bar-mode perturbation saturates in dynamically unstable stars, the triaxial shape is maintained, at least for several rotational periods.  To check the reliability of our numerical integrations, we verify that the general relativistic Kelvin-Helmholtz circulation is well-conserved, in addition to rest-mass energy, total mass-energy, linear and angular momentum.  Conservation of circulation indicates that our code is not seriously affected by numerical viscosity.  We determine the amplitude and frequency of the quasi-periodic gravitational waves emitted during the bar formation process using the quadrupole formula. ",
        "introduction": "\\label{sec:intro}   Stars in nature are usually rotating and subject to nonaxisymmetric rotational instabilities.  An exact treatment of these instabilities exists only for incompressible equilibrium fluids in Newtonian gravity, {\\it e.g.} \\citep{Chandra69,Tassoul,ST}.  For these configurations, global rotational instabilities arise from non-radial toroidal modes $e^{im\\varphi}$ ($m=\\pm 1,\\pm 2, \\dots$) when $\\beta\\equiv T/W$ exceeds a certain critical value. Here $\\varphi$ is the azimuthal coordinate and $T$ and $W$ are the rotational kinetic and gravitational potential binding energies.  In the following we will focus on the $m=\\pm 2$ bar-mode, since it is the fastest growing mode when the rotation is sufficiently rapid.  There exist two different mechanisms and corresponding timescales for bar-mode instabilities.  Uniformly rotating, incompressible stars in Newtonian theory are {\\em secularly} unstable to bar-mode formation when $\\beta \\geq \\beta_{\\rm sec} \\simeq 0.14$.  However, this instability can only grow in the presence of some dissipative mechanism, like viscosity or gravitational radiation, and the growth time is determined by the dissipative timescale, which is usually much longer than the dynamical timescale of the system.  By contrast, a {\\em dynamical} instability to bar-mode formation sets in when $\\beta \\geq \\beta_{\\rm dyn} \\simeq 0.27$.  This instability is independent of any dissipative mechanisms, and the growth time is the hydrodynamic timescale of the system.  The secular instability in compressible stars, both uniformly and differentially rotating, has been analyzed numerically within linear perturbation theory, by means of a variational principle and trial functions, by solving the eigenvalue problem, or by other approximate means.  These techniques have been applied not only in Newtonian theory \\citep{LO67,OB73,FS78,IL90} but also in a post-Newtonian (PN) framework (Cutler and Lindblom 1992; Shapiro and Zane 1998 for incompressible stars) and in full general relativity \\citep{BFG96,BFG98,SF,YE}.  For relativistic stars, the critical value of $\\beta_{\\rm sec}$ depends on the compaction $M/R$ of the star (where $M$ is the gravitational mass and $R$ the circumferential radius at the equator), on the rotational law and on the dissipative mechanism.  The gravitational-radiation driven instability occurs for smaller rotation rates, {\\it i.e.}, for values $\\beta_{\\rm sec} < 0.14$, in general relativity.  For extremely compact stars (Stergioulas and Friedman 1998; Yoshida and Eriguchi 1999) or strongly differentially rotating stars (Imamura et al. 1995), the critical value can be as small as $\\beta_{\\rm sec} < 0.1$. By contrast, viscosity drives the instability to higher rotation rates $\\beta_{\\rm sec} > 0.14$ as the configurations become more compact \\citep{BFG96,BFG98,SZ}.  Determining the onset of the dynamical bar-mode instability, as well as the subsequent evolution of an unstable star, requires a fully nonlinear hydrodynamic simulation.  Simulations performed in Newtonian theory, {\\it e.g.} \\citep{TDM,DGTB,WT,HCS,SHC,HC,PDD,TIPD,NCT} have shown that $\\beta_{\\rm dyn}$ depends only very weakly on the stiffness of the equation of state.  Once a bar has developed, the formation of spiral arms plays an important role in redistributing the angular momentum and forming a core-halo structure. Recently, it has been shown that, similar to the onset of secular instability, $\\beta_{\\rm dyn}$ can be smaller for stars with a higher degree of differential rotation \\citep{TH,PDD}.  To date, the dynamical bar-mode instability has been analyzed mainly in Newtonian theory, since until quite recently a stable numerical code capable of performing reliable hydrodynamic simulations in three dimensions plus time in full general relativity has not existed.  Some recent developments, however, have advanced the field significantly. New formulations of the Einstein equation based on modifications of the standard $(3+1)$ system of equations have resulted in codes which have proven to be remarkably stable over many dynamical timescales, {\\it e.g.} \\citep{SN,BS}. In addition, gauge conditions which allow long-time stable evolution for rotating, self-gravitating systems and are manageable computationally have been developed, {\\it e.g.} \\citep{Shibata99b}.  The purpose of this paper is twofold.  We verify that the point of onset of dynamical bar mode formation, as measured by $\\beta$, decreases with increasing compaction, and we furthermore show that in unstable configurations, the bar persists for at least several rotational periods.  In a previous paper \\citet{SBS00} performed simulations of rapidly and differentially rotating neutron stars in full general relativity. They employed the relativistic code of \\citet{Shibata99a} to study the onset and growth of the dynamical bar-mode instability in relativistic stars.  They found that for compact stars with $M/R \\gtrsim 0.1$, the onset of dynamical instability occurs at $\\beta_{\\rm dyn} \\sim 0.24 - 0.25$, somewhat smaller than the Newtonian value of $\\beta_{\\rm dyn} \\sim 0.26$.  In principle, this reduction of $\\beta_{\\rm dyn}$ could be caused by effects of either general relativity or differential rotation.  In order to isolate the two effects one could study sequences of varying compaction $M/R$ or sequences of varying degree of differential rotation.  We choose the former in this paper and focus on the {\\em transition} to relativistic gravitation as the compaction increases to moderate values.  In fully relativistic evolution codes the Courant condition for the gravitational fields limits the size of the numerical timesteps due to the speed of light\\footnote{This limitation could in principle be avoided by using an implicit finite difference scheme.  However, such schemes are more complicated than explicit schemes, and have not yet been implemented for fully relativistic hydrodynamics in three spatial dimensions}.  As a consequence, the ratio between the dynamical timescale for the matter and the Courant timestep for the fields scales approximately as $(M/R)^{-1/2}$, which makes the calculation prohibitively slow for small compactions.  In this paper, we therefore adopt the first order post-Newtonian (1PN) code of \\citet{SBS98} to study the effect of general relativity on $\\beta_{\\rm dyn}$ for small and moderate compactions.  In this formalism, the relativistic evolution equations for the gravitational fields reduce to elliptic equations, which are solved on individual timeslices and hence have no Courant condition. The numerical timesteps are now limited by the Courant condition for the hydrodynamical evolution equations alone and therefore scale with the dynamical timescale.  We evolve models with compactions $M/R$ between 0.01 and 0.05 for several rotational periods to determine their stability.  In cases in which a bar forms we also follow the nonlinear growth and saturation of the instability.  Should the star retain its bar-like shape for many rotational periods, then it would emit a quasi-periodic gravitational wave signal.  Such a radiator would be potentially a very promising source for the gravitational wave interferometers currently under construction.  This prospect has prompted many researchers to study the evolution of the bar-mode instability, most recently \\citet{NCT} and \\citet{Brown}.  They have performed long-duration Newtonian simulations of dynamically unstable stars that form a bar and then spiral arms.  These configurations eject small amounts of mass and then settle down to triaxial stars.  For the early stage of the evolution, the results of these two groups are very similar, but for later times some differences arise.  According to \\citet{NCT}, the bar gradually decays after a few rotational periods, probably due to numerical errors associated with the unphysical motion of the center of mass, whereas \\citet{Brown} reports that the star remains bar-like for more than 10 rotational periods, and presumably until gravitational radiation ultimately drives the decay of the bar.  Given these discrepancies, which are probably due to numerical inaccuracies, it seems desirable to develop additional criteria to verify the late-time reliability of numerical codes.  In particular, it is possible that numerical viscosity associated with finite-differencing could lead to an artificial decay of bar-modes. In this paper we evaluate the conservation of circulation, which is violated in the presence of viscosity (numerical or otherwise).  We present a method for computing the circulation in general relativity and demonstrate that, in our numerical code, the circulation is well conserved.   This paper is organized as follows.  In Sec.~\\ref{sec:model} we present the basic equations of our 1PN formulation of general relativity and describe our initial data in Sec.~\\ref{sec:indata}. We discuss our numerical results in Sec.~\\ref{sec:stability}, focusing on the dynamical stability of differentially rotating stars.  In Sec.~\\ref{sec:Circularity} we demonstrate the conservation of circulation in our simulations and discuss the emitted gravitational wave signal in Sec.~\\ref{sec:GWaves}.  In Sec.~\\ref{sec:Discussions} we briefly summarize our findings.  Throughout this paper, we use the geometrized units with $G=c=1$ and adopt Cartesian coordinates $(x,y,z)$ with the coordinate time $t$.  Greek and Latin indices take $(t, x, y, z)$ and $(x, y, z)$, respectively. ",
        "conclusions": "\\label{sec:Discussions} We perform 1PN simulations of rapidly and differentially rotating stars to investigate general relativistic effects on the dynamical bar-mode instability for small compactions $M/R \\leq 0.05$.  By combining these PN results with the fully relativistic simulations of \\cite{SBS00} for configurations of higher compaction, we conclude that the critical value of $\\beta = \\beta_{\\rm dyn}$ decreases with increasing $M/R$.  Thus, relativistic gravitation enhances the bar-mode instability.  We also describe how conservation of circulation can be used to check by how much a code is affected by numerical visocity.  In the presence of significant numerical viscosity, long-time evolution calculations become very unreliable and may lead, for example, to erroneous evolution of a saturated bar.  We show that in our calculations the circulation is well conserved, implying that our code is at most very weakly affected by numerical viscosity.  We present a method for computing the circulation which can be applied in Newtonian, post-Newtonian and fully relativistic calculations."
    },
    "0010/astro-ph0010171_arXiv.txt": {
        "abstract": "The determination of the primeval deuterium abundance has opened a precision era in big-bang nucleosynthesis (BBN), making accurate predictions more important than ever before.  We present in analytic form new, more precise predictions for the light-element abundances and their error matrix.  Using our predictions and the primeval deuterium abundance we infer a baryon density of $\\Omega_Bh^2 = 0.020\\pm 0.002$ (95\\% cl) and find no evidence for stellar production (or destruction) of $^3$He beyond burning D to $^3$He.  Conclusions about $^4$He and $^7$Li currently hinge upon possible systematic error in their measurements. ",
        "introduction": "The BBN prediction of a large primeval abundance of $^4$He ($Y_P \\approx 0.25$) was the first success of the hot big-bang model.  For two decades the consistency of the BBN predictions for the abundances of D, $^3$He, $^4$He and $^7$Li with their inferred primeval abundances has been an important test of the standard cosmology at early times ($t\\sim 1\\,$sec).  BBN has also been used to ``inventory'' ordinary matter at a simpler time and to probe fundamental physics (see e.g., Schramm \\& Turner 1998 or Olive et al. 2000).  With the detection of the deuterium Ly-$\\alpha$ feature in the absorption spectra of 3 high-redshift ($z>2$) quasars and the accurate determination of the primeval abundance of deuterium, (D/H)$_P=(3.0\\pm 0.2)\\times 10^{-5}$ (Burles et al. 2000; Tytler et al. 2000; O'Meara et al. 2001) BBN has entered a new precision era (Schramm \\& Turner 1998). Because its abundance depends strongly upon the baryon density, and its subsequent chemical evolution is so simple (astrophysical processes only destroy D), deuterium can accurately peg the baryon density. Once determined, the baryon density allows the abundances of $^3$He, $^4$He and $^7$Li to be predicted.  These predictions can be used to test the consistency of the big-bang framework and to probe astrophysics.  The chemical evolution of $^4$He is simple (stars produce it) and so its predicted abundance, $Y_P = 0.246 \\pm 0.001$ (Lopez \\& Turner 1999), can be used as a consistency test of BBN and the standard cosmology.  Because the $^7$Li abundance in old pop II stars may be depleted, lithium probes both stellar models and the consistency of the standard cosmology.  While the post-big-bang evolution of $^3$He is complex, the sum D+$^3$He can be used to study the chemical evolution of the Galaxy. All stars burn D to $^3$He as they evolve onto the main sequence (MS). Later stages of stellar evolution may produce or destroy $^3$He, depending on stellar mass and subject to uncertainty in modeling. Thus, the evolution of D+$^3$He measures the net stellar production of $^3$He beyond pre-MS burning (Yang et al. 1984), providing an important probe of stellar models.  A key to realizing BBN's full potential in the precision era is accurate and reliable predictions.  With this in mind, we have recently used Monte-Carlo techniques (Burles et al. 1999; Nollett \\& Burles 2000) to link the calculated abundances directly to the nuclear data, making the predictions more reliable.  Previous work (Smith et al. 1993, Fiorentini et al. 1998) was based on fitting cross-section data to standard forms, estimating conservative uncertainties to accommodate most or all of the data for each reaction.  Very recent work (Esposito et al. 2000; Vangioni-Flam et al. 2000) computes only ``maximum'' uncertainties, using upper and lower limits quoted in a compilation of charged-particle reaction rates (Angulo et al. 1999).  In contrast, our procedure ties the abundance errors directly to the experimental measurements by weighting the data by their quoted errors, and furthermore, leads to smaller estimated abundance errors (by factors of 2-3).  In this {\\em Letter} we present our results in the form of analytic fits for the abundances and their error matrix. We then use these predictions to make inferences about the baryon density, the consistency of BBN, $^7$Li depletion and stellar $^3$He production. ",
        "conclusions": "We have presented analytical fits for our new, more accurate predictions of the light-element abundances and their error matrix. These results have already renewed interest in more accurately determining key nuclear rates, and further improvements are likely (see e.g., Schreiber et al. 2000; Rupak 2000).  Using our results and the primeval deuterium abundance from three high-redshift Ly-$\\alpha$ systems we infer $\\Omega_Bh^2 = 0.020\\pm 0.002$ (95\\% cl). For $h=0.7\\pm 0.07$, this implies a baryon fraction $\\Omega_B = 0.041\\pm 0.009$, with the error dominated by that in $H_0$. Measurements of cosmic microwave background (CMB) anisotropy have recently also determined the baryon density.  The physics underlying this method is very different -- gravity-driven acoustic oscillations in the Universe at 500 000 yrs -- but the result is similar: $\\Omega_Bh^2 = 0.032^{+0.009}_{-0.008}$ at 95\\% cl (Jaffe et al. 2000).  While there is about a $2\\sigma$ difference, this first result from CMB anisotropy confirms the long-standing BBN prediction of a low baryon density, and with it, the need for nonbaryonic dark matter.  The sum of D+$^3$He predicted for the deuterium baryon density is consistent with that in the ISM today, implying no significant production (or destruction) of $^3$He beyond pre-MS burning.  The deuterium-predicted $^4$He abundance is an important consistency test of BBN; however, systematic measurement error (the $Y_P$ values from the two key data sets differ by 5 times the statistical error) precludes firm conclusions at this time.  Likewise, the discrepancy between the predicted $^7$Li abundance and the abundance measured in pop II stars has no simple explanation: The discrepancy could indicate that $^7$Li has been depleted in pop II stars by about a factor of two, that uncertainties (in observations or predictions) have been grossly underestimated, or that there is an inconsistency in BBN."
    },
    "0010/astro-ph0010492_arXiv.txt": {
        "abstract": " ",
        "introduction": "The progress made in last years in numerical modeling allows to study the internal structure of galaxies with great detail. This is of particular interest for stellar systems composed by several components, such as spiral galaxies. In the recent years, we produced a self-consistent, 3-D model of disk galaxy made by an exponential disk and a $r^{1/4}$ bulge (Pignatelli \\& Galletta, 1999).  The model fit simultaneously the photometric data (surface brightness, ellipticities and PAs of isophote major axis) and the kinematic data (rotation curves and velocity dispersion curves along various PAs). In such a way, the structure of the luminous matter derived from the photometric analysis is compared with the total matter deduced from the dynamical data and the presence of dark matter may be studied.  To test the model, we selected five galaxies with morphological types spanning from S0/a to Sc and absolute magnitudes from -20.6 to -22.5. They have been selected from HI observations looking for objects with a line width $W_{20} > 350$ km/sec.  \\begin{figure}[h] \\plottwo{ngc3263.ps}{ngc6925.ps} \\caption{B images of two galaxies of the sample: the SBcd galaxy NGC 3263 and the Sbc galaxy NGC 6925.} \\end{figure}  For all these galaxies we obtained images at B and I photometric wave bands using the ESO telescopes at La Silla. The used telescopes were the Danish 1.5m, the NTT+EMMI and the ESO-MPI 2.2m+EFOSC2. The spectra have been collected with the spectrographs Boller\\& Chivens and EFOSC2 of the 2.2m ESO-MPI telescope.  We used the model to evaluate the intrinsic properties of these galaxies, such as the disk/bulge mass ratio and the scale length of the galaxy components.  \\vspace*{-0.4cm} ",
        "conclusions": ""
    },
    "0010/astro-ph0010347_arXiv.txt": {
        "abstract": "The preponderance of binary systems in all known stellar populations makes them exciting {\\it dynamical agents} for research on topics as varied as star formation, star-cluster dynamics and the interiors of young and old stars. Today we know that the Galactic-field binary population is probably a dynamically evolved version of the Taurus--Auriga pre-main sequence population, and that the initial distributions of binary-star orbital elements are probably universal. Furthermore, $N$-body calculations tentatively suggest that OB stars form in energetic binaries near cluster cores, and that binaries with 'forbidden' orbital elements that are produced in stellar encounters, may turn out to be very useful windows into stellar interiors, potentially allowing tests of pre-main sequence evolution theory as well as of models of main-sequence stars. ",
        "introduction": "\\noindent Most 'stars' in the sky are multiple systems, and observations demonstrate that the binary proportion of 'isolated' pre-main sequence stars is typically much higher than in the Galactic-field (GF) (Ghez et al. 1997; K\\\"ohler \\& Leinert 1998; Duchene 1999).  Differences such as this contain clues about the origins of various populations. The disruption of binary systems in their stellar birth aggregates may naturally reduce the binary proportion to the GF level. This is an exciting notion, opening the possibility of learning something about the configuration of the birth aggregate typical for present-day star formation (sf).  The study of the evolution of a primordial binary population in a young cluster thus becomes important for understanding the origin of the distribution of binary-star orbital elements. Cluster dynamics with a substantial primordial binary population differs from that of single-star clusters because the number of interacting systems changes with time owing to binary disruption affecting the relaxation process, and because binary systems contain additional degrees of freedom that single stars do not. These internal degrees of freedom affect the energy exchanges that occur during close encounters, changing the energy spectrum of escaping stars and the energy budget of the whole cluster, ultimately being able to arrest core collapse in massive star clusters (e.g. Giannone \\& Molteni 1985; Giersz \\& Spurzem 2000).  This text addresses related questions, with the caveat that many important and interesting problems remain to be studied. Complementary discussions are available in Kroupa (2000a, KI; Kroupa 2000b, KII), this contribution (KIII) concentrating primarily on the implications of stellar-dynamical interactions on the binary-star orbital characteristics and high-velocity stars.  In the following, the mass, $m$, period, $P$ (always in days), eccentricity, $e$, and mass-ratio, $q=m_2/m_1; m_2\\le m_1$, of a stellar system (a binary, or a single with no $P$, $e$, $q$) are collectively referred to as its {\\it dynamical properties}.  The primordial, or birth, distribution functions of these quantities are the initial mass function (IMF), and the initial period, eccentricity and mass-ratio distributions (IPF, IEF, IMRF, respectively). The distribution functions are written as, $f_x$, where $f_x\\,dx$ is, for example, the number of orbits with orbital parameter $x$ between $x$ and $x+dx$. The total (all periods and all primaries) binary proportion is $f_{\\rm tot} = N_{\\rm bin}/\\left(N_{\\rm bin}+N_{\\rm sing}\\right)$, with $N_{\\rm bin}$ being the number of binaries and $N_{\\rm sing}$ the number of single stars in the sample.  There are two main mechanisms that evolve the dynamical properties: {\\it eigenevolution} through system-internal processes that redistribute angular momentum and energy (e.g. tidal circularisation, disk--companion-star interactions, outflows), and {\\it stimulated evolution} through encounters with neighbours. Eigenevolution affects short-period systems ($P\\simless 10^3$~days), whereas stimulated evolution affects long-period systems.  One key goal of the work reported here is to find out if the IPF, IEF and IMRF are universal, or if systematic variations with sf conditions are evident in the available data (Durisen \\& Sterzik 1994). ",
        "conclusions": ""
    },
    "0010/nucl-th0010053_arXiv.txt": {
        "abstract": "We have studied the $^4{\\rm He}({^3{\\rm He}},\\gamma){^7{\\rm Be}}$, $^3{\\rm He}({^3{\\rm He}},2p){^4{\\rm He}}$, and $^7{\\rm Be}(p,\\gamma){^8{\\rm B}}$ reactions of the solar p-p chain, using microscopic cluster models. Among other results, we showed that the $^6{\\rm Li}+p$ channel has a nontrivial effect on the $^7$Be-producing reaction, that the existence of a resonance in $^6$Be close to the $^3{\\rm He}+{^3{\\rm He}}$ threshold is rather unlikely, and that the correlations between some properties of $^7{\\rm Be}/{^8{\\rm B}}$  and the low-energy cross section of $^7{\\rm Be}(p,\\gamma){^8{\\rm B}}$ might help one to constrain the value of the $S_{17}(0)$ astrophysical S-factor. ",
        "introduction": "The very-low-energy cross sections of the solar nuclear reactions are used as input parameters in solar models \\cite{Bahcall}. In order to get reliable results from these models for the energy generation, neutrino fluxes, etc., of our sun, all input parameters, including those coming from nuclear physics, should be known as precisely as possible. Despite the tremendous improvements in our understanding of these processes in the past decades, some uncertainties still exist, which could considerably influence the predictions of solar models \\cite{Adelberger}.  In order to try to reduce some of these uncertainties, we have performed a systematic study of some of the key reactions of the solar p-p chain by using a microscopic cluster model. The wave function of our model looks like \\begin{equation} \\Psi=\\sum_{L,S}{\\cal A} \\Bigg \\{ \\bigg [ \\Big [\\Phi^{A} \\Phi^{B} \\Big ]_S\\chi_L(\\mbox{\\boldmath $\\rho$}) \\bigg ]_{JM} \\Bigg\\}, \\label{wf} \\end{equation} where ${\\cal A}$ is the intercluster antisymmetrizer, $\\Phi$ are the cluster internal states, $\\mbox{\\boldmath $\\rho$}$ is the intercluster relative coordinate, while $L$, $S$, and $J$ are the orbital angular momentum, intrinsic spin, and total spin, respectively. The internal states of the $A\\leq 4$ clusters are simple $0s$ harmonic oscillator shell-model functions, while the heavier subsystems are described by two-cluster wave functions, similar to Eq.\\ (\\ref{wf}) (for example, $^7$Be inside $^8$B is described as a $^4{\\rm He}+{^3{\\rm He}}$ two-cluster system). The intercluster relative motions $\\chi$, which are the most important degrees of freedom, are treated with high precision, using variational methods \\cite{meth}. ",
        "conclusions": "We have studied some of the important nuclear reactions of the solar p-p chain, using microscopic models. We believe that in order to make significant further advances in the description of these processes, one should probably use a full $A$-body dynamical model. However, the fully consistent description of the bound and scattering states in such an approach will be a difficult task."
    },
    "0010/astro-ph0010421_arXiv.txt": {
        "abstract": "Spanish astronomy has grown rapidly during the 1990s.  The number of papers published per year has doubled, to about 4\\% of the world total.  The number of high-impact papers, however, remains constant at about 1\\% of the world total.  Papers from France, Germany and Italy also have lower impact than their numbers suggest, while for the US, UK, Japan, Netherlands and Australia, the fraction of high-impact papers is approximately the same as the fraction of all papers.  We suggest a few possible explanations for this difference. ",
        "introduction": "Professional astrophysics has a short tradition in Spain compared with other European countries. However, the inauguration of two world-class observatories (La Palma, Calar Alto) in Spain in the 1980s has allowed Spanish astronomy to develop rapidly in recent years, with a large increase in the number of astronomers, in the number of institutions, in public interest, and in funding.  We have made a quantitative study of the productivity of Spanish astronomy during the 1990s, in order to compare its standing with that of astronomy in other countries. ",
        "conclusions": "Spanish astronomy is a young science. The number of professional astronomers, number of papers per astronomer and fraction of papers (compared with the other countries) have all risen by a factor $\\sim$2 in the last decade. The fraction of Spanish astronomical papers has risen from $\\sim$2\\% in 1989 a $\\sim$4\\% in 1998. However, the fraction of high-impact work has remained constant at $\\sim$1\\% during this period. For Spain (and for France, Germany and Italy) the ratio between the number of high-impact papers, and the total number of papers published, is lower than for the US, UK, Japan, Netherlands and Australia. Possible reasons are language bias, smaller research budgets, or less competitive deployment of research resources."
    },
    "0010/astro-ph0010617_arXiv.txt": {
        "abstract": "I review the assumptions and observations that motivate the concept of the extragalactic cosmic background radiation, and the issues of energy accounts and star formation history as a function of galaxy morphological type that figure in the interpretation of the measurements of the extragalactic infrared background. ",
        "introduction": "It is useful to begin by recalling the basic assumptions and observations that lead us to the concept of the extragalactic cosmic background radiation, as opposed to radiation surface brightness that may be some highly variable function of position and direction.  Deep counts of objects detected at a broad range of wavelengths, from gamma ray sources to radio galaxies, are close to isotropic across the sky. It is an excellent bet therefore that the integrated radiation from observed sources plus those too faint to be detectable as individual objects also is quite to isotropic. This allows us to think of the local extragalactic radiation background as a function of one variable, the radiation energy density $u_\\nu$ per interval of frequency $\\nu$. The fluctuations around the mean as a function of position in the sky are important too, as a measure of large-scale structure, but $u_\\nu$ is the center of attention in these Proceedings.  The argument for large-scale homogeneity --- against a universe with a radial density gradient and us at the center --- is less direct but I think persuasive; my review of the considerations is in Peebles (1993). If we are persuaded then we conclude that within our Hubble length space is filled with a near uniform sea of radiation with spectral energy density $u_\\nu$: the cosmic extragalactic background radiation.  If the propagation of the radiation is described by a metric theory then it satisfies the Liouville or brightness theorem. If the metric describes a homogeneous isotropic spacetime then the geometry is fixed by the expansion factor $a(t)$, a function of the proper world time $t$ alone, together with the radius of curvature $a(t)R$ of sections of constant time, where the comoving radius $R$ is a constant. In this spacetime the radiation energy density $u(t)=\\int d\\nu\\, u_\\nu$ integrated over frequency at time $t$ is an integral over the history of production and absorption of radiation, \\begin{equation} u(t) = \\int ^t dt'\\, j(t')\\, (a(t')/a(t))^4. \\end{equation} At tine $t'$ the net rate of production of radiation (emitted minus absorbed) per unit proper volume is $j(t')$, and $j(t')\\, (a(t')/a(t))^3$ is the rate of production of energy per comoving volume normalized to the time $t$ of observation. The remaining factor in the integrand, $a(t')/a(t)=(1+z)^{-1}$, where $z$ is the redshift at the epoch $t'$ observed at time $t$, represents energy lost due to the cosmological redshift.  If spacetime were static, $a$ independent of time, equation~(1) says $j$ could not have been constant: there would have to have been a characteristic time at which star formation commenced. The point, associated with the name Olbers, is not often mentioned now; an edifying discussion is to be found in Bondi (1960). In the classical steady state cosmology (which also is well described by Bondi) the universe is expanding, $a\\propto e^{Ht}$, where $H$ is Hubble's constant. This makes the integral converge even when $j$ is constant, stars forming at a fixed mean rate per physical volume back to the indefinitely remote past. But we know now this is not a viable picture: Cowie and Lilly describe in these Proceedings observations of galaxies and an intergalactic medium at high redshift that are distinctly different from what is observed nearby; the more youthful appearance of objects at high redshift agrees with the interpretation that they are seen closer to the time when the structure we see started forming. In the general relativistic Friedmann-Lema\\^\\i tre model with a classical stress-energy tensor that satisfies $\\rho +3p>0$ the integral in equation~(1) has to have a lower limit, at the singular start of expansion at $a(t)=0$. In the eternal inflation scenario (Linde~1990) this unsatisfactory situation is relieved by the return to a steady state philosophy: the lower limit to the integral extends back along our world line to the remote past. ",
        "conclusions": "It has been widely and properly noted that we have significant observational evidence galaxies as we know them formed recently, at $z\\la 1$: the energy in the infrared extragalactic background, without large redshift dimming, is comparable to the binding energy of the heavy elements in normal $L\\ga L_\\ast$ galaxies, and the time integral of the seen star formation rate, which implies most stars are young, agrees with the seen mass in stars. But I have argued other lines of evidence indicate the cohort of stars and their heavy elements that formed at $z<1$ had to have ended up somewhere other than the high surface brightness parts of normal galaxies, or else are subdominant additions to the normal galaxies. True coincidences happen, of course: we have a sensible case for similar contributions to the sub-mm background by two quite different sources --- the radiation from stars and from AGNs --- with quite different fractions processed through dust.  The lesson is that untangling the relations among the extragalactic radiation background, the histories of star formation in different environments, and the accumulations of stars and heavy  elements, is likely to be a complicated operation. This is hardly surprising, considering that we live in a complicated universe. But the suite of observational evidence in hand is rich, rapidly growing, and fascinating to see sorted out."
    },
    "0010/astro-ph0010567_arXiv.txt": {
        "abstract": "We present observations with the adaptive optics assisted integral field spectrograph \\oasis\\ of the \\M double nucleus in the spectral domain around the Calcium triplet at a spatial resolution better than $0\\farcs5$~FWHM. These data are used to derive the two-dimensional stellar kinematics within the central 2\\arcsec. Archival \\wfpc/\\hst\\ images in the F300W, F555W and F814W bands are revisited to perform a photometric decomposition of the nuclear region. We also present  \\stis/\\hst\\ kinematics obtained from the archive.  The luminosity distribution of the central region is well separated into the respective contributions of the bulge, the nucleus including P1 and P2, and the so-called UV peak. We then show, using the \\oasis\\ kinematical maps, that the axis joining P1 and P2, the two local surface brightness maxima, does not coincide with the kinematic major-axis, which is also the major-axis of the nuclear isophotes (excluding P1). We also confirm that the velocity dispersion peak is offset by $\\sim 0\\farcs2$ from the UV peak, assumed to mark the location of the supermassive black hole. The newly reduced {\\tt \\stis/\\hst} velocity and dispersion profiles are then compared to \\oasis\\ and other published kinematics. We find significant offsets with previously published data. Simple parametric models are then built to successfully reconcile all the available kinematics.  We finally interpret the observations using new N-body simulations. The nearly keplerian nuclear disk of M31 is subject to a natural $m=1$ mode, with a very slow pattern speed (3~km/s/pc for $M_{\\rm BH} = 7\\, 10^7$~\\Msun), that can be maintained during more than a thousand dynamical times. The resulting morphology and kinematics of the mode can reproduce the M~31 nuclear-disk photometry and mean stellar velocity, including the observed asymmetries. It requires a central mass concentration and a cold disk system representing between 20 and 40\\% of its mass. Such a slow mode could be excited when interstellar clouds from the more external gaseous disk infall towards the centre.  Nuclear disks formed from accreted gas are possible candidates for the precursors of these types of structure, and may be common in central regions of galaxies. ",
        "introduction": "The role of supermassive black holes (hereafter SBHs) as power sources for active galactic nuclei is now widely accepted. The number of SBH candidates is growing rapidly and with more than 20 objects with good black hole mass estimates, one can start to search for statistical relationships between the black hole mass and the surrounding galaxy (de Zeeuw \\cite{Zee00} and references therein). Preliminary studies indicate that the SBH mass is correlated with the luminosity of the host galaxy (Magorrian \\etal\\ \\cite{Mag98}) and, with much less scatter, with its velocity dispersion (Ferrarese \\& Merritt \\cite{Fer00}, Gebhardt \\etal\\ \\cite{Geb00}). This suggests that the SBH mass is of the order of $10^{-3}$ times the mass of the bulges (Merritt \\& Ferrarese \\cite{Mer00}). If this relationship is confirmed, it would have important implications on our understanding of the formation of SBHs in galaxies. However the statistics is still biased to high SBH mass because of observational as well as modeling limitations. Extending the study to a larger number of objects and to lower SBH mass will require a large observational effort.  Detailed study of individual objects is fundamental not only to better constrain the SBH mass, but also to provide observational support on the origin of a possible relation between the SBH and global properties of the host galaxy. Theoretical work suggests that SBHs may influence the central morphology of galaxies through scattering of centrophilic stars (see e.g. the review by Merritt \\cite{Mer99}). Nuclear bars or spirals have been observed in the central region of galaxies, and thus proposed as a triggering mechanism for the growth of SBHs. Although this does not solve the problem of driving the dissipative component within the central parsec, it strongly suggests that the hypothesis of axisymmetry and steady-state widely used to model galaxy cores and constrain the SBH mass may not be adequate.  In this paper we will focus our attention on the nucleus of \\MP, which is an ideal case for such a detailed study: it is well resolved with a diameter of approximately 4\\arcsec\\ ($\\sim 15$~pc), even at ground-based spatial resolution, it is bright ($m_V \\sim 13$), does not suffer from dust absorption (contrarily to the Galactic centre) and has a favourable inclination ($i \\sim 70\\degr$). There is a long history of kinematical observations of this nucleus, with first evidence for a $\\sim10^7$~\\Msun\\ central SBH by Kormendy (\\cite{Kor88}) and Dressler \\& Richstone (\\cite{DR88}). It stood out then as one of the most convincing cases, since it was inferred from stellar kinematics obtained from high signal-to-noise ratio spectroscopic observation at 4~pc ($\\sim 1\\arcsec$) resolution. However the modeling was based on the unrealistic assumption of spherical symmetry. Upgrading the modeling to an axisymmetric (or triaxial) geometry would have been possible, but it was already known since the Stratoscope~II observations (Light \\etal\\ \\cite{LDS74}), that the light distribution was not even symmetric. This asymmetry was resolved in an intriguing double structure by Lauer~\\etal~(\\cite{Lau93}) using the \\hst\\ (pre-COSTAR) {\\tt WFPC} imaging capabilities. In the \\hst\\ original and post-COSTAR images (Lauer \\etal\\ \\cite{Lau98}), the double nucleus appears as a bright peak (P1) offset by $\\sim0\\farcs5$ from a secondary fainter peak (P2), nearly coinciding with the bulge photometric centre, and the suggested location of the SBH inferred from the previous spectroscopic observations (see the discussion in Kormendy \\& Bender \\cite{KB99}). \\hst\\ images from the far-UV to near-IR (King \\etal\\ \\cite{Kin95}; Davidge \\etal\\ \\cite{Dav97}) and long-slit spectra (Kormendy \\& Bender \\cite{KB99}) all demonstrate that P1 has the same stellar population as the rest of the nucleus, and that a nearly point-like source produces a UV excess close to P2 (King \\etal\\ \\cite{Kin95}).  The first available two-dimensional maps of the stellar kinematics of \\MP, obtained by Bacon \\etal\\ (\\cite{Bac94}) with the TIGER integral field spectrograph, showed another unexpected feature: while the stellar velocity field is roughly centred on P2, the peak in the velocity dispersion map is offset by $\\sim0\\farcs7$ on the anti-P1 side. Further \\hst\\ spectrographic observations were conducted with {\\tt FOS}, as well as \\foc\\ (Statler \\etal\\ \\cite{Sta99b}). In the following we will solely refer to the long-slit \\foc\\ observations of Statler \\etal\\ (\\cite{Sta99b}) as they are the only published data. At the \\hst\\ resolution, the velocity curve presents a strong gradient and the zero velocity point is offset by $\\sim 0\\farcs16$ from P2 towards P1. The velocity dispersion peak reaches a value of $\\sigma$ with $440\\pm70$~\\kms\\ (to be compared with the best corresponding ground-based value of $\\sim 248$~\\kms, Kormendy \\& Bender \\cite{KB99}). In the Statler \\etal\\ (\\cite{Sta99b}) data, the dispersion peak is nearly centred on P2 (within 0\\farcs06). The \\foc\\ measurements however suffer from a low signal-to-noise ratio (S/N hereafter) ($\\sim 14$ per pixel at P1/P2), and were obtained via a complex data reduction procedure to palliate some calibration problems. The shapes of the velocity and velocity dispersion curves should thus be confirmed with better S/N data.  Such data have been obtained by Kormendy \\& Bender (\\cite{KB99}) using the \\sis/\\cfht\\ spectrograph although with a significantly lower spatial resolution of $0\\farcs64$ FWHM. They confirmed that the dispersion peak is offset by $\\sim 0\\farcs2$ (in the direction opposite to P1) from their assumed velocity centre, and that the nucleus is significantly colder than the surrounding bulge.  On the modeling and theoretical fronts, the situation remains uncertain. There is a consensus regarding the existence of a SBH of a few $10^7$~\\Msun\\ although published values range from 3 to $10\\,10^7$~\\Msun. Such a black hole mass is needed to explain the high value of the velocity dispersion and the fast rotating stellar disk. Note however that the precise SBH mass cannot be accurately derived since we do not yet have a self-consistent model that can account for the observed properties. The SBH is assumed to coincide with the centre of the UV peak, near P2, and possibly with the recently uncovered hard X-ray emission detected by Chandra (Garcia \\etal\\ \\cite{Gar00}). In that scheme the position of the velocity dispersion peak should coincide with the location of the SBH, roughly consistent with the kinematics presented by Statler \\etal\\ (\\cite{Sta99b}), but not with the ground-based observations of Bacon \\etal\\ (\\cite{Bac94}) and Kormendy \\& Bender (\\cite{KB99}).  The nature of P1 and the observed dynamical pecularities remain a puzzle. While various possibilities have been discussed, only two hypotheses have been studied in some detail. Tremaine (\\cite{Tre95}) first proposed a model where an eccentric disk of stars orbiting the SBH at P2 produced the observed accumulation of light at P1. This parametric model was able to reproduce the light distribution and some of the kinematical features. Emsellem \\& Combes (\\cite{EC97}) built the first (and still only) self-consistent models for the nucleus of M~31, via N-body simulations. They performed simulations of a stellar cluster captured in the SBH potential and showed that the available observational properties could be reproduced well with properly tuned orbital elements for the falling cluster. However the timescale for the disruption of the cluster is quite small ($\\sim$ \\ten{5} years), significantly weakening this scenario. More problematic is the observed homogeneity in the colours of P2 and P1 whose explanation would require some more fine tuning of the stellar population of the cluster.  As emphasized by Statler \\etal\\ (\\cite{Sta99b}) and Kormendy \\& Bender (\\cite{KB99}), the Tremaine (\\cite{Tre95}) model then remained the only attractive solution: it naturally predicts the same stellar population for P1 and the rest of the nucleus and seems to be, at least qualitatively, compatible with the observed kinematics. Statler~(\\cite{Sta99a}) noted that it may also help to explain a wiggle observed in the \\foc\\ velocity profile near the location of P1. In Tremaine's original paper, some preliminary suggestions were given for the possible formation of such an eccentric disk. These ideas have yet to be confirmed by more realistic self-consistent models. Such an investigation is also important to determine if eccentric disks are common in SBH environments.  The goal of this paper is twofold: (i) add more observational constraints to the photometry and kinematics of \\M nucleus and eliminate the uncertainty regarding the location of the velocity dispersion peak; (ii) investigate Tremaine's model or alternative solutions in more detail.  To achieve these goals, we have performed new observations with the adaptive optics assisted integral field spectrograph \\oasis\\ (Section~\\ref{sec:oasis}). Although these new kinematical maps have a factor two better spatial resolution than previous two-dimensional spectroscopic data (Bacon \\etal\\ \\cite{Bac94}), they cannot compete with the \\hst\\ spatial resolution. The relative centring and orientation of the various key features observed in the nucleus are also critical. We therefore performed a detailed analysis of the photometry using archival \\hst/\\wfpc\\ images (Sect.~\\ref{sec:photometry}). We thus combined them with new velocity and velocity dispersion profiles derived from the recently released \\stis/\\hst\\ data (Section~\\ref{sec:stis}). Together, these data provide new constraints on the dynamics of \\M (Sect.~\\ref{sec:results}). We finally interpret the observations in terms of slow $m=1$ modes and present new N-body simulations to support this scenario in Section~\\ref{sec:simu}. Conclusions are given in Sect.~\\ref{sec:conclusion}. ",
        "conclusions": "\\label{sec:conclusion}  In this paper, we have first tried to clarify a number of issues concerning the morphology and colours of the nucleus of M~31. The photometry in the nuclear region of M~31 can thus be decomposed in 3 distinct components: \\begin{itemize} \\item The bulge: a hot slowly rotating system with central colour values of $(V-I)_B = 1.26$, $(U-V)_B = 1.98$ \\item The nucleus: exhibiting two peaks in the $I$ band, the so-called P1 and P2, both being offcentred with respect to the centre of the bulge isophotes. Colours are suprisingly homogeneous throughout the nucleus after the bulge contribution has been subtracted, confirming previous results, as e.g. emphasized in KB99. However, contrarily to the analysis of Lauer \\etal\\ (\\cite{Lau98}), we found that the nucleus is redder than the bulge in both $V-I$ and $U-V$ with $(V-I)_N = 1.30$, $(U-V)_N = 2.36$. \\item A central blue excess, the so-called UV peak, mostly apparent in the F300W/WFPC2 image, where it is just resolved. This component was first detected and discussed by King \\etal\\ (\\cite{Kin95}). It has an axis ratio of about $0.7$, a major-axis FWHM of $\\sim 0\\farcs21$ and a PA of $\\sim 62\\degr \\pm 8$. \\end{itemize}  We have then presented new \\oasis\\ and \\stis\\ data which complete our knowledge of the complex kinematics within the central 2'' of \\MP. Taking the centre of the UV peak in the F300W WFPC2 image as our reference zero point, we have derived the respective offsets required to reconcile the FOC data of Sta+99, the SIS data of KB99, and the newly presented STIS and \\oasis\\ kinematics. The main results are summarized here: \\begin{itemize} \\item The kinematic axis (PA$=56.4\\degr$) is nearly coincident with the major-axis of the nucleus but not with the P1-P2 axis (PA$= 42\\degr$). \\item The strong asymmetry of the velocity profile along the P1--P2 axis observed by Sta+99 is confirmed. On the P1 side, the maximum velocity in the \\stis\\ kinematics ($201\\pm 5$~\\kms) is reached at $0\\farcs63$ (the turnover being at a radius of $0\\farcs42$ with $V = 197\\pm5$~\\kms), while on the other side the minimum velocity($-292\\pm21$~\\kms) is at $-0\\farcs235$. \\item The velocity dispersion peak is offset from the UV peak on the anti-P1 side. It reaches $321\\pm33$~\\kms\\  at $-0\\farcs235$ in \\stis\\ measurements. \\item The velocity dispersion peak is consistent with being the result of pure velocity broadening, and is not consistent with a hot system located at the location of the UV peak. High velocities seem to be associated with P2 which is observed to be offset from the UV peak on the anti-P1 side. \\end{itemize}  We finally performed new N-body simulations of $m=1$ modes in a thin disk including a SBH in its centre. We have found that such $m=1$ modes can be maintained from more than a thousand dynamical times, when the disk mass represents between 20 and 40\\% of the central mass, allowing self-gravity to counteract the differential precession of the disk. The pattern speed of this $m=1$ mode is small, of the order of 3~km/s/pc in a case designed to resemble M~31's nucleus. For the present paper, we have only presented a model where the $m=1$ mode was excited by launching particles along elliptical orbits. This model reproduces reasonably well the observed asymmetry in the surface brightness distribution and mean stellar velocity, although we had to remove some particles near the SBH to avoid numerical artefacts. A detailed comparison with the observed stellar velocity dispersion requires a realistic treatment of the particles near the SBH. The existence of an $m=1$ mode in the nucleus of M~31 is consistent with the presence of a SBH with a mass in the range $\\sim 3.5-8.5\\, 10^7$ M$_\\odot$.  The model proposed here relies on a relatively thin and cold nuclear stellar disk. How can this disk be formed and maintained? A likely possibility is that the central region is subject to accretion and infall of gas from the nearby disk of M~31, which is gas rich. Already, within the few 10-100~pc dust lanes, and CO molecular clouds are observed (Melchior \\etal\\ \\cite{Mel00}). The infall of gas clouds could both trigger the $m=1$ instability, and also progressively reform the cold stellar disk."
    },
    "0010/astro-ph0010444_arXiv.txt": {
        "abstract": "{ This paper is part of a series reporting the results of the ATESP radio survey obtained at 1.4~GHz with the Australia Telescope Compact Array. The survey consists of 16 radio mosaics with $\\sim 8\\arcsec \\times 14\\arcsec$ resolution and uniform sensitivity ($1 \\sigma$ noise level $\\sim$79~$\\mu$Jy) over the whole area of the ESO Slice Project redshift survey ($\\sim 26$~sq.~degrees at $\\delta \\sim -40\\degr$). The ATESP survey has produced a catalogue of 2960 radio sources down to a flux limit ($6\\sigma$) of $\\sim 0.5$ mJy. \\\\ \\indent In this paper we present the 1.4 GHz $\\log N - \\log S$ relation derived from the ATESP radio source catalogue. The possible causes of incompleteness at the faint end of the source counts are extensively discussed and their effects are quantified and corrected for. The ATESP counts are consistent with others reported in the literature, even though some significant discrepancies are present at low fluxes (below a few mJy). We investigate whether such discrepancies may be explained in terms of field-to-field anisotropies, considering the fact that all the existing sub-mJy surveys cover small areas of sky (from a fraction of square degree to a few square degrees). We stress that the ATESP survey, covering 26~sq.~degrees, provides the best determination of source counts available today in the flux range $0.7\\la S_{1.4\\;\\rm{GHz}}\\la 2$ mJy. ",
        "introduction": "\\label{sec-intr}  The statistical study of radio sources allows us to address a number of astrophysical and cosmological problems, ranging from the evolution of classical radio galaxies and quasars to the properties of starburst galaxies as a function of cosmic time. The radio source counts are the most immediate product which can be derived from a radio survey and reflect the statistical properties of the radio source populations. The slope of the counts is determined essentially by the relative contribution of different types of sources at every flux, which is the result of their luminosity functions at various redshifts. The source counts therefore represent the most immediate observational constraint to evolutionary models of radio sources. An accurate determination of the source counts, their slope and normalization, is however necessary in order to make this constraint really useful. The source counts, together with measured local radio luminosity functions, can then be used to make predictions (about the redshift distribution, for instance) which can be verified once an extensive program of optical identification and spectroscopy has been completed in some specific flux range. \\\\ \\indent Normalized differential source counts derived from deep 1.4 GHz surveys show a flattening below a few mJy. This change in slope is usually interpreted as the result of the emergence of a new population which does not appear at higher flux densities, where the counts are dominated by the classical powerful radio galaxies and quasars ($99\\%$ above 60 mJy, Windhorst et al. 1990). This new population is thought to be a mixture of several different types of objects (faint AGNs, normal spirals and ellipticals, starburst galaxies, etc.) whose relative importance changes as a function of the flux itself. Unfortunately the existing samples of faint radio sources are confined to regions of relatively small angular sizes and no complete optical follow--up of such samples has been carried out yet (see {\\it e.g.} Gregorini \\& Prandoni 2000 for a review). It follows that conclusions about the composition of the faint radio sources are based on small and only partially identified samples, biased by the fact that only the brightest optical counterparts have spectral information. Many questions are therefore still open about the true nature and evolution of the faint radio source population and very little is known about the luminosity properties and redshift distribution of such sources. \\\\ \\indent The ATESP 1.4 GHz survey (Prandoni et al. 2000a, paper I) - carried out with the Australia Telescope Compact Array in the past years - consists of 16 radio mosaics with $\\sim 8\\arcsec \\times 14\\arcsec$ resolution and uniform sensitivity ($1 \\sigma$ noise level ranging from 69 $\\mu$ Jy to 88 $\\mu$Jy depending on the radio mosaic) covering a narrow strip of $26\\degr \\times 1\\degr$ at $\\delta \\sim -40\\degr$. The ATESP survey has produced a catalogue of 3172 radio components, corresponding to 2960 distinct radio sources, down to a flux limit ($6\\sigma$) of $\\sim 0.5$ mJy (Prandoni et al. 2000b, Paper II). The ATESP sample represents the largest sample (26~sq.~degr.) sensitive to sub--mJy fluxes available so far. It offers a unique opportunity to investigate the nature and evolution of the mJy and sub-mJy populations and provides a very accurate determination of the radio source counts down to sub-mJy fluxes (the most accurate in the range $0.7 - 2$ mJy). In this respect the ATESP survey fully complements the FIRST (White et al. 1997), which provides the most accurate determination of the source counts between 2 and 30 mJy. \\\\  This paper is organized as follows. The catalogue completeness is discussed in Sect.~\\ref{sec-corr}. The ATESP number counts are derived in Sect.~\\ref{sec-counts} and the possible effect of clustering on the faint end of the counts  is discussed in Sect.~\\ref{sec-clust}. Results are summarized in Sect.~\\ref{sec-summary}. ",
        "conclusions": "\\label{sec-summary}  We have presented the $\\log{N} - \\log{S}$ relation derived from the ATESP catalogue. Particular emphasis has been given to the possible causes of incompleteness at the faint end of the source counts. In particular we have discussed the effect of systematic under-estimations in the measured flux densities due to clean bias and bandwidth smearing and we have quantified the incompleteness of the catalogue in terms of total flux density (the so--called resolution bias). Once corrected for such effects, the ATESP counts are consistent with the counts derived from other recent surveys. The only discrepant counts are those obtained from the Phoenix Deep Survey (Hopkins et al. 1998), which are systematically higher. It is worth to notice that the ATESP counts point towards lower fluxes ($S\\la 1$ mJy) for the upturn than those ($S\\la 5$ mJy) indicated by the Windhorst et al. (1990) best fit.  \\\\ \\indent At fluxes below 1 mJy, the 1.4 GHz number counts derived from different surveys show a large spread. Such a spread can be due to technical reasons, like different corrections for resolution bias or no correction at all. The effect of field-to-field anisotropies and/or clustering could also play an important role, since the faintest samples typically cover very small regions of sky ($<< 1$ sq.~degree). This latter effect does not apply to the ATESP sample which covers 26 sq.~degrees. We have investigated such a possibility by splitting the ATESP catalogue in smaller sub--samples. In particular we have tested two typical scales for sub--mJy radio surveys: a few~sq.~degrees for the largest surveys ($1-2$~sq.~degr. in this case) and $\\sim 0.33$~sq.~degrees for the deepest surveys. As a result we have found that on both scales the spread in the counts can be simply explained in terms of Poissonian fluctuations, at least down to the flux densities probed by the ATESP. The only exception is represented by the Phoenix Deep Survey whose counts remain systematically higher than the ATESP counts. Such a discrepancy is statistically relevant and draws us to conclude that clustering and/or some other unknown systematic effect should play an important role in the case of the PDF sample. \\\\ \\indent The ATESP survey provides the best determination of the 1.4 GHz source counts in the flux range 0.7--2 mJy, and in this respect it complements the FIRST (White et al. 1997) which provides the most accurate counts available at higher fluxes (2--30 mJy). The ATESP source counts can thus be used to set observational constraints on the evolutionary models for the mJy and sub--mJy populations."
    },
    "0010/astro-ph0010358_arXiv.txt": {
        "abstract": "The `maximum' disk hypothesis of galactic disks imbedded in dark matter halos is examined. First, decompositions of the rotation curves of NGC\\,2613, 3198, 6503, and 7184 are analyzed. For these galaxies the radial velocity dispersions of the stars have been measured. If the parameters of the decompositions are chosen according to the `maximum' disk hypothesis, the Toomre $Q$ stability parameter is sytematically less than one, which is a strong argument against the `maximum' disk hypothesis. Next, density wave theory arguments are used to describe the morphology of the spiral arms of NGC\\,3223, 157, and 7083. It is shown that the `maximum' disk hypothesis is not consistent with the observed morphologies of the galaxies. ",
        "introduction": "An important aspect of the dark matter problem of spiral galaxies is the question to what extent the galaxies are dominated by dark matter even in their inner parts, where the optically visible disks reside. The presence of dark matter is usually deduced from the decompositions of the rotation curves of the galaxies. These are, however, highly ambiguous, and `maximum' disk versus submaximal disk decompositions of the rotation curves of spiral galaxies have been discussed at great length in the literature. The general opinion is that the inner parts of the galaxies are not dominated by dark matter (cf.~Bosma (1999) for a recent review). However an increasing number of arguments have been put forward, which",
        "conclusions": ""
    },
    "0010/astro-ph0010391_arXiv.txt": {
        "abstract": "We have used the {\\it Submillimeter Wave Astronomy Satellite} to map the ground-state $1_{10}\\rightarrow1_{01}$ transition of ortho-\\water\\ at 557 GHz in the Orion molecular cloud.  \\water\\ emission was detected in Orion over an angular extent of about 20\\arcmin, or nearly 3 pc. The water emission is relatively weak, with line widths (3--6 km s$^{-1}$) and $V_{\\rm LSR}$ velocities (9--11 km s$^{-1}$) consistent with an origin in the cold gas of the molecular ridge.  We find that the ortho-\\water\\ abundance relative to H$_2$ in the extended gas in Orion varies between 1 and 8$\\times10^{-8}$, with an average of 3$\\times10^{-8}$. The absence of detectable narrow-line ortho-\\wateriso\\ emission is used to set a $3\\sigma$\\ upper limit on the relative ortho-\\water\\ abundance of 7$\\times10^{-8}$. ",
        "introduction": "We present a map of the water emission from the central portion of the Orion molecular cloud obtained with the {\\it Submillimeter Wave Astronomy Satellite} (\\swas).  These results provide the first definitive detection of water in the extended, cold molecular gas. The Orion molecular cloud has been the focus of many studies of water, and besides the strongly masing transition at 22 GHz, numerous transitions of \\water\\ and \\wateriso\\ have been detected from the ground \\citep*{wat80,phi80,kui84, cer94,gen96,cer99}, from the {\\it Kuiper Airborne Observatory} \\citep{zmu95,tim96}, and from the {\\it Infrared Space Observatory} \\citep{har98, gon98, dis98}.  With the exception of the broad \\wateriso\\ line emission observed toward BN/KL by \\citet{zmu95}, all of the transitions detected have upper states lying from 200 to 700 K above the ground state and many are likely masing. In most cases this emission arises from the hot gas associated with BN/KL, however \\cite{cer94} report spatially extended emission from the $3_{13}\\rightarrow 2_{20}$ transition of water at 183 GHz.  \\swas\\ observed the $1_{10} \\rightarrow 1_{01}$ transitions of ortho-\\water\\ at 556.936 GHz and of ortho-\\wateriso\\ at 547.676 GHz.  These transitions have upper state energies of only 27 K above the ortho-\\water\\ ground state, and thus are unique probes of water in the cooler molecular gas. The emission obtained with \\swas\\ toward BN/KL \\citep{mel00a} is composed of both a broad velocity feature associated with the shocked gas and a narrow feature that likely arises from the quiescent gas in the Orion molecular ridge.  The abundance of water relative to \\hh\\ in the shocked gas is of order $4\\times10^{-4}$, making water one of the most abundant molecules in the hot gas. The focus of this Letter is the kinematically narrow and spatially extended component of the water emission detected in Orion. ",
        "conclusions": "\\swas\\ has made the first detection of water emission from the cold, quiescent gas in the Orion molecular cloud.  The water emission seen by \\swas\\ has line widths and velocities characteristic of the emission from the molecular ridge. The average abundance of ortho-\\water\\ in the cold gas is $3\\times10^{-8}$, with variations between 1 and 8$\\times10^{-8}$. As discussed earlier, the abundance derived from the \\water\\ emission can be compromised by several physical processes, however a firm  $3\\sigma$\\ upper limit on the relative water abundance 3\\farcm2 south of BN/KL of $7\\times10^{-8}$ is established from the \\wateriso\\ observation.  Uncertainties in the water abundance arise entirely from systematic errors associated with uncertainties in our estimates of the density and gas column density.  For instance, if a significant fraction of the gas column density is associated with much lower density gas that we assumed in our model, then our estimate of the ortho-water abundance will be underestimated.  Specifically, if one-half of the gas column density originated in low-density gas, our estimate of the water abundance will be too low by approximately a factor of two.  Since \\citet{ber96} concluded that it is unlikely that the low density gas contributes significantly to the total gas column density toward the Orion core, we believe that our estimates of the water abundance cannot be greatly in error.  \\citet{cer94} estimated the water abundance in Orion based on their observations of the  $3_{13} \\rightarrow 2_{20}$ transition of para-\\water\\ at 183 GHz.  Although this masing transition arises from an upper state lying 200 K above the ground state, the emission is spatially extended and, away from BN/KL, has relatively narrow line widths. \\citet{cer94}  estimate a relative water abundance of at least 10$^{-5}$. However, if the abundance of water in the cold molecular gas were as large as they estimate, then \\wateriso\\ emission should have been readily detected by \\swas\\ and be of comparable strength to the detected \\water\\ emission.  In addition, the \\water\\ line observed by \\swas\\ should be between 5 and 100 times stronger than observed, depending on position. The 183 GHz line must trace a warmer, denser component of the gas than the 557 GHz line - as suggested by the anomalous line velocity reported by \\citet{cer94}  - so that the high abundance estimate does not apply to the bulk of the cloud material, which is at low temperature.  The water abundance we derive for the quiescent gas in Orion is many orders of magnitude smaller than that derived for the hot gas in Orion and other massive star forming regions \\citep{mel00a,cer99,dis98b}.  Chemical models \\citep*{ber98} predict equilibrium gas-phase water abundances of order $5\\times10^{-7}$ in the quiescent gas, similar to the results found for the extended envelope surrounding SgrB2 \\citep{zmu95}, but about 10 times larger than what we estimate for Orion.  Further discussion of the chemical implications of these results is presented in \\citet{ber00}."
    },
    "0010/astro-ph0010028_arXiv.txt": {
        "abstract": "This review aims to give an overview of the contribution of the {\\it Hubble Space Telescope} to our understanding of the detailed properties of Local Group dwarf galaxies and their older stellar populations.  The exquisite stable high spatial resolution combined with photometric accuracy of images from the {\\it Hubble Space Telescope} have allowed us to probe further back into the history of star formation of a large variety of different galaxy types with widely differing star formation properties. It has allowed us to extend our studies out to the edges of the Local Group and beyond with greater accuracy than ever before.  We have learnt several important things about dwarf galaxy evolution from these studies. Firstly we have found that no two galaxies have identical star formation histories; some galaxies may superficially look the same today, but they have invariably followed different paths to this point. Now that we have managed to probe deep into the star formation history of dwarf irregular galaxies in the Local Group it is obvious that there are a number of similarities with the global properties of dwarf elliptical/spheroidal type galaxies, which were previously thought to be quite distinct. The elliptical/spheroidals tend to have one or more discrete episodes of star formation through-out their history and dwarf irregulars are characterized by quasi-continuous star-formation. The previous strong dichotomy between these two classes has been weakened by these new results and may stem from the differences in the environment in which these similar mass galaxies were born into or have inhabited for most of their lives. The more detailed is our understanding of star formation processes and their effect on galaxy evolution in the nearby Universe the better we will understand the results from studies of the integrated light of galaxies in the high-redshift Universe. ",
        "introduction": "This review is a survey of the {\\it Hubble Space Telescope} (HST) studies of resolved stellar populations of nearby galaxies that have determined accurate global star formation histories going back several Gyr. The determination of such a detailed the star formation history depends upon the ability to accurately photometer many Gyr old stars.  This was possible from ground based imaging for our nearest companions, the dwarf spheroidals, and the Magellanic Clouds. HST, with it's outstanding combination of lower sky brightness, high resolution, and a stable and constant point spread function, has allowed us to to probe further out into the Local Group and even beyond this to look at a much more diverse sample of galaxies than previously possible, and it has also allowed us renewed insights into our nearest neighbours.  I am leaving out any discussion of the considerable body of literature on the HST observations of the stellar populations of the Magellanic Clouds ({\\it e.g.}, Holtzman {\\it et al.} 1999; Panagia {\\it et al.} 2000) which technically do not count as dwarf galaxies ({\\it e.g.}, Tammann 1994; Binggeli 1994) but are often assumed to be so.  Because these galaxies are so large on the sky it is difficult for HST by itself to gain a perspective of the global star-formation properties, although attempts are being made ({\\it e.g.}, Smecker-Hane {\\it et al.} 1999).  There are also numerous HST studies of stellar populations at the small scale of individual star clusters and H~II regions in the Clouds ({\\it e.g.}, Da~Costa~1999; Massey~1999; Hunter~1999).  Studies of the stellar populations of the Magellanic Clouds could easily take up an entire review, not to say a conference, in their own right. I have also neglected considerable work on resolved stellar populations of dwarfs in the UV ({\\it e.g.} Brown {\\it et al.} 2000; Cole {\\it et al.} 1998), because the detailed connection to quantifying a star formation history is unclear. This is also true of studies of star clusters around nearby dwarf galaxies ({\\it e.g.}, Da~Costa~1999; Hodge {\\it et al.} 1999; Mighell, Sarajedini \\& French 1998).  \\begin{table} \\begin{center} \\caption{The Local Group}  \\begin{tabular}{llllccl}  name& type    & $M_V$   & $\\Sigma_0$& M$_{tot}$& M$_{tot}$/M$_{HI}$ &[Fe/H] \\\\[3pt] &\t  &\t    &mag arcsec$^{-2}$&10$^6\\Msun$&& dex  \\\\ \\\\ {\\it Spiral galaxies:}   &&&&&&\\\\ M~31\t\t& Sb      & $-$21.2 & 10.8&2$\\times 10^6$& 0.002& +0.2\t\\\\ Milky Way\t& Sbc\t  & $-$20.9: &     & 10$^6$& 0.004 & ~~~0.\t\t\\\\ M~33\t\t& Sc\t  & $-$18.9 & 10.7&10$^5$&0.02 &$-$0.2 \t\t\\\\ \\\\ {\\it Irregular galaxies:}&&&&&&\\\\ NGC~3109\t& Irr\t  & $-$15.7 & 23.6$\\pm0.2$&6550\t& 0.11\t&$-1.2\\pm0.2^{a}$\\\\ LMC\t\t& Irr\t  & $-$18.1 & 20.7&6000\t& 0.5\t&$-$0.7\t\t\\\\ SMC\t\t& Irr\t  & $-$16.2 & 22.1& 2000& 0.25\t&$-$1.0\t\t\\\\ \\\\ {\\it Dwarf Ellipticals:} &&&&&&\\\\ M~32\t\t& dE2\t  & $-$16.7 & $<$11.5:&2120& $<$0.001&$-1.1\\pm0.2$\\\\ NGC~205\t\t& dE5\t  & $-$16.6 & 20.4$\\pm0.4$& 740\t& 0.001\t&$-0.8\\pm0.1$\\\\ NGC~185\t\t& dE3     & $-$15.5 & 20.1$\\pm0.4$& 130\t& 0.001\t&$-1.2\\pm0.15$\\\\ NGC~147 \t& dE4\t  & $-$15.5 & 21.6$\\pm0.2$& 110\t& $<$0.001&$-1.1\\pm0.2$\\\\ Sagittarius\t& dE7\t  & $-$13.4 & 25.4$\\pm0.3$& 500:& $<$0.0001&$-1.0\\pm0.2$\\\\ \\\\ {\\it Dwarf Irregulars:}  &&&&&&\\\\ NGC~6822\t& dIrr\t  & $-$15.2 & 21.4$\\pm0.2$&1640 & 0.08\t&$-0.7\\pm0.2^{a}$\\\\ IC~10 \t\t& Irr\t  & $-$15.7 & 22.1$\\pm0.4$& 1580& 0.10\t&$-0.7\\pm0.15^{a}$\\\\ IC~1613\t\t& dIrr\t  & $-$14.7 & 22.8$\\pm0.3$& 795 & 0.07\t&$-1.1\\pm0.2^{a}$\\\\ IC~5152\t\t& dIrr\t  & $-$14.8 &     & 400 & 0.15\t&$-0.6\\pm0.2^{a}$\\\\ Sextans~B\t& dIrr\t  & $-$14.2 &    & 885  & 0.05\t&$-1.1\\pm0.3^{a}$\\\\ Sextans~A\t& dIrr\t  & $-$14.6 & 23.5$\\pm0.3$& 395 & 0.20\t&$-1.4\\pm0.2^{a}$\\\\ WLM   \t\t& dIrr    & $-$14.5 & 20.4$\\pm0.05$& 150 & 0.40\t&$-1.1\\pm0.2^{a}$\\\\ Phoenix\t\t& dIrr/dE & $-$10.1 &     &  33 & 0.006:&$-1.9\\pm0.1$\\\\ Pegasus\t\t& dIrr\t  & $-$12.9 &     &  58 & 0.09\t&$-1.0\\pm0.14^{a}$\\\\ LGS~3\t\t& dIrr/dE & $-$10.5 & 24.7$\\pm0.2$&  13 & 0.03\t&$-1.8\\pm0.3$\\\\ Leo~A\t\t& dIrr\t  & $-$11.4 &     &  11 & 0.72\t&$-1.6\\pm0.15^{a}$\\\\ SagDIG\t\t& dIrr\t  & $-$12.3 & 24.4$\\pm0.3$&  9.6& 0.92\t&$-1.5\\pm0.3^{a}$\\\\ DDO~210\t\t& dIrr\t  & $-$10.0 & 23.0$\\pm0.3$&  5.4& 0.35\t&$-1.9\\pm0.12$\\\\ EGB 0427+63\t& dIrr\t  & $-$12.6 & 23.9$\\pm0.1$&     &\t&$-1.4\\pm0.1^{a}$\\\\ \\\\ {\\it Dwarf Spheroidals:} &&&&&&\\\\ Fornax\t\t& dE3\t  & $-$13.2 & 23.4$\\pm0.3$&  68 & $<$0.001&$-1.3\\pm0.2$\t\\\\ Ursa~Minor\t& dE5\t  & $-$8.9  & 25.5$\\pm0.5$&  23 & $<$0.002&$-2.2\\pm0.1$\t\\\\ Draco\t\t& dE3\t  & $-$8.8  & 25.3$\\pm0.5$&  22 & $<$0.001&$-2.1\\pm0.15$\\\\ Leo~I\t\t& dE3\t  & $-$11.9 & 22.4$\\pm0.3$&  22 & $<$0.001&$-1.5\\pm0.4$\t\\\\ Sextans\t\t& dE4\t  & $-$9.5  & 26.2$\\pm0.5$&  19 & $<$0.001&$-1.7\\pm0.2$\t\\\\ Carina\t\t& dE4\t  & $-$9.3  & 23.9$\\pm0.4$&  13 & $<$0.001&$-2.0\\pm0.2$\t\\\\ Sculptor\t& dE\t  & $-$11.1 & 23.7$\\pm0.4$&  6  & $<$0.004&$-1.8\\pm0.1$\t\\\\ Antlia\t\t& dE3\t  & $-$10.8 & 24.3$\\pm0.2$&  12 & 0.08\t  &$-1.8\\pm0.25$\\\\ Tucana\t\t& dE5\t  & $-$9.6  & 25.1$\\pm0.06$&     & \t  &$-1.7\\pm0.15$\\\\ Cetus\t\t& dE4\t  & $-$10.1 & 25.1$\\pm0.1$&     & \t  &$-1.9\\pm0.2$\t\\\\ Leo~II\t\t& dE0\t  & $-$9.6  & 24.0$\\pm0.3$&  10 & $<$0.001&$-1.9\\pm0.1$\t\\\\ And~I\t\t& dE0\t  & $-$11.9 & 24.9$\\pm0.01$&     & \t  &$-1.5\\pm0.2$\t\\\\ And~II\t\t& dE3\t  & $-$11.1 & 24.8$\\pm0.05$&     & \t  &$-1.5\\pm0.3$\t\\\\ And~III\t\t& dE6\t  & $-$10.3 & 25.3$\\pm0.05$&     &         &$-2.0\\pm0.2$ \\\\ And~V\t\t& dE3\t  & $-$9.1  & 24.8$\\pm0.2$&    &\t  &$-1.6\\pm0.2$\t\\\\ And~VI (Peg dSph)& dE3\t  & $-$11.3 & 24.3$\\pm0.05$&     &\t  &$-1.6\\pm0.2$\t\\\\ And~VII (Cass dSph)& dE3     & $-$12.0 & 23.5$\\pm0.05$&     &\t  &$-1.4\\pm0.3$\t\\\\  \\end{tabular} $^a$ these values were converted from [O/H] measurements, assuming constant [Fe/O]=0. \\end{center} \\end{table}  \\subsection{Dwarf Galaxy Types:}  Classification is a difficult and often emotive buisness, and I do not attempt to seriously address this issue, except to make it easier for me to refer to the sample of galaxies in the Local Group in broad terms, and with my particular interest in their star-forming properties.  So, bearing this caveat in mind let me introduce the four different classes of dwarf galaxies that should cover anything out there:  {\\it Dwarf Irregular} (dI) galaxies are arguably the most common type of galaxy in the Universe ({\\it cf}. Ellis 1997), they are not clustered around larger galaxies, but appear to have a fairly random distribution through out the Local Group, and indeed in the Universe. They are usually loosely structured late-type gas rich systems with varying levels of star formation occurring in a haphazard manner across the galaxy.  The velocity field of the HI gas in these systems can be dominated by random motions rather than rotation ({\\it e.g.}, Lo, Sargent \\& Young 1993, but see Skillman 1996), for the fainter dIs ({\\it e.g.}, Leo~A; v$_{rot} \\sim 5$km/s), but for the more massive dIs ({\\it e.g.}, NGC~6822, Sextans~A) solid body rotation is clearly seen, with amplitudes of $30-40$km/s.  {\\it Blue Compact Dwarf} (BCD) galaxies are gas rich systems dominated by a region of extremely active star formation, and resembling the massive HII regions which can be found in larger galaxies. They are thought to be forming stars at a rate which they can only maintain for a short period ({\\it e.g.}, Searle, Sargent \\& Bagnuolo 1973).  This type of galaxy may be a dI undergoing a period of particularly active star formation ({\\it e.g.}, Tolstoy 1998a).  Within the Local Group, IC~10 is a fairly good approximation of what we expect a BCD to look like (see van den Bergh 2000; Hunter {\\it et al.} in prep.), and perhaps also IC~5152. The more distant BCDs could easily be embedded in larger low surface brightness galaxies, which are easier to see within the Local Group. There are several, classical, examples of BCDs just beyond the Local Group ({\\it e.g.} NGC~1569 \\& VII~Zw403).  {\\it Dwarf Elliptical} (dE) galaxies are basically low luminosity Elliptical galaxies, with smooth surface brightness distributions ({\\it e.g.}, Ferguson \\& Binggeli 1994).  They are typically dominated by an old stellar population, but as Baade already noticed in 1951, they are subject to the same extreme variations of stellar population as other dwarf galaxy types. Baade (1951) found that the archetypal dEs NGC~185 and NGC~205 contain B stars along with gas and dust.  Recent detailed analysis suggests that several epochs of star formation over long time scales are needed to explain the characteristics of dE stellar populations ({\\it e.g.}, Ferguson \\& Binggeli 1994; Han {\\it et al.} 1997). Most of the bright dEs (M$_B < -16$) appear to have nuclei, and there is some evidence that these are dynamically separate super-massive star clusters ({\\it e.g.}, M~54 in Sagittarius, Ibata et al. 1994; and NGC~205, Carter \\& Sadler 1990).  dEs are strongly clustered with the largest galaxies, and four of the five dEs in the Local Group are found in proximity to M~31.  {\\it Dwarf Spheroidal} (dSph) galaxies are basically low-surface brightness, non-nucleated dEs. Many argue that the dSph are merely the low luminosity tail of the dE galaxy class ({\\it e.g.}, Ferguson \\& Binggeli 1994), and the fact that they are {\\it often} clustered around bigger galaxies such as our Galaxy, M~31, and perhaps also NGC~3109 tends to support this. However it is also possible that at least a fraction of this class fit into a common evolutionary scenario with dIs and BCDs, their past star-formation having been dominated by bursts ({\\it e.g.} Carina).  There are a number of so called {\\it transition} objects, such as Phoenix, Antlia and LGS~3 which are hard to fit unambiguously into either the dE or dI class. They have very little or no present-day star formation, and yet they have associated HI gas, so it seems a fair assumption that they will form stars again before too long, and will then clearly belong to the dI class.  \\subsection{The Properties of Local Group Galaxies}  Our local neighbourhood, the Local Group, is arguably as representative a piece of the Universe as any ({\\it e.g.}, van den Bergh 2000).  What we learn about the properties of star formation and galaxy evolution here can justifiably be extrapolated to explain what we see in the distant, early Universe.  That which is on our doorstep provides us with the chance to properly understand the dominant physical processes in great detail.  As with galaxy classification, there are different selection criteria which vary the number of dwarf and irregular galaxies included in a census of Local Group members.  I have chosen to follow dynamical arguments ({\\it e.g.}, Irwin 1998), which tends to allow a slightly more distant limit to the distance from the centre than van den Bergh (2000). There is no straight forward boundary between the Local Group and the Sculptor, and many of the galaxies in this region could ``belong to either'', by dynamical arguments. Listed in Table~1 are some basic properties of the galaxies I have assumed to be members of the Local Group.  For each galaxy there is listed the absolute V magnitude (M$_V$), the central surface brightness in V ($\\Sigma_0$), the total mass (M$_{tot}$), and the fraction of this total in HI gas, when I could find them. Also listed is an estimate of the mean metallicity of each galaxy, given as [Fe/H], for the sake of uniformity.  Where the metallicity was ``converted'' from [O/H], this is noted. The rest of the values are predominantly based upon the colour of the red giant branch, and thus ought to be treated as lower limits. A lot of detail is glossed over in presenting a uniform ``metallicity'' for a range of galaxy type, as here ({\\it cf}, Skillman 1998 for much more detail on this complex topic).  Most of the data in Table~1 came from Mateo (1998), which contains a very complete collection of what is known about Local Group dwarf galaxies.  Mateo also provides explanations of the error bars, and where these data originally come from.  The purpose of Table~1 is largely illustrative and I would recommend anyone who would like to use the values in this table to look them up in Mateo, and even better still in the original references provided there.  I also recommend the detailed descriptions of the individual galaxies in van den Bergh (2000).  The data for the more massive galaxies in the Local Group: the Milky Way, M31, M33 and the LMC/SMC are extracted from various standard sources, and not meant to be definitive, merely to illustrate the scale sizes between dwarf galaxies and their large neighbours in the Local Group.  The mass of the Local Group is dominated by three large spiral galaxies, namely our Galaxy, M~31 and M~33. However, the largest population {\\it by number} are the dwarf type galaxies (see Table~1). The Local Group contains several examples of each class of dwarf galaxy, as defined above. All dwarf galaxy types could plausibly come from the same type of progenitor but for reasons of differences in initial dark matter content, or environment or chance encounters with other galaxies follow different evolutionary paths which result in the different present-day properties ({\\it e.g.}, Ferrara \\& Tolstoy 2000; Binggeli 1994; Davies \\& Phillips 1988)  \\begin{figure} \\vskip 1cm \\psfig{figure=horse_fig.ps,height=15cm,width=12cm} \\caption{ This is a WIYN 3.5m telescope image of the Pegasus dwarf taken in 0$^{\\prime\\prime}_.$6 seeing, from Gallagher {\\it et al.} (1998). This is a good example of a low surface brightness dI in the Local Group. It shows the bright core where star formation is occurring, and the more extended dE-like main body of the galaxy.  The field of view shown here is 4$^{\\prime}_.8\\times6^{\\prime}_.$7 on a side, which corresponds to 1$\\times$1.5~kpc at the distance of Pegasus (760~kpc).  The location of the WFPC2 pointing is also shown. } \\end{figure}  The total mass of a galaxy has a critical impact on the ability of that galaxy to form stars. It determines how effectively gas can be compressed to increase the efficiency of star formation and also how hard it is for supernova explosions to disrupt or even rid the system of gas delaying or preventing future star formation (e.g, Mac Low \\& Ferrara 1999; De Young \\& Heckman 1994; Dekel \\& Silk 1986).  The lowest mass dwarf galaxies are often at the hairy edge of being able to retain their gas whilst forming stars, and many clearly have lost the battle. For the lowest mass galaxies any small perturbation can have a dramatic impact on the evolution of these systems, and it is this sensitivity to initial conditions and random events which may explain some of the dispersion of the values in Table~1 (see Ferrara \\& Tolstoy 2000).  One of the few obvious correlations between dwarf galaxy properties is that between absolute magnitude (M$_V$) and global metallicity ({\\it e.g.}, Skillman, Kennicutt \\& Hodge 1989).  If a reasonably uniform mass to light (M/L) ratio can be assumed, then this can be interpreted as a mass-metallicity relation. Those galaxies which fall significantly off the relation ({\\it e.g.} extreme BCDs) can be assumed, with very good reasons, to have a significantly different M/L. Indeed some scatter in the M/L of average galaxies might well account for the scatter seen in the mass-metallicity relation ({\\it e.g.}, Ferrara \\& Tolstoy 2000).  A large uncertainty in proving or disproving this is the lack of reliable measurements of total (or dynamical) masses of dwarf galaxies.  Basically, the global star-formation properties of a galaxy appear to be dominated by the total mass.  This may perhaps vary when galaxies interact and merge, but it is difficult to disentangle the effects of merging two small galaxies to form a bigger one and the temporary boost to the star formation rate from the merger.  These two effects might balance each other - because taken at face value the luminosity/mass-metallicity relation seems to say that a galaxy knows how big it is from the earliest times. If a lot of dwarfs were added together they would retain the global metallicity of the original pieces, whilst increasing in luminosity. This simple arithmetic could also argue against merging having a significant impact upon Local Group galaxies, at least in recent times. ",
        "conclusions": "A survey of the resolved stellar populations of all the galaxies in our Local Group provides a uniform picture of the global star formation properties of galaxies with a wide variety of mass, metallicity, gas content etc., and makes a sample that ought to reflect the star formation history of the Universe and give results which can be compared to high redshift survey results ({\\it e.g.}, Steidel {\\it et al.} 1999). Initial comparisons suggest these different approaches do not yield the same results (Tolstoy 1998b; Fukugita {\\it et al.} 1998), but the errors are large due to the lack of {\\it detailed} star formation histories of nearby galaxies.  The CMDs presented here whilst beautiful and dramatic represent the tip of the iceberg for the Local Group. Most of the observations reported here consisted of 1 or 2 orbits of integration time per filter.  To really complete a detailed census of the nearby resolved stellar populations we need to go as deep as the sensitivity limit given in Figure~2 for all Local Group galaxies, to detect the oldest MSTOs. This includes the large Local Group galaxies, such as M~31, which along with our Galaxy represent the dominant mode of star formation in the Local Group. With data like this we will know the star formation history of the Local Group, going back to the earliest times.  We have also shown that data from ground based telescopes in excellent seeing with active optics can compete with HST images, and indeed make an excellent complement, by being more blue sensitive, and having a larger field of view. It is only with the deepest exposures of the most crowded regions that HST is still the undisputed winner, because it is hard to gain enough exposure time in excellent stable seeing conditions on the ground, as no where under the Earth's atmosphere can ideal conditions be guaranteed. It looks promising that Adaptive Optics on large ground based telescopes will one day rival HST supremacy, but this is an endeavour that will be restricted to infra-red wavelengths for the foreseeable future. HST must lead the way to extend detailed star formation studies past the Galaxy and its immediate satellites."
    },
    "0010/astro-ph0010052_arXiv.txt": {
        "abstract": "We discuss optical and infrared photometric properties of stars matched in the Two Micron All Sky Survey (2MASS) and the Sloan Digital Sky Survey (SDSS) commissioning data for $\\about$50 deg$^2$ of sky along the Celestial Equator centered at $l=150^\\circ, b=-60^\\circ$. About 98\\% ($\\about$63,000) of objects listed in the 2MASS Point Source Catalog in the analyzed area are matched within 2 arcsec to an SDSS source. The matched sources represent 8\\% of the $\\about$800,000 SDSS sources in this area. They are predominantly red sources, as expected, and 15\\% of them are resolved in SDSS imaging data, although they are detected as point sources in 2MASS data. The distribution of positional discrepancies for the matched sources, and the astrometric statistics for the multiply observed SDSS sources, imply that the astrometric accuracy of both surveys is about 0.1 arcsec per coordinate (rms).  For about 14,000 stars with the smallest photometric errors ($\\la$ 10\\%) in both surveys, we present optical and infrared color-magnitude and color-color diagrams. We use optical (SDSS) colors to identify the stellar spectral sequence and show that stars of different spectral types can have similar infrared colors, thus making the classification of stars based on only 2MASS data very difficult. However, a broad separation into ``early'' and ``late'' spectral types (relative to type K0) is possible with a reliability of $\\about$95\\% even with 2MASS colors alone.  The distributions of matched sources in color-magnitude and color-color diagrams are compared to the predictions of a stellar population synthesis code. We find that the models are in fair overall agreement with the data. In particular, the total number counts agree to better than 10\\%, and the morphologies of the color-magnitude and color-color diagrams appear similar. The most significant discrepancies are found for the number ratio of ``early'' to ``late'' type stars (by about a factor of 2) and in the colors of M stars (up to 0.2 mag). The first disagreement indicates that some parameters of the standard Galactic structure model and/or initial mass function can be improved, and the second disagreement emphasizes known difficulties with the modeling of stellar atmospheres for cool stars. ",
        "introduction": "The Two Micron All Sky Survey (2MASS, Skrutskie {\\em et al.} 1997) and the Sloan Digital Sky Survey (SDSS, York {\\em et al.} 2000) are producing eight-color optical and infrared photometry for millions of Galactic and extragalactic sources. It is of obvious scientific interest to positionally match the sources observed in both surveys and to study their distribution in color-magnitude and color-color diagrams. Due to the large number of sources and very accurate photometry and astrometry, such a study can, for example, reliably reveal detailed properties of ``normal\" stars which should constitute the majority of matched sources. This information is invaluable for studying stellar and Galactic structure and evolution, and also makes feasible efficient searches for peculiar, or previously unknown objects. For example, the selection of recently defined L dwarfs (Kirkpatrick {\\em et al.} 1999) was based on both optical and infrared photometry, an approach still followed in the current searches for L and T dwarfs (e.g. Leggett {\\em et al.} 2000).  In this paper we present preliminary results for matched sources from the recent 2MASS Second Incremental Data Release\\footnote{http://www.ipac.caltech.edu/2mass/releases/second/index.html} and SDSS commissioning data. A brief description of both surveys is provided in \\S 2. We describe the matching and the basic statistics in \\S 3. A more detailed study of optical-infrared properties of stars is presented in \\S 4, and in \\S 5 we compare data with models.  \\section {  The Data }  \\subsection {2MASS}  The 2MASS is using two 1.3-meter telescopes, one at Mt. Hopkins, AZ, and one at CTIO, Chile, to survey the entire sky in near-infrared light\\footnote{http://www.ipac.caltech.edu/2mass/overview/about2mass.html}. The Northern telescope has been observing regularly since June 1997, while the Southern facility began operation in March 1998. Each telescope's camera is equipped with three $256\\times256$ arrays (the pixel size is 2 arcsec) of HgCdTe detectors and observes simultaneously in the $J$ (1.25 $\\mic$), $H$ (1.65 $\\mic$), and $K_s$ (2.17 $\\mic$) bands.  The detectors are sensitive to point sources brighter than about 1 mJy at the $10\\sigma$ level, corresponding to limiting (Vega-based) magnitudes of 15.8, 15.1, and 14.3, respectively.  Point-source photometry is repeatable to better than 10\\% precision at this level, and the astrometric uncertainty for these sources is less than 0.2 arcsec.  When completed, 2MASS will catalog $\\sim$300 million stars as well as several million galaxies, and will cover at least 95\\% of the sky.  A new data installment, the 2MASS Second Incremental Data Release, was recently made available to the public. Spanning 47\\% of the sky, it contains photometry and astrometry for over 162 million point sources as well as 1.9 million resolved sources.   \\subsection {SDSS}  The SDSS is a digital photometric and spectroscopic survey which will cover one quarter of the Celestial Sphere in the North Galactic cap and produce a smaller area ($\\sim$ 225 deg$^2$) but much deeper survey in the Southern Galactic hemisphere (York {\\em et al.} 2000\\footnote{see also http://www.astro.princeton.edu/PBOOK/welcome.htm}). The flux densities of detected objects are measured almost simultaneously in five bands ($u'$, $g'$, $r'$, $i'$, and $z'$; \\cite{F96}) with effective wavelengths of 3540~\\AA, 4760~\\AA, 6280~\\AA, 7690~\\AA, and 9250~\\AA, 95\\% complete\\footnote{These values are determined by comparing multiple scans of the same area obtained during the commissioning year. Typical seeing in these observations was 1.5$\\pm$0.1 arcsec.} for point sources to limiting magnitudes of 22.1, 22.4, 22.1, 21.2, and 20.3 in the North Galactic cap\\footnote{We refer to the measured magnitudes in this paper as $u^*, g^*, r^*, i^*,$ and $z^*$ because the absolute calibration of the SDSS photometric system is still uncertain at the $\\sim 0.03^m$ level. The SDSS filters themselves are referred to as $u', g', r', i',$ and $z'$. All magnitudes are given on the AB$_\\nu$ system (Oke \\& Gunn 1983, for additional discussion regarding the SDSS photometric system see \\cite{F96} and Fan 1999).}. The survey sky coverage of about $\\pi$ steradians (10,000 deg$^2$) will result in photometric measurements to the above detection limits for about 50 million stars. Astrometric positions are accurate to about 0.1 arcsec per coordinate for sources brighter than $r^* \\sim$ 20.5$^m$, and the morphological information from the images allows robust  star-galaxy separation to $r^* \\sim$ 21.5$^m$.  \\section         {          Matched Data                      }  \\subsection {              SDSS Data     }  We utilize a portion of SDSS imaging data from two commissioning runs (numbered 94 and 125) taken during the Fall of 1998. The data in each run were obtained in six parallel scanlines\\footnote{See also http://www.astro.princeton.edu/PBOOK/strategy/strategy.htm}, each 13.5 arcmin wide, along the Celestial Equator ($-1.2687^\\circ <$ $\\delta_{J2000}$ $< 1.2676^\\circ$). The six scanlines from each run are interleaved to make a filled stripe. The portion of data used in this work extends from $\\alpha_{J2000}$ = 0$^h$ 24$^m$ ($l=108^\\circ$, $b=-62^\\circ$) to $\\alpha_{J2000}$ = 3$^h$ 0$^m$ ($l$=177$^\\circ$, $b=-49^\\circ$). The seeing in both runs was variable between 1 and 2 arcsec (FWHM) with the median value typically 1.5 arcsec. The total area which overlaps with public 2MASS data is 47.41 deg$^2$ and includes about 800,000 SDSS sources to the $6\\sigma$ detection limit in at least one of the SDSS bands.  The SDSS color-color and color-magnitude diagrams which summarize photometric properties of unresolved sources are shown in Figure \\refSDSSall. In this and all other figures, we correct the data for the interstellar extinction\\footnote{The Galactic structure model discussed in \\S5 implies that the most sources in this sample are further than 200 pc away, which justifies this correction.} determined from the maps given by Schlegel, Finkbeiner \\& Davis (1998). Typical values for high-latitude regions discussed in this work are $A_{r^*}$ = 0.05-0.15 mag ($A_{r^*} = 0.84 A_V$). Throughout this work we use the ``point spread function'' (PSF) magnitudes\\footnote{Note that SDSS photometric system uses asinh magnitudes (\\cite{LGS99}). For reference, zero flux corresponds to asinh magnitudes of 23.40, 24.22, 23.98, 23.51, and 21.83 in $u^*, g^*, r^*, i^*$, and $z^*$, respectively.} (measured by fitting a PSF model of a double Gaussian and applying an aperture correction) as computed by the photometric pipeline (``photo'', version v5\\_0; for details see \\cite{Lupton00}). Photometric errors are typically 0.03$^m$ at the bright end ($r^* < 20^m$), relevant in this work (for more details see Ivezi\\'c {\\em et al.} 2000).  The top left panel of Figure 1 displays the \\r\\ vs. \\gr\\ color-magnitude diagram for $\\about$ 25,000 objects observed in 3 deg$^2$ of sky during SDSS commissioning run 94. There are at least three distinct groups of sources. The red sources with \\gr\\ $\\about$ 1.4 are disk stars, and are dominated by stars of spectral type M, see e.g. Fukugita {\\em et al.} (1996) and Lenz {\\em et al.} (1998). The blue ridge with \\gr\\ $\\about$ 0.4 is dominated by F and G disk stars at the bright end (\\r $\\la$ 18), and by halo stars at the faint end (note that the median \\gr\\ color for halo stars is bluer than for disk stars, due to the lower metallicity in the halo).  As shown by Fan (1999), the contribution of low-redshift ($z \\la 2$) quasars (QSOs) becomes increasingly important at the faint end of the blue ridge. The majority of stars observed by SDSS are on main sequence; with the aid of models discussed in \\S5 we estimate that the fraction of stars which are not on main sequence is $\\sim$1.0\\%. The latter are dominated by giants and subgiants with 0.4 $<$ \\gr $<$ 0.8 (\\about 90\\%) and horizontal branch stars with \\gr $<$ 0.5 (\\about 10\\%).  The three remaining panels show three projections of the SDSS color-color diagrams for objects, marked as dots, brighter than 20$^m$ in each of the 3 bands used to construct each diagram. Red is always towards the upper right corner. Objects that have \\ug\\ and \\gr\\ colors similar to low-redshift QSOs (\\ug $<$ 0.4, -0.1 $<$ \\gr $<$ 0.3, \\ri $<$ 0.5), and are brighter than the limit for quasars in the SDSS spectroscopic survey (\\i $<$ 19), are marked by open circles.  The locus of ``normal\" stars stands out in all three diagrams. As shown by Fukugita {\\em et al.} (1996), Krisciunas, Margon \\& Szkody (1998), Lenz {\\em et al.} (1998), and Fan (1999), the position of a star in these diagrams is mainly determined by its spectral type. For reference, the positions of several spectral types taken from Fukugita {\\em et al.} (1996) are marked next to the stellar locus in the \\ug\\ vs. \\gr\\ and \\ri\\ vs. \\gr\\ color-color diagrams (labels are slightly offset for clarity).  For most of its length, the locus in the \\gr\\ vs. \\ug\\ diagram is made of stars with spectral types ranging from F to late K. A-type stars are visible at \\ug $\\about$ 1.0--1.2 and \\gr $\\about$ -0.3--0.0 (for a detailed study of A stars detected in SDSS commissioning data see Yanny {\\em et al.} 2000), and late K and M stars are found at the red end of the stellar locus. Different M spectral subtypes cannot be distinguished in the \\gr\\ vs. \\ug\\ diagram, and these stars are better separated in the \\ri\\ vs. \\gr\\ diagram. M stars have a constant \\gr\\ color (\\about 1.4) due to strong TiO bands in their spectra, and a value of \\ri\\ color that depends strongly on spectral subtype. In the \\iz\\ vs. \\ri\\ diagram, stars earlier than K type are found in a small region around \\ri $\\about$ 0.2 and \\iz $\\about$ 0.0. For most of its length, the \\iz\\ vs. \\ri\\ stellar locus represents late K and M stars.  \\subsection      {         2MASS       Data          }  We positionally match data from the two SDSS commissioning runs to 2MASS Point Source Catalog (PSC) data with 0$^h$ 24$^m$ $<$ $\\alpha_{J2000}$ $<$ 3$^h$ 0$^m$ and $-1.2687^\\circ <$ $\\delta_{J2000}$ $< 0^\\circ$ (2MASS data for 0$^\\circ$ $<$ $\\delta_{J2000}$ $< 1.2676^\\circ$ are not yet available). The region with 1$^h$ 51$^m$ $<$ $\\alpha_{J2000}$ $<$ 1$^h$ 57$^m$ is missing from the 2MASS catalog. The resulting overlapping area (47.41 deg$^2$) includes 64,695 2MASS sources.   \\subsection   { The Positional Matching of SDSS and 2MASS Catalogs }  We consider as matches all sources from both catalogs whose listed positions agree to better than 2 arcsec. There are 63,341 such sources, or 97.9\\% of the 2MASS PSC sources and 8\\% of the SDSS sources, in the overlapping area. The probability of finding an SDSS source within a circle with 2 arcsec radius is \\about 1\\%, and the probability of associating a 2MASS source with multiple SDSS sources is thus negligible ($<$ 10$^{-4}$). With the further constraints on the final sample, discussed in \\S4, the random association rate for matched sources also becomes inconsequential ($<$ 10$^{-5}$).  The distributions of RA and Dec differences between the SDSS and 2MASS catalog positions are shown in Figure \\refRADec. The thin lines show the normalized histograms for all matched sources, and the thick lines show histograms for the $\\about 14,000$ sources with at least 10$\\sigma$ detections in both surveys (see the next section), which are not saturated or blended with other sources in the SDSS images. Both histograms are normalized by the total number of sources. Equivalent Gaussian widths determined from the interquartile range are 0.30 arcsec for the distributions of all sources and 0.15 arcsec for the sources with the high signal-to-noise ratio\\footnote{The contribution of the proper motions to this scatter is negligible due to the short time span between the SDSS and 2MASS observations.}  (SNR). As the multiple SDSS commissioning observations of the same area show that the positions of sources with high SNR are reproducible to 0.10 arcsec per coordinate (Pier {\\em et al.} 2000), this implies that the astrometric accuracy of both 2MASS and SDSS is about 0.10 arcsec for bright sources (positional root-mean-square scatter per coordinate). However, note that the histograms displayed in Figure \\refRADec\\ show a 0.05 arcsec offset from the origin for both coordinates. A comparison of the SDSS and VLA FIRST survey (Faint Images of the Radio Sky at Twenty-cm, Becker {\\em et al.} 1995) data shows the same offset (though with less statistical significance due to the smaller number of matched sources), indicating that this is a systematic error in the SDSS astrometry. We are currently investigating this problem.    \\subsection     {      Unmatched  2MASS PSC  Sources        }  There are 1354 2MASS sources (2.1\\% of the whole sample) which do not have a cataloged SDSS source within 2 arcsec. Visual inspection of SDSS images shows that 70\\% of these do not have an optical counterpart at the cataloged 2MASS position. Such sources are either spurious 2MASS detections (roughly 80\\% of the unmatched 2MASS sources have SNR $<$ 10 in all 2MASS bands), extremely red sources (e.g. stars heavily obscured by circumstellar dust), or asteroids which moved between the times of the SDSS and 2MASS observations. There is no significant correlation between the number of unmatched sources and the distance from the Ecliptic Plane.  15\\% of the unmatched 2MASS sources are associated in SDSS images with saturated stars. The centroids of saturated stars in SDSS have a large positional uncertainty which sometimes may exceed the matching cutoff (2 arcsec). 10\\% of the unmatched 2MASS sources are associated in SDSS images with complex structures consisting of several blended sources, and which were not deblended by the processing software.  The remaining 5\\% of the unmatched sources are blended with a satellite trail or a diffraction spike of a nearby bright star in the SDSS data. This category represents only 0.1\\% of the total number of 2MASS sources.    \\section {     Optical and Infrared Properties of Stellar Sources }   The combined SDSS and 2MASS eight-color photometry can be used to construct a multitude of point source color-magnitude and color-color diagrams. The five SDSS and three 2MASS bandpasses are shown in Figure \\reffilters. All curves include typical atmospheric extinction and the transmission of the entire instrumental system. The solid curves correspond to the overall efficiencies of the SDSS system (Gunn {\\em et al.} 1998), and the dot-dashed lines show the 2MASS responses normalized to 0.5 (M. Cohen, priv. comm.).  Since the thrust of this work is to provide the accurate optical and infrared colors of normal stars, we apply several conditions on the matched sources in order to select a subsample of stars with the best photometric data.  \\begin{enumerate}  \\item The 2MASS and SDSS positions must agree to better than 1 arcsec. This limit corresponds to a $\\about$5$\\sigma$ cut.  \\item Sources must not be saturated in any band in the SDSS data. This is roughly equivalent to the condition $i^* > 14$, because the majority of matched sources are red.  \\item Sources must have high-SNR 2MASS detections. We impose this limit by requiring $J < 15.8$, $H < 15.1$ and $K_s < 14.3$, which corresponds to $10\\sigma$ detections. Since the majority of matched sources have $J-K_s$ colors bluer than 1.5 (see below), the faint limit is in practice determined by the condition on $K_s$ flux.  \\item Sources must be unresolved in SDSS data. The SDSS star-galaxy separation is robust to $r^* \\about 21.5$, which is significantly fainter than the faint limit of the resulting sample, and only a negligible number of galaxies are likely to be present in the final sample.  \\end{enumerate}  These limits result in a sample of 13,924 sources. Note that the fourth condition does not necessarily exclude all extragalactic sources since QSOs are typically optically unresolved sources. However, low-redshift QSOs can be easily recognized in the SDSS \\gr\\ vs. \\ug\\ color-color diagram (c.f. Figure 1) as shown by Fan (1999). There are 100 matched sources in this region of the color-color diagram, but only seven of these are bright enough to satisfy the (conservative) 2MASS SNR limits (\\# 3 above). These sources, together with the optically classified galaxies, will be discussed in a separate publication.  Figure \\refcmd\\ displays color-magnitude diagrams for the 13,917 selected stellar sources. The SDSS and 2MASS photometry is better than 10\\% for the majority of this sample due to the imposed cuts. The upper left panel shows an \\r\\ vs. \\gr\\ diagram and can be compared to the analogous diagram for all SDSS sources displayed in the upper left panel in Figure \\refSDSSall. The bright limit on the \\r\\ magnitude is correlated with the \\gr\\ color because the red sources become saturated in SDSS data at a fainter \\r\\ magnitude than the blue sources. Similarly, the faint limit depends even more strongly on the \\gr\\ color because only comparatively bright blue objects can pass the 2MASS faint limit cutoffs due to their steeply falling spectral energy distributions\\footnote{Due to these cutoffs, essentially all matched stars belong to the Galactic disk. With the aid of models discussed in \\S5, we estimate that $\\sim$97\\% of these stars are on main sequence.}. Essentially the same effects can be seen in the other three color-magnitude diagrams which combine optical and infrared data. Note the sharp faint limit in the $J$ vs. $J-K_s$ diagram because it is determined by the $K_s < 14.3$ condition, as discussed above.   The color-color diagrams for the selected subsample are shown in Figure \\refccd. The upper left panel shows an analogous diagram to the one displayed in the lower left panel of Figure \\refSDSSall. Note that there is no significant difference in the appearance of the stellar locus. However, as is already visible in Figure \\refcmd, the matched sample has a larger fraction of red stars (see below). The following three panels show optical-infrared color-color diagrams. Note that the stellar locus is clearly visible in all of them, confirming that the optical and near-infrared broad-band colors of normal stars can be described to within 0.1-0.2 magnitudes in terms of temperature as a single parameter. In particular, the near-infrared fluxes can be predicted for the majority of sources to 0.1-0.2 mag from the optical fluxes alone. However, the converse is not true; stars with varying optical colors may occupy the same region in near-infrared color-color diagrams (n.b. in the Rayleigh-Jeans tail color is independent of temperature). This is true for some stars even in the case of optical-infrared color-color diagrams. For example, all M stars occupy a very small region in the $J-K_s$ vs. \\gr\\ diagram shown in the middle left panel (\\gr = 1.4, $J-K_s$=0.85). While their \\gr\\ color can be determined from their $J-K_s$ color, it is impossible, for example, to constrain their \\ri\\ color, which can vary by more than a magnitude.  The lower right panel displays the $J-K_s$ vs. $r^*-K_s$ color-color diagram. This diagram was used by Kirkpatrick {\\em et al.} (1999) in their search for L dwarfs; we present it here as reference (note that Kirkpatrick {\\em et al.} display this graph with reversed $J-K_s$ axes). Kirkpatrick {\\em et al.} show that the M spectral sequence begins around $r^*-K_s$ = 2.0, and ends around $r^*-K_s = 8.0$, where the L dwarf sequence begins. L dwarfs are very rare and are not apparent on the diagram. The position of Gl 229B, the first discovered T dwarf (Nakajima {\\em et al.} 1995), is marked by an open circle at $r^*-K_s = 9.1$ and $J-K_s = -0.1$.  The lower left panel shows a color-color diagram constructed from 2MASS data alone. The stellar locus is not as well defined as in the optical and optical-infrared color-color diagrams, and the position of a star in this diagram is not strongly correlated with its spectral type. However, the structure seen in the $H-K_s$ vs. $J-H$ color-color diagram indicates that there is at least a weak correlation between the spectral type and the position in this diagram. Understanding such a correlation could allow for very efficient studies of the Galactic disk structure based on only 2MASS data; unlike SDSS, 2MASS will extend over the entire sky. We attempt to quantify such a correlation by separating sources based on their optical colors, and then studying their distribution in the $H-K_s$ vs. $J-H$ color-color diagram.  The \\ri\\ vs. \\gr\\ color-color diagram can be used to separate the stars into rough spectral classes (c.f. \\S 3.1). We divide the sample into  \\begin{enumerate} \\item Stars with spectral types earlier than G5 by the condition $\\gr \\le 0.5$. There are 2894 stars (21\\%) in this subsample. \\item Stars with spectral types between G5 and K5: $0.5 < \\gr \\le 1.0$. There are 3731 stars (27\\%) in this subsample. \\item Stars with spectral types between K5 and M5: $\\gr > 1.0$ and $\\ri \\le 1.0$. There are 4393 stars (32\\%) in this subsample. \\item Stars with spectral types later than M5: $\\gr > 1.0$ and $\\ri > 1.0$. There are 2873 stars (20\\%) in this subsample.  \\end{enumerate}   The distribution of stars in the $H-K_s$ vs. $J-H$ and $J-K_s$ vs. $r^*-K_s$ color-color diagrams is shown separately for each subsample in the four panels of Figure \\refcontours. The spectral types are plotted in the same order as listed above, with stars earlier than type G5 shown in the top panel. Stars from each optically selected spectral subsample are plotted as dots, and the contours represent the distribution of the whole sample. The contour levels are 0.5, 0.3, 0.1 and 0.03 of the peak density. The separation of the four classes is much better defined in the $J-K_s$ vs. $r^*-K_s$ diagram, because the range of colors is much larger than in the $H-K_s$ vs. $J-H$ diagram. The two peaks in the $H-K_s$ vs. $J-H$ source distribution correspond to stars earlier than G5 (blue peak), and to stars with spectral types later than K5 (red peak). Stars with spectral types between G5 and K5 populate the area between the peaks, and the stars with spectral types later than M5 populate the region with the reddest $J-K_s$ and $r^*-K_s$ colors.  We have also attempted a finer binning in the \\ri\\ vs. \\gr\\ color-color diagram and found that the 2MASS data alone does not allow for a more precise spectral classification than that presented in Figure \\refcontours. Nevertheless, we find that the 2MASS color $J-K_s = 0.6$ provides a good overall separation of stars with $\\gr < 0.7$ from stars with $\\gr \\ge 0.7$ (c.f. Figure 5). There are 5099 stars (37\\%) with $\\gr < 0.7$ in the subsample of 13917 stars with good photometry, and 93\\% of these also have $J-K_s < 0.6$. Of the remaining 8818 stars with $\\gr \\ge 0.7$, there are 96\\% with $J-K_s \\ge 0.6$. These numbers show that there is a $\\about95\\%$ probability of success in predicting whether the SDSS \\gr\\ color is greater or smaller than 0.7 from the information specifying whether the 2MASS $J-K_s$ color is greater or smaller than 0.6. The SDSS \\gr\\ color of 0.7 separates stars with spectral type earlier than K0 from stars with later spectral types. This boundary is fuzzy to within 2-3 spectral subtypes due to the effects of varying metallicity and gravity on stellar colors. Such a method of separating early from late type stars may prove useful in selecting subsamples of 2MASS data for the studies of Galactic disk structure.   \\section { Comparison of Data with a Stellar Population Synthesis Code  }  We compare the number counts and the source distribution in color-color diagrams using the code of Fan (1999, hereafter F99) for simulating SDSS observations of unresolved sources, which we have extended to the 2MASS bands (Finlator 2000). F99 compared several simulated stellar populations to data obtained by the ESO Imaging Survey Patch B (Prandoni {\\em et al.} 1998), and found that the position of the simulated stellar locus in SDSS color-color space agreed very well with the data. Nevertheless, he also found that the number counts could differ by up to $\\sim$20\\%. In the following section we present an expanded analysis of the model predictions, made possible by the new 2MASS and SDSS data, and discuss similar discrepancies. Detailed discussion of this model is given by F99, and here we only provide a brief summary.   \\subsection{       Generating a Simulated Population      }  The population synthesis code for generating a simulated stellar population at a given Galactic latitude and longitude is based on the following inputs/steps:  \\begin{enumerate}  \\item A Galactic structure model based on the star count models of Bahcall \\& Soneira (1981, 1984) is used to generate the number of halo and disk stars as a function of Galactic latitude and longitude.  The disk scale height is assumed to increase with stellar age (i.e. the scale height is correlated with the spectral type), and the Galactic halo is flattened. The initial mass function is taken from Haywood {\\em et al.} (1997) and is assumed to be valid both for disk and halo stars.  \\item To generate the properties of a specified stellar population, the simulation generates the mass, age, and metallicity of each star with the aid of the Padova stellar evolutionary tracks and isochrones (Bertelli et al. 1994). The resulting $T_{eff}$, [Fe/H], and log($g$) are then used to select a stellar atmosphere model in order to determine the colors for each star.  \\item The hybrid model stellar atmospheres compiled by Lejeune, Buser, \\& Cuisinier (1997a, 1997b) are used to compute the SDSS and 2MASS colors of stars, given their effective temperature $T_{eff}$, metallicity [Fe/H], and surface gravity log($g$).  \\item The SDSS AB magnitudes are computed as described in F99, and the 2MASS Vega magnitudes, $m$, are computed using \\begin{equation} m = -2.5 \\log \\frac{\\int \\mathrm{d}(\\log(\\nu))f_{\\nu}S_{\\nu}} {\\int \\mathrm{d}(\\log(\\nu))S_{\\nu}} - 48.60 + m_0, \\end{equation} where $m_0(J)=-0.89$, $m_0(H)=-1.37$, and $m_0(K_s)=-1.84$ represent the offsets from the AB magnitude system. These offsets were computed  by assuming that the flux from Vega is constant within any 2MASS band and equal to 1595, 1024, and 667 Jy in the $J$, $H$, and $K_s$ bands, respectively (M. Cohen, priv. comm.).  \\item Photometric errors in SDSS bands are added as described in Appendix A of F99. The 2MASS photometric errors are derived from an online plot of photometric repeatability versus magnitude\\footnote{http://www.ipac.caltech.edu/2mass/releases/second/doc/figures/secii2f7.gif}. Errors are $\\about0.03^m$ at the bright end, and then rise to $\\about0.1^m$ at $J=15.8$, $H=15.1$, $K_s=14.3$.  \\end{enumerate}    \\subsection {Comparison of Simulated Observations with the Data   }  Figure \\refdatamodels\\ compares the photometric properties of the matched sources with a population simulated at $(l,b)=(150^\\circ,-60^\\circ)$ (the properties of a sample are strong functions of the Galactic coordinates) and spanning the same area on the sky (47.41 deg$^2$). The panels on the left show the diagrams for the matched objects, and those on the right show the simulated population. To ensure the same selection effects for both samples, we require that the model fluxes satisfy $z^* > 14$, $J < 15.8$, $H < 15.1$, and $K_s < 14.3$. These conditions result in 12,740 simulated sources, only 9\\% fewer than the observed value of 13,917 matched sources. Since the selection effects are quite complicated, and include two independent surveys, this close agreement is quite encouraging. We note that the simulated population does not include ``exotic'' objects such as white dwarfs and QSOs.  The top row in Figure \\refdatamodels\\ compares optical colors. The overall agreement is satisfactory, although there are fewer simulated stars with \\ri $\\about$ 0.8-0.9, yielding a discontinuity in the stellar locus in the \\gr\\ vs. \\ri\\ diagram. This is due to the different model libraries used for objects with T$_{eff} >$ 3500 K and T$_{eff} <$ 3500 K, as discussed by F99. The middle row in Figure \\refdatamodels\\ shows an optical-infrared color-color diagram. The overall morphology is well reproduced by the simulated population, except for the M stars ($r^*-K_s \\ga 3$) which have notably different colors. While the observations suggest that infrared colors should not vary with $r^*-K_s$ color for $r^*-K_s \\ga 3$, the simulated M stars become bluer in $J-K_s$ after this point. We will further discuss this discrepancy below. The bottom row in Figure \\refdatamodels\\ compares the $z^*$ vs. $r^*-K_s$ color-magnitude diagrams. Again, the overall morphology is reproduced, but the simulation does not seem to generate enough bright stars with intermediate $r^*-K_s$ colors.  To provide a more quantitative comparison of the data with the models, we display histograms of the matched and simulated objects in Figure \\refhistograms. The top four plots show that the simulation overestimates the overall counts of blue stars while underestimating the red star counts. While the observed ratio of stars redder and bluer than $J - K_s = 0.6$ is 1.73, the simulated ratio is 0.98. The top right panel clearly shows the lack of simulated stars with \\ri $\\about$ 0.8-0.9 mentioned earlier.  The disagreement between the observed and simulated ratio of red to blue stars could be due either to incorrect parameters of the Galactic structure model, and to an incorrect shape of the initial mass function employed in the simulation. The blue stars have higher luminosities than do the red stars and thus are observed to larger distances. The majority of simulated red stars with $J - K_s > 0.6$ have distances ranging from 200 to 800 pc (median value of 500 pc), while the majority of blue stars are in the distance range 800-2000 pc (median value of 1000 pc). Decreasing the disk scale height for red stars, and increasing it for blue stars in the expression for stellar density as a function of distance from the Galactic plane (F99, eqs. 5 and 6) can increase the simulated ratio of red to blue stars. However, these adjustments would involve changing the age of stars,  resulting in a change in the total number of simulated stars.  Similarly, one can increase the simulated ratio of red to blue stars by increasing the slope of the initial mass function (IMF) at the low-mass end. The simulated sample is dominated by stars with masses $\\la 1 \\Mo$. The majority of stars with $J - K_s > 0.6$ have masses between 0.3 \\Mo\\ and 0.75 \\Mo, while the stars with $J - K_s < 0.6$ have masses between 0.75 \\Mo\\ and 1.1 \\Mo. In this mass range, the simulation adopts $dN/dM \\propto M^{-1.7}$. We have examined a set of models with $dN/dM \\propto M^{-n}$ and found that $n=2.5\\pm0.1$ produces the observed ratio of red to blue stars. However, the total number of stars becomes too low and another model adjustment is needed. Thus, to meaningfully constrain the free parameters requires an extensive modeling effort beyond the scope of this work. A detailed analysis of SDSS stars counts and their implications for the Galactic structure model will be reported in Chen {\\em et al.} (2000).  Another model difficulty is related to the properties of the stellar models, as opposed to the Galactic distribution of stars. As discussed above, the $J-K_s$ colors of simulated M stars becomes bluer for $r^*-K_s \\ga 3$, while the observations suggest that they should stay constant at roughly 0.85. The bottom two panels of Figure 8 show the $J-K_s$ histograms for two subsamples of M stars: the left panel shows stars with $3.5 < \\r -K_s < 4$ ($\\about$ M3) and the right panel shows stars with $4.5 < \\r -K_s < 5$ ($\\about$ M5). It is evident that the $J-K_s$ color for simulated early M stars is significantly redder (0.15$^m$) than observed, while it is bluer than observed for the mid-to-late M stars. Such discrepancies between the data and simulations for M stars are not surprising because these stars are notoriously difficult to model due to their complicated chemistry (Allard {\\em et al.} 1997). Particularly problematic are incomplete and uncertain opacity tables, as well as the possibility of dust condensation in the outer cool parts of the atmosphere, an effect which is usually ignored. ",
        "conclusions": "This preliminary study indicates the enormous potential of combined 2MASS and SDSS data for stellar and Galactic structure studies. The results of the positional matching of sources presented here indicate that both surveys will provide outstanding astrometry ($\\about 0.1$ arcsec rms per coordinate) for an unprecedented number of objects. Eight-color accurate photometry (better than 0.1 mag) for millions of stars is bound to place studies of  stellar and Galactic structure and evolution at an entirely new level. Such a large number of stars will enable detailed studies of the stellar initial mass function and its possible Galactic variation, of the metallicity gradients in the Galaxy, and similar other types of analyses which were until now impeded by the lack of large and reliable data sets. For example, by analyzing sources from less than 1\\% of the final 2MASS-SDSS overlapping area, we find that either the parameters of the standard Galactic structure model, or the conventional initial mass function, must be changed to make simulations fit the data. Also, the wide and detailed spectral coverage based on eight fluxes provides strong tests for the stellar structure and atmosphere models. As we have shown here, the near-infrared model colors for M dwarfs are significantly different from the observed colors.  In addition to the testing of stellar and Galactic structure models as discussed here, such a large and uniform data set is ideal for discovering objects with unusual colors (e.g. T Tau stars, Herbig Ae/Be stars, brown dwarfs, etc.). An analysis of such outliers in color-color space will be presented in a forthcoming publication.    \\vskip 0.4in \\leftline"
    },
    "0010/astro-ph0010264_arXiv.txt": {
        "abstract": "The CELESTE experiment uses the heliostats of an old solar farm in the French Pyrenees to detect gamma ray air showers by the atmospheric Cerenkov technique. Observations of the TeV blazar Markarian 421 have been made with the fully instrumented CELESTE experiment since December 1999. The detection of gamma ray emission from this source at energies greater than 50 GeV is presented here. A comparison is made with the light curve from the CAT experiment at the same site which shows correlation between the observed gamma ray fluxes and the detection of short duration flaring episodes. ",
        "introduction": " ",
        "conclusions": ""
    },
    "0010/astro-ph0010114_arXiv.txt": {
        "abstract": "We have written a light curve synthesis code that makes direct use of model atmosphere specific intensities, in particular the {\\sc NextGen} model atmosphere grid for cool giants ($T_{\\rm eff}\\le 6800$~K and $\\log(g)\\le 3.5$, Hauschildt et al.\\ \\cite{hauschildt99}).  We point out that these models (computed using spherical geometry) predict a limb darkening behaviour that deviates significantly from a simple linear or two-parameter law (there is less intensity at the limb of the star). The presence of a significantly nonlinear limb darkening law has two main consequences.  First, the ellipsoidal light curve computed for a tidally distorted giant using the {\\sc NextGen} intensities is in general different from the light curve computed using the same geometry but with the black body approximation and a one- or two-parameter limb darkening law. In most cases the light curves computed with the {\\sc NextGen} intensities have deeper minima than their black body counterparts. Thus the light curve solutions for binaries with a giant component obtained with models with near linear limb darkening (either black body or plane-parallel model atmosphere intensities) are biased. Observations over a wide wavelength range (i.e.\\ both the optical and infrared) are particularly useful in discriminating between models with nearly linear limb darkening and the {\\sc NextGen} models. Second, we show that rotational broadening kernels for Roche lobe filling (or nearly filling) giants can be significantly different from analytic kernels due to a combination of the nonspherical shape of the star and the radical departure from a simple limb darkening law.  As a result, geometrical information inferred from $V_{\\rm rot} \\sin i$ measurements of cool giants in binary systems are likewise biased. ",
        "introduction": "The  study of close binary stars is of interest for several reasons. For example, the understanding the structure and evolution of stars is a basic goal of stellar astronomy, and is required in most other branches of astronomy.  Critical tests of evolution theory (i.e.\\ predicting the radius and luminosity of a star as a function of its mass and age) for stars other than the Sun are practical only for a small set of eclipsing binary stars (see, e.g., Pols et al.\\ \\cite{pols97}; Schr\\\"oder et al.\\ \\cite{schroder97}; Lacy et al.\\ \\cite{lacy00}).  In addition, our knowledge of the masses of stellar black holes (and many neutron stars and white dwarfs as well) depends on our ability to interpret ellipsoidal and usually non-eclipsing light curves (e.g.\\ Avni \\& Bahcall \\cite{avni75}; Avni \\cite{avni78}; McClintock \\& Remillard \\cite{mcclintock90}; Haswell et al.\\ \\cite{haswell93}; Shahbaz et al.\\ \\cite{shahbaz94}; Sanwal et al.\\ \\cite{sanwal96}; Orosz \\& Bailyn \\cite{orosz97}). Finally, considerable observational effort has been put forth recently to use detached eclipsing binaries as extragalactic distance indicators (Mochejska et al.\\ \\cite{mochejska98}; Ribas et al.\\ \\cite{ribas00}).  Since close binary stars are of such importance, it is crucial that we have the ability to construct accurate synthetic light curves for a variety of close binaries.  The light curve expected from a particular close binary depends on the system geometry (i.e.\\ the figures of the stars, their relative sizes, separation, viewing angle, etc.) and on the radiative properties of the stars, which are set mainly by their effective temperatures, surface gravities and chemical compositions.  The equations describing the basic system geometry for a close binary are reasonably simple and have been known for a long time (e.g.\\ see the text by Kopal \\cite{kopal59}).  In practice, however, the direct computation of light curves requires a computer, and codes to compute light curves have been in use since the late 1960s (e.g.\\ Lucy \\cite{lucy67}; Hill \\& Hutchings \\cite{hill70}; Wilson \\& Devinney \\cite{wilson71}; Mochnacki \\& Doughty \\cite{mochnacki72}; Avni \\& Bahcall \\cite{avni75}; see also the review by Wilson \\cite{wil94}). On the other hand, the detailed computation of stellar atmosphere models which describe the specific intensity of radiation emitted by the stellar surfaces is quite involved.  As a result, approximations are frequently used in the computations of light curves.  The Planck function is usually used to compute the normal monochromatic intensity of a surface element with an effective temperature $T_{\\rm eff}$: \\begin{equation} I_0\\propto[\\exp(hc/k\\lambda T_{\\rm eff})-1]^{-1} \\end{equation} where $\\lambda$ is the effective wavelength of the observation and $h$, $c$, and $k$ are the usual physical constants.  After $I_0$ is found, the specific intensities for other emergent angles are computed using a simple parameterisation (the ``limb darkening law''): \\begin{equation} I(\\mu) = I_0[1-x(1-\\mu)]. \\label{linear} \\end{equation} The coefficient $x$ depends on the temperature, gravity, and chemical composition of the star being modelled.  There are also many two-parameter limb darkening laws including a ``quadratic'' law: \\begin{equation} I(\\mu) = I_0[1-x(1-\\mu)-y(1-\\mu)^2], \\label{quad} \\end{equation} the ``square root'' law (D\\'{\\i}az-Cordov\\'es \\& Gim\\'enez \\cite{diaz92}): \\begin{equation} I(\\mu) = I_0[1-x(1-\\mu)-y(1-\\sqrt{\\mu})], \\label{sqrt} \\end{equation} or the ``logarithmic'' law (Klinglesmith \\& Sobieski \\cite{klinglesmith70}): \\begin{equation} I(\\mu) = I_0[1-x(1-\\mu)-y\\mu\\ln\\mu]. \\label{log} \\end{equation} There are many tabulations of the coefficients for these various laws for a wide variety of temperatures, gravities, and chemical compositions (e.g.\\ Al-Naimiy \\cite{alnaimiy78}; Wade \\& Rucinski \\cite{wade85}; Van Hamme \\cite{vanhamme93}; Claret \\cite{claret98}).    We discuss in this paper our technique for computing light curves using model atmosphere specific intensities directly, thereby eliminating the need for the black body approximation and the one- or two-parameter limb darkening laws.  Although our emphasis here is on stars with low effective temperatures and surface gravities ($T_{\\rm eff}\\le 6800$~K and $\\log g\\le 3.5$), the technique is quite general. In the next section we discuss some previous work in this area and give details of our method.  We then discuss some of the basic results and their implications, and give a summary of this work.  We have also included an appendix to the paper which gives some of the details of our light curve code not directly related to the stellar atmospheres. ",
        "conclusions": "In this paper we have presented a way to include specific intensities from detailed model atmosphere computations into a light curve synthesis code.  We have shown that using the model atmosphere intensities directly is almost required for cool giants since the limb darkening behaviour for a cool low-gravity atmosphere is radically different than simple one- or two-parameter limb darkening laws (Fig.\\ \\ref{fig1}). This departure from the near-linear limb darkening law is a consequence of the spherical geometry used in the computation of the {\\sc NextGen} models. Other workers have computed spherical model atmospheres for cool giants before (e.g.\\ Scholz \\& Tsuji \\cite{scholz84}; Scholz \\& Takeda \\cite{scholz87}; Plez et al.\\ \\cite{plez92}; J{\\o}rgensen et al.\\ \\cite{jorgensen92}), and the strongly nonlinear limb darkening behaviour has previously been noted by Scholz \\& Takeda (\\cite{scholz87}), and more recently by Hofmann \\& Scholz (\\cite{hofmann98}). We have shown that this strongly nonlinear limb darkening has a large effect on the light curve amplitude.  In general, the light curves computed using the {\\sc NextGen} intensities have larger amplitudes than the light curves computed using the same geometry but with  black body intensities and a one- or two-parameter limb darkening law.  Also, the strongly nonlinear limb darkening has an effect on rotational broadening kernels in that for spherical stars the kernel computed using the {\\sc NextGen} intensities will be narrower than a numerical or an analytic kernel that has a linear limb darkening law. If the star fills a substantial fraction of its Roche lobe, then the broadening kernel will also be different than the analytic one owing to the nonspherical shape of the star.   Thus we basically have two different types of models with which to fit close binary light curves:  our ELC code with the {\\sc NextGen} intensities, and codes that use either black body intensities or plane-parallel model atmosphere intensities and near-linear limb darkening (e.g.\\ W-D, ELC in black body mode, ELC with the Kurucz table, etc.). It should be possible to test which class of models provides the better description of the observational data available for binaries with cool giants.  In this regard, we have shown that observations in several bandpasses are useful in discriminating between the models. For example, Figs.\\ \\ref{fig4} and \\ref{rzscuti} show how the difference in the depths of the minima between the {\\sc NextGen} and Kurucz light curves depends on the bandpass, while Fig.\\ \\ref{fitfig} shows that black body light curves provide a poor simultaneous fit to {\\sc NextGen} $V$ and $J$ light curves of a binary like \\object{T CrB}.  There is no shortage of potentially good binary stars to study.  For example, \\object{T CrB}, as we have already noted, has $V$, $I$, $J$, and $K$ light curves available (although the light curves are perhaps somewhat noisy). \\object{RZ Ophiuchi} is an eclipsing binary consisting of a hot $\\approx$B star and a K supergiant.  The orbital period is 262 days.  Reasonably good light curves in the bluer filters exist (e.g.\\ Knee et al.\\ \\cite{knee86}; Olson \\cite{olson93}), and additional observations in $J$, $H$, and $K$ would be extremely valuable. Kruszewski \\& Semeniuk (\\cite{kruszewski99}) present a catalog of poorly studied eclipsing binaries with good parallaxes measured by the {\\em Hipparcos} mission. Many of these binaries have long periods (more than 10 days) and contain evolved cool components.   Needless to say, we encourage observers to systematically study these binaries in multiple colours. With a binary of known distance, one has the added challenge of obtaining the correct integrated flux.  Our interest should not be confined to binaries with a supergiant component.  For example, there appears to be a small class of ``long period'' Algol type binaries (periods between about 10 and 20 days) where the mass losing cool star appears to be slightly underfilling its Roche lobe.  Some examples are \\object{WW Andromedae} (Olson \\& Etzel \\cite{olson93a}), \\object{S Cancri} (Olson \\& Etzel \\cite{olson93b}), and \\object{DN Orionis} (Etzel \\& Olson \\cite{etzel95}).  The cool stars in these three systems have surface gravities near $\\log g\\approx 2.5$, so the sphericity effects are not as pronounced as they are in systems like \\object{T CrB}.  On the other hand, all three of these binaries have excellent five-colour ($I$(Kron)$ybvu$) photometry, and \\object{S Cnc} and \\object{DN Ori} are totally eclipsing and are double-lined. Olson and Etzel report that in each case, the cool star seems to underfill its Roche lobe by about 10 per cent.  It would be worthwhile to re-evaluate the light curve solutions for these three stars using ELC and the {\\sc NextGen} intensities to see if the sphericity effects can account for the apparent slight underfilling of the Roche lobes by the cool stars."
    },
    "0010/astro-ph0010322_arXiv.txt": {
        "abstract": "We report the discovery of the gamma-ray burst \\object{GRB~000131} and its optical afterglow. The optical identification was made with the VLT 84 hours after the burst following a BATSE detection and an Inter Planetary Network localization. \\object{GRB~000131} was a bright, long-duration GRB, with an apparent precursor signal 62 s prior to trigger. The afterglow was detected in ESO VLT, NTT, and DK1.54m follow-up observations. Broad-band and spectroscopic observations of the spectral energy distribution reveals a sharp break at optical wavelengths which is interpreted as a Ly$\\alpha$ absorption edge at 6700~{\\AA}. This places \\object{GRB~000131} at a redshift of $4.500 \\pm0.015$. The inferred isotropic energy release in gamma rays alone was $\\sim 10^{54}~{\\rm erg}$ (depending on the assumed cosmology). The rapid power-law decay of the afterglow (index $\\alpha$=2.25, similar to bursts with a prior break in the lightcurve), however, indicates collimated outflow, which relaxes the energy requirements by a factor of $<$ 200. The afterglow of \\object{GRB~000131} is the first to be identified with an 8-m class telescope. ",
        "introduction": "Since the first detections of soft X-ray \\cite{costa97}, optical \\cite{paradijs97}, and radio \\cite{frail97} afterglows to gamma-ray bursts (GRBs), 19 optical afterglows have been detected.  For the current sample of 12 GRBs with well-determined redshifts, the median redshift is z$_{\\rm median}$ = 1.1 with a very large root-mean-square variation of $\\sim$0.8. While a photometric redshift of $\\approx$ 5 has been proposed for \\object{GRB~980329} \\cite{fruchter99}, the highest spectroscopic redshift determination for a GRB so far is $z = 3.418$ for GRB~971214 \\cite{kulkarni98}. This redshift was determined from the spectrum of a faint Lyman Break galaxy that was found at the position of the optical afterglow (see also Odewahn et al., 1998). The highest redshift determined directly from absorption lines in the spectrum of an optical afterglow is $z = 2.0404\\pm0.0008$ for \\object{GRB~000301C} \\cite{smette00,jensen00}. Several authors have speculated about the possibility of detecting GRBs at even higher redshifts than achieved so far. Wijers et al.~\\cite*{wijers98} and Lamb \\& Reichart~\\cite*{lamb00} estimate that GRBs should be detectable at very high redshifts ($z\\gtrsim 5$) based on the assumption that GRBs are associated with star formation and by using models for the cosmic star formation rate as a function of redshift. Blain \\& Natarajan~\\cite*{blain00} propose to use the opposite strategy, namely to use the observed redshift distribution of GRBs to measure the (uncertain) cosmic star formation rate as a function of redshift.  Unfortunately, afterglows have been detected in only $\\lesssim 40$ percent of all well-localized (SAX, RXTE, IPN) GRBs (e.g., Fynbo et al.~2000). It is not clear why most GRBs with no apparent optical afterglow (`dark bursts') are so abundant. Possible explanations are rapid decay \\cite{groot98}, reddening in the immediate environment and host galaxy (eg., Jensen et al.~\\cite*{jensen00}), intrinsic optical faintness \\cite{taylor00}, very high redshifts \\cite{lamb00} or a combination of these effects. To shed light on dark bursts, one obvious strategy is to set deeper optical limits by using large telescopes and to determine the decay slopes and colours (see Fynbo et al.~2000).  In this Letter we present the first attempt to identify the afterglow of a GRB with an 8-m class telescope. The attempt was successful in that the observations led to the discovery of the afterglow and the subsequent determination of the decay slope, spectral energy distribution and record high redshift. In the following we document the gamma-ray, optical, and infrared observations, and discuss the properties of the gamma-ray burst and afterglow. ",
        "conclusions": "Based on the temporal and positional coincidence of the transient object with \\object{GRB~000131}, and the fact that it was located at a cosmological distance, we conclude that this transient object was the optical afterglow of \\object{GRB~000131}. Assuming a Hubble constant of 65 km~s$^{-1}$~Mpc$^{-1}$, $\\Omega_0=0.3$ and $\\Omega_\\Lambda=0.7$, a redshift of 4.50 implies a distance modulus of 48.24. Hence, the total energy release assuming isotropic emission is $\\approx$1.1$\\times$10$^{54}$ erg. This is only a factor of $\\sim$3 smaller than the isotropic equivalent energy for the most energetic event so far, \\object{GRB~990123} \\cite{andersen99,1999Natur.398..389K}. However, the combination of the decay slope $\\alpha$ and the constraints on the spectral slope $\\beta$ indicate that the afterglow was due to a collimated jet. The upper limit on the break epoch, t$_{\\rm break}$ $<$ 3.5 days, implies a lower limit on the jet opening angle of $\\theta >$ 7$^{\\circ}$ $n^{1/8}$ $>$ 7$^{\\circ}$, where $n$ is the density of the ambient medium (in units of cm$^{-3}$), which is larger than 1 in star-forming regions. From the lower limit on the opening angle we infer a lower limit on the released energy of $\\sim5\\times10^{51}$ erg.  This work shows that 8-m class telescopes may be used successfully to detect faint R~$>23$ optical afterglows even in fields with very bright stars. This is necessary in order to resolve why $\\gtrsim60\\%$ of all attempts to detect optical afterglows of GRBs are currently unsuccessful \\cite{fynbo00}. The apparent brightness of the \\object{GRB~000131} afterglow was similar to that of \\object{GRB~990123}. \\object{GRB~000131} does therefore provide the first observational evidence that it is possible to obtain high resolution optical spectra of GRBs at very high redshift, if the afterglow is identified at an early time. High redshift GRBs will therefore no doubt prove to be an extremely valuable tool for not only the understanding of the GRB environment but also for the study of the Lyman forest and of cosmology."
    },
    "0010/astro-ph0010608_arXiv.txt": {
        "abstract": "{ We describe the first XMM-Newton observations of the starburst galaxy \\ngc253. As known from previous X-ray observations, \\ngc253 shows a mixture of extended (disk and halo) and point-source emission. The high XMM-Newton throughput allows a detailed investigation of the spatial, spectral and variability properties of these components simultaneously. We characterize the brightest sources by their hardness ratios, detect a bright X-ray transient $\\sim$70\\arcsec\\ SSW of the nucleus, and show the spectrum and light curve of the brightest point source ($\\sim$30\\arcsec\\ S of the nucleus, most likely a black-hole X-ray binary, BHXRB). The unresolved emission of two disk regions can be modeled by two thin thermal plasma components (temperatures of $\\sim$ 0.13 and 0.4 keV) plus residual harder emission, with the lower temperature component originating from above the disk. The nuclear spectrum can be modeled by a three temperature plasma ($\\sim$ 0.6, 0.9, and 6 keV) with the higher temperatures increasingly absorbed. The high temperature component most likely originates from the starburst nucleus, as no non-thermal component, that would point at a significant contribution from an active nucleus (AGN), is needed. Assuming that type\\,{\\rm II}a supernova remnants (SNRs) are mostly responsible for the E$>$4 keV emission, the detection with EPIC of the 6.7\\,keV line allows us to estimate a supernova rate within the nuclear starburst of 0.2\\,yr$^{-1}$. The unprecedented combination of RGS and EPIC % also sheds new light on the emission of the complex nuclear region, the X-ray plume and the disk diffuse emission. In particular, EPIC images reveal that the limb-brightening of the plume is mostly seen in higher ionization emission lines, while in the lower ionization lines, and below 0.5\\,keV, the plume is more homogeneously structured. The plume spectrum can again be modeled by a three temperature thermal plasma containing the two low temperature nuclear components (though less absorbed) plus an unabsorbed 0.15~keV component similar to the disk spectra. This points to new interpretations as to the make up of the starburst-driven outflow. ",
        "introduction": "X-ray emission from starburst galaxies is known to be complex, revealing both point sources and diffuse X-ray emission in abundance. X-ray binaries, supernovae, supernova remnants and nuclear sources dominate the point-like source contribution, while the hot phases of the interstellar medium (ISM), in the form of hot outflows (or winds), more bound coronal features and diffuse  emission within the disk, make up the diffuse, gaseous component.  The nearby edge-on galaxy \\ngc253 is perhaps the classic example (along with \\m82) of a starburst galaxy, and as such, has received a great deal of attention from X-ray observatories over the years. Initial {\\em Einstein} observations (Fabbiano \\& Trinchieri \\cite{fab84}; Fabbiano \\cite{fab88}), along with seeing several bright point-like disk sources, discovered large plumes of diffuse emission extending above and below the disk of the galaxy.  This emission is thought to be due to ``mass-loading\" of a hotter wind with cooler ambient gas, i.e. shock heated and accelerated cooler interstellar and circumnuclear gas (Heckman et al. \\cite{hec90}; Strickland et al. \\cite{str00}).  Several studies were made of \\ngc253 with ROSAT. Both the PSPC and HRI data were presented by Read et al. (\\cite{rea97}) and Dahlem et al. (\\cite{dahlem}). An extensive ROSAT point source catalogue of \\ngc253 by Vogler \\& Pietsch (\\cite{vog99}) made it possible to separate the point source and diffuse X-ray emission, allowing insights into the spatial, spectral and timing properties of the many point sources within \\ngc253. The `nuclear', likely starburst-associated source appeared to be extended, and the brightest point-like source, lying some ~30\\arcsec\\ south of the nucleus, at the border of a plume of diffuse emission, was thought to be a plausible BHXRB candidate. Structure in the diffuse emission could now be studied in detail, Pietsch et al. (\\cite{pie00}) reporting different diffuse emission components in the nucleus, disk and halo. Especially of note was the discovery of coronal diffuse emission bubbling out of the disk, via galactic {\\em fountains} and {\\em chimneys} (essentially formed by localized high-activity star-forming regions within the disk), then falling ballistically back to the plane (e.g. Norman \\& Ikeuchi \\cite{nor89}). Also observed was a hollow-cone shaped diffuse plume of emission extending up to $\\sim$700\\,pc along the SE minor axis, thought to be due to the interaction of the galactic superwind and the dense disk ISM.  More recent high spectral resolution observations with ASCA (Ptak et al. \\cite{pta97}) have revealed strong O, Ne, Fe, Mg, S and Si emission lines in the integrated \\ngc253 spectrum, and these results have been backed up by BeppoSAX observations (Cappi et al. \\cite{cappi}), where for the first time, the 6.7\\,keV Fe K line has been detected (Persic et al. \\cite{per98}). \\chandra\\ observations allowed for the first time a detailed study of the SE plume, tentatively examined with ROSAT, and showed it to be a well-collimated, limb-brightened kpc-scale conical outflow, which closely follows in morphology the known H$\\alpha$ outflow. Furthermore, several point sources are seen, the nuclear feature being partly separated into a number of distinct sources (Strickland et al. \\cite{str00}).  In this letter we report the results of the performance verification phase observation of \\ngc253 to demonstrate XMM-Newton's capabilities of spatially resolved spectroscopy in a mixture of point sources and diffuse emission. ",
        "conclusions": ""
    },
    "0010/astro-ph0010336_arXiv.txt": {
        "abstract": "We model the evolution of canonical pulsars from the death line to millisecond pulsars through the X-ray neutron star stage of accretion from a low-mass companion. We trace this evolution in magnetic field strength starting at $B=10^{12}$ to $10^8$ G and in period of about 1 s to milliseconds. Important factors are accretion rate and the decay rate of the magnetic field. A broad swathe is traced in the $B-P$ plane according to the value of these factors, which represent different conditions of the binary pair. An important ingredient is the dependence of the stellar moment of inertia on rotation frequency (time). ",
        "introduction": "Pulsars are believed to have formed from the collapsed cores of massive stars, and are spun up by angular momentum conservation to moderate frequencies of 30 Hz or so, like the Crab pulsar. Thereafter, they radiate their angular momentum through the strong magnetic dipole field which is fixed in the star, ever more slowly as the rotational rate decreases. Depending on the magnetic field strength and  frequency, the radiation mechanism turns off and canonical pulsars disappear from the radio sky at frequencies of 1 Hz or there about, depending on the magnetic field strength. They have crossed what has been referred to as the ``death line''. Millisecond pulsars are thought to originate from these radio silent stars, being spun up by accretion from a low-mass companion. Their magnetic fields are typically $10^8$ to $ 10^9$ Gauss, as compared to $10^{11}$ to $10^{13}$  Gauss for canonical pulsars. Either the field decays ohmically during the long epoch in which they are radio and X-ray silent, or the field is partially destroyed by the accretion process, or by some as yet unknown mechanism. In this intermediate accretion phase, they are X-ray neutron          stars. The missing link between canonical pulsars with mean period of 0.7 s and millisecond pulsars was recently discovered, an X-ray star with period of 2.5 ms (\\cite{klis98:b,chakrabarty98:a,klis00:a})  So far as we know, the evolution from canonical to millisecond pulsar has never been modeled. We do so here. We find essentially a continuum of evolutionary tracks according to the rate at which matter is accreted from the companion, and the rate at which the magnetic field decays.  The evolutionary tracks essentially fill all the space in the B-P plane, starting at the death line at fields typical of canonical pulsars, and extending downward in field strength, broadening to fill the space on both sides of the deathline, and extending to the small periods of  millisecond pulsars.     All are potential tracks of some particular binary pair, since accretion rates vary by several orders of magnitude and presumably so do decay rates of the magnetic field. ",
        "conclusions": "We have computed the evolutionary tracks in the $B-P$ plane due to mass accretion onto neutron stars beginning at the death line with a typical field strength of $10^{12}$ Gauss, to shorter periods and low fields. According to the assumed accretion rate and decay constant for the magnetic field, the tracks indicate that the individual binaries with characteristics ranging from Z to Atoll sources will evolve along paths that cover a broad swathe in the $B-P$ plane. These include tracks of X-ray stars corresponding to low accretion rates that follow a path beyond the death line in the so-called ``graveyard'.  We have assumed two  particular forms for the law of decay of the magnetic field. (1) The field  approaches an asymptotic value of $10^8$ Gauss such as is typical of millisecond pulsars. This assumption leads to a particular form for the termination of evolutionary tracks. All accreters, no matter what the accretion rate, will end with millisecond periods, unless accretion ceases beforehand. (2) If instead, we had assumed a purely exponential decay, the tracks would not tend to an asymptotic value, but would continue to decrease in the strength of $B$. The tracks would still cover a broad swathe in the $B-P$ plane.  But one would conclude that only the higher accretion rate binaries, particularly the Z-sources, could produce millisecond period neutron stars. If acccretion continues for too long a time, the neutron star will be carried to very low fields and across the death line, or an overcritical mass will have been accreted, leading instead to a black hole."
    },
    "0010/astro-ph0010100_arXiv.txt": {
        "abstract": "We briefly discuss our survey of Extremely Red Objects (EROs) with HST/NICMOS and deep ground-based images and summarize its results. We use the high-resolution NICMOS images to sub-divide EROs into morphological classes, tentatively arguing that about 60 \\% of EROs (18 out of 30) are early-type galaxies. About 20 \\% (6 out of 30 objects) appear to be flattened disk systems. Another 10 \\% (3 out of 30) have an irregular morphology. Thus, combining disk and irregular systems, our data are compatible with 30 \\% of EROs being dusty starbursts. A small fraction (3 out of 30) appear point-like even on the HST images and could therefore be either compact high redshift galaxies or stars in the Milky Way halo. ",
        "introduction": "The advent of high quality near-infrared detectors has allowed a veritable breakthrough in the study of extremely red objects (hereafter EROs), {\\it i.e.} galaxies characterized by $R-K$ colors of 5 or redder. Clearly, a natural interpretation of their red colors is that EROs are elliptical galaxies with relatively old stellar populations at a redshift between 1 and 2.  This has been confirmed by Spinrad et al. (1997) for the ERO with regular morphology LBDS 53W091. However, Graham \\& Dey (1996) have shown that another ERO, HR10, actually has a spectrum with emission lines. Cimatti et al. (1998) definitely confirmed the dusty nature of HR10. Clearly, on the basis of colors alone it is not possible to discriminate between a dusty starburst or an old stellar population at $z>1$. On the other hand, spectroscopy of a large sample of EROs is, unfortunately, prohibitive even with 8-meter class telescope ({\\it e.g.} Cimatti et al. 2000).  One important difference between LBDS 53W091 and HR 10 is their morphology.  At HST resolution, the former appears regular while the latter has an irregular morphology.  This has led us to adopt as a working hypothesis the assumption that morphology with HST can help identify the nature of EROs. Therefore, we have carried out a survey of EROs in fields where HST/NICMOS images were available.  In Section 2, we describe our survey and some basic results. Section 3 is devoted to a discussion of morphologies. Section 4 sums up. ",
        "conclusions": "Our results suggest that some early-type galaxies were in place at high redshift. We detect a deficit of red objects which may imply either that some early-type galaxies formed later, or that observational effects or interactions interve to make the colors bluer and thus cause some early-type galaxies to drop from the E/S0 ERO sample.  We find a number density of about 0.5 arcmin$^{-2}$ (at H$_{160} = 24.5$) of EROs that appear to be as flattened as a disk. They could be massive disks already in place at $z>1$. We expect these objects as well as the irregular EROs to be characterized by a dusty starburst spectrum.  The major limitation of our study is the relatively small area of the survey especially in light of the large field-to-field fluctuations discovered by Daddi et al. (2000) for the bright-end of the ERO distribution. For this reason, we are planning to increase the area of our survey by a factor four in the near future."
    },
    "0010/astro-ph0010270_arXiv.txt": {
        "abstract": "\\ni In currently popular cosmological scenarios -- all variants of the cold dark matter (CDM) cosmogony -- some time beyond a redshift of 15, stars within the numerous small halos that condense with virial temperatures $\\sim 10^4\\,$K created the first heavy elements; these protogalactic systems, together perhaps with an early population of mini-quasars, generated the ultraviolet radiation and mechanical energy that reheated and reionized the cosmos. The history of the Universe during and soon after these crucial formative stages is recorded in the thermal, ionization, and chemical state of the all-pervading intergalactic medium (IGM), which contains most of the ordinary baryonic material left over from the big bang. Throughout the epoch of structure formation, the IGM becomes clumpy under the influence of gravity, and acts as a source for the gas that gets accreted, cools, and forms stars within subgalactic fragments, and as a sink for the metal enriched material, energy, and radiation which they eject. ",
        "introduction": "\\ni At epochs corresponding to $z\\sim 1000$ the intergalactic medium is expected to recombine and remain neutral until sources of radiation and heat develop that are capable of reionizing it. The detection of transmitted flux shortward of the \\lya wavelength in the spectra of sources at $z\\sim 6$ implies that the hydrogen component of this IGM was ionized at even higher redshifts. It appears then that substantial sources of ultraviolet photons and mechanical energy were already present when the Universe was less than 6\\% of its current age, perhaps quasars and/or young star-forming galaxies: an episode of pregalactic star formation may provide a possible explanation for the widespread existence of heavy elements (like carbon, oxygen, and silicon) in the IGM, while the integrated radiation emitted from quasars is likely responsible for the reionization of intergalactic helium at later times.  Popular cosmological models predict that some time beyond a redshift of 15 the gas within halos with virial temperatures $T_v\\ga 10^4\\,$K [or, equivalently, with masses $M\\ga 10^9\\, (1+z)^{-3/2}h^{-1}\\,\\msun$ comparable to present-day dwarf ellipticals] cooled rapidly due to the excitation of hydrogen \\Lya by the Maxwellian tail of the electron distribution, and fragmented. Massive stars formed with some initial mass function (IMF), synthesized heavy elements, and exploded as Type II supernovae (SNe) after a few $\\times 10^7\\,$yr, enriching and heating the surrounding medium. Whilst collisional excitation of molecular hydrogen may have allowed the gas in even smaller systems [virial temperatures of only a few hundred K, corresponding to masses around $10^7\\,(1+z)^{-3/2}h^{-1}\\,\\msun$] to cool and form stars at earlier times (Abel, Ciardi, and Ferrara, this volume), H$_2$ molecules are efficiently photo-dissociated by stellar UV radiation, and such negative `feedback' is likely to have suppressed molecular cooling and further star formation inside very small halos. One should note that while numerical N-body$+$hydrodynamical simulations have convincingly shown that the IGM is expected to fragment into structures at early times in CDM cosmogonies (e.g. Cen \\etal 1994; Zhang \\etal 1995), the same simulations are much less able to predict the efficiency with which the first gravitationally collapsed objects lit up the Universe at the end of the `dark age'. ",
        "conclusions": ""
    },
    "0010/astro-ph0010046_arXiv.txt": {
        "abstract": "We present high-resolution {\\sl HST} imaging in the optical ({\\sl WFPC2}) and near-infrared ({\\sl NICMOS}) of a disk region 1 kpc NE of the starburst core in the nearby galaxy M82.  This region, M82 ``B,'' has been suspected to be a fossil starburst site in which an intense episode of star formation occurred over 100 Myr ago, and our new observations confirm this intepretation.  M82 thus presents us with the opportunity to observe both active and evolved starburst environments at close range.  The surface brightness of M82 B is well above normal for galactic disks and comparable to the core surface brightnesses in spiral galaxies.  Its intrinsic surface brightness at an age of 10 Myr was comparable to that found in the present-day nuclear starburst, indicating an event of comparable amplitude.  We find a large, evolved system of super star clusters in M82 B.  Using size as a criterion to distinguish cluster candidates from point sources, we identify a total of 113 super star cluster candidates.  We use a two-color {\\it BVI} diagram and evolutionary spectral synthesis models to separately estimate the extinction and age of each cluster. The clusters range in absolute magnitude from $M_V^0 = -6$ to --10, with a peak at --7.5.  The derived age distribution suggests steady, continuing cluster formation at a modest rate at early times ($> 2$ Gyr ago), followed by a concentrated formation episode $\\sim 600$ Myr ago and more recent suppression of cluster formation.  The peak episode coincides with independent dynamical estimates for the last tidal encounter with M81, which presumably induced the starburst.  Our {\\it J} and {\\it H} band observations resolve the bright giant population in M82's disk for the first time.  Star formation evidently continued in M82 B until about 20 -- 30 Myr ago, but none is found associated with the youngest generations in the nuclear starburst (age $\\lesssim 15$ Myr).  After correcting the cluster luminosity function to a fiducial age of 50 Myr, we find that the bright end is characterized by a power-law slope with $\\alpha = -1.2 \\pm 0.3$, similar to that of other young cluster systems in interacting galaxies.  There is tentative evidence for broadening of the luminosity function due to dynamical destruction of lower mass clusters.  Cluster sizes ($2.34 \\le R_{\\rm eff} \\lesssim 10$ pc, or $2.4 \\lesssim R_{\\rm core} \\lesssim 7.9$ pc) and estimated masses (a median of $10^5 M_\\odot$) are consistent with values found for young super star cluster populations in M82's core and other galaxies and with the progenitors of globular clusters. ",
        "introduction": "\\subsection{A fossil starburst?} \\label{intro1.sect}  M82 is the prototype starburst galaxy.  Observations at all wavelengths from radio to X-rays (reviewed in Telesco 1988 and Rieke et al.\\ 1993) are consistent with the following scenario.  During the last several 100 Myr, tidal interactions between M82 and M81 and/or other member galaxies of the M81 group have channeled large amounts of gas into the central regions of M82.  This has induced a starburst which has continued for up to about 50 Myr at a star formation rate of $\\sim 10$ M$_\\odot$ yr$^{-1}$.  Energy deposited by supernovae, at a rate of $\\sim 0.1$ supernova yr$^{-1}$ (e.g.\\ O'Connell \\& Mangano 1978, hereinafter OM; Rieke et al.  1980, McLeod et al.\\ 1993), drives a large-scale galactic wind along the minor axis of M82 (e.g.\\ Lynds \\& Sandage 1963, Fabbiano \\& Trinchieri 1984, Watson, Stanger, \\& Griffiths 1984, McCarthy, Heckman, \\& van Breugel 1987, Shopbell \\& Bland-Hawthorn 1998, Strickland, Ponman, \\& Stevens 1997).  All of the bright radio and infrared (IR) sources associated with the active starburst are confined to the galaxy's center.  They lie within a radius of $\\sim 250$ pc and mostly suffer heavy extinction by dust, estimated to be in the range ${\\rm A}_{\\rm V} \\sim$ 5--25 (e.g.  Telesco et al.\\ 1991, McLeod et al.\\ 1993, Satyapal et al.\\ 1995).  They correspond spatially with bright optical structures, labeled M82 A, C, and E in OM.  These are probably those parts of the starburst core which happen to be the least obscured along the line of sight (O'Connell et al.\\ 1995).  However, there is good photometric and spectroscopic evidence that this was not the only major starburst episode to have occurred in M82.  A region about 1 kpc NE from the galactic center, M82 B (cf.\\ OM and Fig. 1), has exactly the properties one would predict for a fossil starburst with an amplitude similar to the active burst in M82 A, C and E.  M82 B has an intrinsic surface brightness of $\\mu_V^0 \\sim 16.5$ mag arcsec$^{-2}$ (measured in a $7''$ aperture).  Although this is fainter than the intrinsic {\\it V}-band surface brightness of M82 A, C, and E in the active starburst, it is nonetheless much higher than the maximum brightnesses encountered in spiral disks, even when seen edge-on.  For the disks of typical edge-on Sb-Sd galaxies, de Grijs (1998) finds central surface brightnesses of order $\\mu_B \\sim 20 - 22$ and $\\mu_I \\sim 18 - 20$ mag arcsec$^{-2}$.  The surface brightness of M82 B is, in fact, comparable to the {\\it nuclear} brightnesses of bright Sb-Sc galaxies measured with similar resolution (e.g.\\ de Jong 1996).  By extrapolating region B's surface brightness back to an age of 10 Myr, using the evolutionary synthesis models of Bruzual \\& Charlot (1996), we estimate that its surface brightness was of order $\\mu_V^0 \\sim 14.5$ mag arcsec$^{-2}$, similar to that presently observed in the active starburst M82 A.  The spectrum of M82 B is likewise consistent with a fossil starburst event, entirely comparable to that now transpiring in M82 A.  It is dominated by Balmer absorption lines and shows a large Balmer discontinuity (OM, Marcum \\& O'Connell 1996).  Emission lines are very weak, unlike in M82 A, which has intense line emission.  These are the hallmarks of the anomalous ``E+A'' spectra found in distant galaxy clusters that exhibit the Butcher-Oemler effect (Butcher \\& Oemler 1978, Oemler 1992).  Such spectra are generally interpreted as the signature of a truncated burst of star formation that occurred 100--1000 Myr earlier (e.g.\\ Couch \\& Sharples 1987, Dressler \\& Gunn 1990).  These starbursts are thought to be part of the process by which disk galaxies are converted to elliptical or lenticular galaxies and probably result from tidal interactions, mergers, or perhaps ram-pressure stripping by the intergalactic medium (Butcher \\& Oemler 1978, Oemler 1992, Barger et al.\\ 1996).  From 20--40 \\AA\\ resolution spectrophotometry, OM and Marcum \\& O'Connell (1996) find evidence of a sharp main-sequence truncation in M82 B, corresponding to a turnoff age of $\\sim$ 100--200 Myr and an average extinction of $A_V \\sim 0.6$ mag.  Region A's spectrum, on the other hand, indicates a very young population ($\\sim 5$ Myr) and is more heavily affected by extinction ($A_V \\sim 2.2$ mag).  Apart from the physical properties of M82 B, three other considerations support the notion that multiple starbursts have occurred in M82. First, the triggering gravitational interactions between M82 and other members of the M81 group and the associated tidal gas flows have been proceeding for more than 200 Myr (e.g.\\ Yun, Ho, \\& Lo 1994), in which time multiple starbursts could easily have occurred.  Second, starbursts are unlikely to be synchronized throughout the disturbed galaxy, and Satyapal et al.\\ (1997) find evidence of outward propagation of star formation in the dense core of M82.  Finally, it is likely that starbursts are strongly self-limited, or quenched, by supernova-driven outflows, which remove the remaining cool gas from the immediate starburst region (e.g.\\ Chevalier \\& Clegg 1985, Doane \\& Mathews 1993). The remarkable minor-axis wind in M82 is a dramatic example of this process.  However, the disturbed conditions near an early burst may discourage re-ignition at the same site when cool gas inflows resume, shifting the location of active star formation.  Although other interesting examples of relatively nearby post-starburst systems have been located (e.g.\\ Vigroux, Boulade, \\& Rose 1989, Oegerle, Hill, \\& Hoessel 1991, Caldwell et al.\\ 1993, Caldwell \\& Rose 1997) none of these, nor any other nearby starburst, offer the opportunity to study two discrete starbursts at such close range as M82.  The active starburst in the inner 250 pc of M82 has been studied thoroughly, whereas the fossil starburst in region B has not received much attention.  The aim of this paper is to study the remains of this earlier episode in detail, with emphasis on young star cluster candidates.  \\subsection{Super Star Clusters in M82}  Using {\\sl Hubble Space Telescope (HST)} imaging of the bright central regions A, C, and E of M82, O'Connell et al.\\ (1995) resolved these into a swarm of over 100 young star cluster candidates, with a mean $L_V \\sim 4 \\times 10^6 L_\\odot$, brighter than any globular cluster in the Local Group.  Satyapal et al.\\ (1997) found evidence in the near-IR of a dozen compact clusters lying deeper within the heavily obscured core of the galaxy near A, C, and E.  The core clusters show an age dispersion of 6 Myr and indications of outward propagation of star formation.  Gallagher \\& Smith (1999) determined a significantly older age of 60 Myr for the luminous cluster F, which is located 440 pc South-West of the nucleus.  Such ``super'' star clusters (SSCs) have been discovered, mostly with {\\sl HST}, in other interacting, amorphous, dwarf, and starburst galaxies (e.g.\\ Arp \\& Sandage 1985, Melnick, Moles, \\& Terlevich 1985, Meurer et al.\\ 1992, Holtzman et al.\\ 1992, Whitmore et al.\\ 1993, 1999, Hunter et al.\\ 1994, 2000, O'Connell, Gallagher, \\& Hunter 1994, Ho \\& Filippenko 1996b, Conti, Leitherer, \\& Vacca 1996, Ho 1997, Watson et al.\\ 1996, Carlson et al.\\ 1998).  Their diameters, luminosities, and -- in several cases -- masses are consistent with these being young {\\it globular} clusters formed as a result of recent gas flows (e.g.\\ van den Bergh 1995, Meurer 1995, Ho \\& Filippenko 1996a, Hunter et al.\\ 2000). In the case of M82 F, Smith \\& Gallagher (2000) obtained a dynamical mass of $2 \\times 10^6 M_\\odot$.  It is possible that a large fraction of the star formation in starbursts takes place in the form of such concentrated clusters.  The discovery that globular cluster formation, once thought to occur only during early stages of galaxy evolution, continues today is one of {\\sl HST}'s main contributions to astrophysics to date.  M82 is one of the few galaxies containing a rich, starburst-induced cluster system which is near enough that the internal structure of individual SSCs can be studied.  Most cluster systems are too distant for such work, even with {\\sl HST}.  Apart from their intrinsic interest, SSCs can be individually age-dated and are therefore important tracers of the history of the starburst process across the face of a galaxy like M82.  If region B is indeed a fossil starburst site, it probably contains evolved clusters which originally had properties similar to those now observed in region A.  The presence of both the active and the fossil starburst sites in M82 therefore provides a unique arena for the study of the stellar and dynamical evolution of star cluster systems.  Even in the time interval of 0.3--1 Gyr that has probably elapsed since the starburst in region B, evolutionary effects might be detectable because of the dense environment of the starburst.  In order to explore the fossil starburst site and to detect candidate SSCs, we therefore made new, multiband optical/near-IR observations of region B with {\\sl HST}.  In \\S 2, we provide an overview of the observations and the image processing techniques applied.  \\S\\S 3 and 4 describe the general characteristics of region B; \\S 5 gives an outline of the analysis techniques used to select our star cluster candidates and obtain the results presented in the subsequent sections.  We report the first detection of the resolved field stellar population in the disk of M82 in \\S 6.  A discussion of the ages of the star cluster sample, in terms of evolutionary spectral synthesis models is presented in \\S 7.  The cluster luminosity functions and estimated masses are discussed in \\S\\S 8 and 9, respectively.  In \\S 10 we present a detailed analysis of the cluster profiles.  Finally, in \\S 11 we summarize our results and conclusions. ",
        "conclusions": "\\label{summ.sect}  We have presented high-resolution {\\sl HST WFPC2} and {\\it NICMOS} imaging of two adjacent fields in the disk 1 kpc NE of the center of M82, the prototype local starburst galaxy.  This region, M82 ``B,'' has been suspected to be a site where an intense episode of star formation occurred over 100 Myr ago.  Our new observations confirm that M82 B is a ``fossil'' of an ancient starburst with an amplitude entirely comparable to that now transpiring in the core of M82.  It is an analog of the anomalous E$+$A systems found in distant galaxy clusters.  M82 thus presents us with the unique opportunity to observe both active and evolved starburst environments at close range.  The main results of our multi-passband study are summarized below:  \\begin{itemize}  \\item The surface brightness of M82 B is well above normal for galactic disks and comparable to the core surface brightnesses in spiral galaxies.  Its intrinsic surface brightness at an age of 10 Myr was comparable to that found in the present-day nuclear starburst, indicating an event of comparable amplitude.  \\item Because of the bright unresolved background light and the large and highly variable extinction within M82, we have been forced to use specialized methods to identify and photometer sources.  We tested selection and photometry methods using artificial sources or real star field images from {\\sl HST} injected into our observed images.  The proximity of M82, however, is a key advantage relative to studies of more distant galaxies systems because we can use source sizes as a criterion to distinguish stars from star clusters.  \\item We find a large, evolved system of SSCs in M82 B.  Our cluster candidate samples consist of well-defined sources with Gaussian $\\sigma_{\\rm G} \\ge 1.25$ pixels (0\\asec0569, or 1.0 pc), $V \\le 22.5$ mag, and relatively smooth backgrounds.  These samples contain 43 and 70 cluster candidates in regions B1 and B2 (nearest the M82 nucleus), respectively.  \\item We use a two-color {\\it BVI} diagram and evolutionary spectral synthesis models to separately estimate the extinction and age of each cluster.  The clusters range in absolute magnitude from $M_V^0 = -6$ to --10, with a peak at --7.5.  The derived age distribution suggests steady, continuing cluster formation at a modest rate at early times ($> 2$ Gyr ago), followed by a concentrated formation episode $\\sim 600$ Myr ago and more recent suppression of cluster formation.  The peak episode coincides with independent dynamical estimates for the last tidal encounter with M81, which presumably induced the starburst.  \\item Our {\\it J} and {\\it H} band observations resolve the stellar population in M82's disk for the first time.  Comparing our near-IR color-magnitude diagram to recent Padova isochrones, the detected stars appear to be cool, helium-burning bright giants.  Star formation evidently continued in M82 B until about 20 -- 30 Myr ago.  However, significant star formation has not occurred in M82 B in the last 10--15 Myr, during which the nuclear starburst has been highly active.  \\item After correcting the cluster luminosity function to a fiducial age of 50 Myr, we find that the bright end is characterized by a power-law slope with $\\alpha = -1.2 \\pm 0.3$, similar to that of other young cluster systems in interacting galaxies.  There is tentative evidence for broadening of the luminosity function due to dynamical destruction of lower mass clusters.  \\item The radial luminosity profiles of the brightest clusters are more closely approximated by modified Hubble or ``Elson'' functions with $\\gamma = $ 2.0 or 2.6 than by Gaussians.  Some of the clusters show evidence for asymmetries or subclustering, possibly an indication of mergers.  Cluster sizes ($2.34 \\le R_{\\rm eff} \\lesssim 10$ pc, or $2.4 \\lesssim R_{\\rm core} \\lesssim 7.9$ pc) and estimated masses (a median of $10^5 M_\\odot$) are consistent with values found for young super star cluster populations in M82's core and other galaxies and with the progenitors of globular clusters.  \\end{itemize}  When we combine the age estimates for the evolved cluster sample in M82 B, for the integrated spectrum of M82 B (Marcum \\& O'Connell 1996), and for the resolved cool giants in the disk of region B, the following picture for the star formation history in its disk emerges.  The last tidal encounter between M82 and M81 about 500 Myr ago had a major impact on what was probably an otherwise normal, quiescent, disk galaxy.  It caused a concentrated burst of star formation activity, as evidenced by the peak in the age distribution of the cluster sample in M82 B.  Comparison of the cluster ages with the integrated light dating ($\\sim 100$--200 Myr) suggests that field star formation may have continued at a high rate after cluster formation had begun to decline, but the uncertainties in the methods are too large to be certain.  The enhanced cluster formation decreased rapidly within a few hundred Myr of its peak.  However, field star formation continued, probably at a much lower rate, in M82 B until $\\sim 20$ Myr ago.  It has evidently been suppressed during the last $\\sim 10 - 15$ Myr, during which the starburst in the core of M82 has been most active.  Evidence for supernova remnants in the parts of region B2 nearest the starburst core (de Grijs et al.\\ 2000) but not in region B1 indicates that disk star formation during the last 50 Myr was more active nearer the nucleus. The current starburst is probably related to late infall of tidally disrupted debris from M82 itself.  The evidence for decoupling between cluster and field star formation is consistent with the view that SSC formation requires special conditions, e.g.  large scale gas flows, in addition to the presence of dense gas (cf.  Ashman \\& Zepf 1992, Elmegreen \\& Efremov 1997).  A strong tidal interaction could easily produce an off-nuclear starburst at a site like M82 B.  However, the M82 B burst could also have been part of a larger scale event encompassing the center of the galaxy as well.  Given the high extinction and the dominance by much younger concentrations of stars, it would not be easy to identify older clusters in the starburst core if they exist.  All of our observational evidence points to a scenario in which the intermediate-age star cluster population in M82 B is entirely comparable to SSC populations at a younger age, as seen in M82 A and other galaxies.  They will likely evolve into Galactic globular cluster analogs over a Hubble time.  \\paragraph{Acknowledgements} - We thank Noella D'Cruz for helping us to check our optical photometry, Andrew Stephens and Jay Frogel for making their {\\sl NICMOS} calibration coefficients available prior to publication, Allan Sandage for loan of the plate shown in Fig.  1, Lynn Matthews for providing an independent check on our source size estimates, M\\'arcio Catelan for useful discussions about the interpretation of the stellar disk light in M82, and Mark Whittle for many insightful and stimulating discussions.  This research was supported by NASA grants NAG 5-3428 and NAG 5-6403, and has made use of NASA's Astrophysics Data System Abstract Service and of {\\sl HST} archival data at the STScI."
    },
    "0010/astro-ph0010516_arXiv.txt": {
        "abstract": "The main goal of the FIRST-APM QSO Survey (FAQS) survey is to compile the most complete sample of Bright QSOs, located in the area covered by the Second Byurakan Survey (SBS). Here we report the first results of an ongoing study based on the cross-identification of the FIRST radio catalog and the APM optical catalog. The overlapping sky area between  FIRST and SBS is about 700 deg$^{2}$. The compiled list of sources for this overlapping region contains $\\sim 400$ quasar candidates brighter than $B=18\\fm5$. About 90 objects were already spectroscopically classified.  During 1999-2000, we observed spectroscopically more than 150 FAQS objects with the 2.1m telescope of the Guillermo Haro Observatory (GHO).We have found 51 new QSOs (4 BAL QSOs), 13 Seyfert Galaxies (5 NLSy1's), 23 emission line galaxies, 3 BL Lac objects and 57 stars. ",
        "introduction": "Optically selected quasar samples are the most complete ones. However, every survey technique has a redshift and a luminosity-dependent selection biases (Wampler and Ponz 1985). Only the combination of the different search techniques in different spectral ranges, will yield a complete sample of quasars, with an adequate representation of the properties inherent to them (Hartwick \\& Shade 1990; Chavushyan 1995). We have started a multiwavelength search for QSOs in the well optically investigated SBS sky area. (Stepanian et al. 1999, and references therein).  The FIRST radio survey (Becker et al. 1995) provides a new resource for constructing a large quasar sample. With positions accurate to 1\\arcsec ~and a point source sensitivity limit of 1 mJy, it goes 50 times fainter in flux than any previous radio survey. The FBQS (Gregg et al. 1996; White et al. 2000) is a QSO survey based on FIRST. Unfortunately, for the FBQS area there is a lack of complete optically selected samples of QSOs. Hence, it is very difficult to estimate the completeness of this latter survey, as well as, the number of missed quasars in the optical FBQS sample. ",
        "conclusions": ""
    },
    "0010/astro-ph0010385_arXiv.txt": {
        "abstract": "We present a model for the hydrogenation states of Polycyclic Aromatic Hydrocarbons (PAHs) in the diffuse interstellar medium. First, we study the abundance of hydrogenation and charge states  of PAHs due to photo-ionization, photo-dissociation in the interstellar UV field, electron recombination and chemical reactions between PAH cations and H or H$_2$. For PAH cations, we find that the dehydrogenation effects are dominant. The hydrogenation state of PAHs  depends strongly on the H density, the size of the molecule and UV field. In  diffuse clouds with  low H density and normal UV radiation, PAHs containing less than 40 C are completely or strongly dehydrogenated whereas at high H density, they are normally hydrogenated. The partially dehydrogenated species dominate in intermediate density clouds. PAHs above 40 C are quite stable and are fully hydrogenated, which would favor their spectroscopic search in near IR surveys of Diffuse Interstellar Bands (DIBs). ",
        "introduction": "In recent years, it has been suggested that a population of large carbonaceous molecules, such as Polycyclic Aromatic Hydrocarbons (PAHs) and fullerene compounds could play a significant role in the physics and chemistry of interstellar and circumstellar environments. These molecules have been proposed as the sources of the broad infrared emission features (Duley \\& Williams 1981, L\\'eger \\& Puget 1984, Allamandola, Tielens \\& Barker 1985) and of the diffuse interstellar absorption bands (L\\'eger \\& d'Hendecourt 1985, van der Zwet \\& Allamandola 1985, Crawford, Tielens \\& Allamandola 1985). PAHs would represent the most abundant molecules in the interstellar medium after H$_2$ and CO (L\\'eger \\& d'Hendecourt 1985). They could exist under various charge and hydrogenation states depending on local conditions (density, UV flux...). Environment studies of some DIBs suggest carriers with ionization potential of 10-13~eV such as PAH cations (Cami et al. 1997, Sonnentrucker et al. 1997, 1999).  The aim of this work is to propose a description of the abundance of different hydrogenation states for some small and medium PAHs $\\leq$ 50C under the physical conditions present in the interstellar medium, especially in diffuse environments where UV photons strongly influence the dissociation of PAHs.  We first compare the effect of physical processes involving PAH cations and find the balance dominated by  photo-dissociation and hydrogenation processes. Then we calculate the distribution of hydrogenation states for re\\-presentative PAH cations. ",
        "conclusions": "\\subsection{Dependence on H density and on molecule size}  The hydrogenation state strongly depends on the H density. Two processes are competing: that of the UV radiation which leads to the dehydrogenation of  PAH cations and the reaction between PAH cations and H which favors hydrogenated states.  \\begin{figure*} \\hspace{-0.5cm} \\begin{tabular}{cc} \\psfig{file=Cg282fig2a.ps,height=6.4cm} & \\psfig{file=Cg282fig2b.ps,height=6.4cm} \\\\ \\psfig{file=Cg282fig2c.ps,height=6.4cm} & \\psfig{file=Cg282fig2d.ps,height=6.4cm} \\\\ \\end{tabular} \\label{coro} \\label{ova}  \\caption{Different hydrogenation states of a) the pyrene cation C$_{16}$H$_{10}^+$, b) the coronene cation C$_{24}$H$_{12}^+$, c) the ovalene cation C$_{32}$H$_{14}^+$ and d) the PAH 50C cation as a function of H density $n_{\\rm H}$ in the presence of UV (Gondhalekar spectrum multiplied by a constant factor $\\it \\Phi_\\nu$). At low densities, strongly dehydrogenated species dominate whereas at $n_{\\rm H}$ $>$ 20, 30, 5 cm$^{-3}$ respectively for the pyrene cation, the coronene cation, the ovalene cation, partially dehydrogenated species are important. The normal hydrogenated PAH 50C cation is stable over a large range of densities.}  \\label{pyr} \\label{PAH50}  \\end{figure*} As we can see in the figures 2, three phases exist for each hydrogenation state of PAHs except for PAHs $>$ 40C. The UV intensity is assumed to be constant while the H density varies. At low density, PAH cations are completely or strongly dehydrogenated but at high density, they are normally hydrogenated. At intermediate density, PAH cations are partially dehydrogenated.  For PAH cations $\\geq$ 50C, the completely hydrogenated state is favored (cf. Fig.\\ref{PAH50}d). This can be explained by the fact that large molecules have more degrees of freedom available to store the excess energy brought by the UV photon, which decreases the photodissociation rate.  \\subsection{Search for PAHs and link with DIBs}  Small PAHs $<$ 30C will be difficult to search because their abundance is distributed over a number of hydrogenation states. Their superimposed absorptions can contribute to the envelope of quasi continuum extinction in addition to dust due to a large number of states and isomers. PAHs $>$ 40C are stable and fully hydrogenated. Their signature can be searched in the fundamental band transition expected in the far red  and near infra-red around 1 $\\mu$m (Foing\\- \\& Ehrenfreund 1994). Rotational contours for such PAH $>$ 40C would lead to FWHM broadening of 2-4 cm$^{-1}$ depending on the exact size and geometry of the molecule (Ehrenfreund \\& Foing 1996). These transitions can be searched in near IR surveys of DIBs. Laboratory spectroscopy and dissociation and hydrogenation rates are also needed for guiding the search for PAHs in space."
    },
    "0010/astro-ph0010499_arXiv.txt": {
        "abstract": "We have studied the fate of initial magnetic fields in the hot halo gas out of which the visible parts of galaxies form, using three-dimensional numerical MHD-experiments.  The halo gas undergoes compression by several orders of magnitude in the subsonic cooling flow that forms the cold disk. The magnetic field is carried along and is amplified considerably in the process, reaching $\\mu$G levels for reasonable values of the initial ratio of magnetic to thermal energy density. ",
        "introduction": "The origin of large scale magnetic fields in disk galaxies is an unsettled issue. Proposed solutions to this problem fall in two categories. On the one hand, in-situ dynamo action has been invoked to exponentially amplify an initially weak, small scale, seed field (for recent reviews, see Beck et al. 1996; Kulsrud 1999). \\nocite{Beck+Brandenburg+Moss+;1996,Kulsrud;1999} On the other hand, large scale pregalactic fields in the halos of forming galaxies might be compressed and stretched sufficiently during the formation and evolution of galactic disks to explain the observed large scale fields (see e.g.~Howard \\& Kulsrud 1997). \\nocite{Howard+Kulsrud;1997}  Observations of $\\mu$G level large scale fields in slowly rotating irregular galaxies (Chy{\\.z}y et al.~2000), \\nocite{Chyzy+Beck+Kohle+;2000} high redshift objects (e.g.~Perry et al.~1993), \\nocite{Perry+Watson+Kronberg+;1993} and in the intracluster medium (Eilek 1999; Colgate \\& Li 2000), \\nocite{Eilek;1999,Colgate+Li;2000} pose fundamental problems for the classical $\\alpha-\\Omega$ galactic dynamo as a cornerstone for creating magnetic fields on galactic (and larger) scales. In addition, it seems difficult to attribute the strong vertical field found in the central molecular zone of our Galaxy to a galactic dynamo, which may be suggestive of a primordial origin for this component of the galactic magnetic field (Sofue \\& Fujimoto 1987; Chandran, Cowley, \\& Morris 2000). \\nocite{Sofue+Fujimoto;1987,Chandran+Cowley+Morris;2000}  The observational evidence for magnetic fields in cluster halos and the association between these fields and outflows from AGNs suggests that similar processes may seed protogalactic halos with large scale magnetic fields. The compression and stretching of the field during the galaxy formation process further amplifies the halo field and could possibly account for the observed fields in present day disk galaxies. The conventional criticism that this kind of scenarios will result in an incorrect parity with respect to the galactic plane may be shown to result from overly simplified assumptions about the initial field.   Only in the case of a field that is both weakly inclined and initially located near the center of the potential well does it follow that the wound-up field is predominantly of odd parity.  %  \\begin{figure} \\plottwo{B_topo.ps}{eB.ps} \\caption[]{ (a) The magnetic energy isosurface (dark gray) shows the asymmetry of the field as it gets dragged into the cold disk (light gray) with the cooling flow (for 3-D color renderings see \\verb'www.astro.ku.dk/'${\\mathtt\\sim}$\\verb'aake/talks/IAU2000'). Field lines show that the polarity of the field changes in the halo, above the plane of the disk. (b) The growth of the average magnetic energy in a volume enclosing the cold disk and the halo gas immediately above and below the disk. The initial magnetic pressure in the ``dynamic'' experiment is 1\\% of the thermal pressure, while the kinematic case starts off with 0.01\\%. } \\label{fig:Btopo} \\end{figure}  This has prompted us to study the compression and wind-up of initial halo fields during the build-up of a disk galaxy in a cooling flow. By adopting a smooth background potential and by using sufficient numerical resolution (enhanced by grid stretching) we achieve that the disk forming at the bottom of the potential well has the characteristic size, mass, and specific angular momentum of a typical disk galaxy (the initial spin parameter $\\lambda = {J E^{1/2} / G M^{5/2}}$ is equal to 0.056). The hot halo gas is threaded by a random magnetic field with approximately constant $P_{\\rm mag}/P_{\\rm gas}$, plus a large scale component of similar energy content that has a non-zero inclination and offset from the rotation axis. ",
        "conclusions": "\\label{sec:discussion}  If the initial field has a significant inclination and offset from the center, the resulting wound-up field has a reversal that is offset into the halo (see Fig.~1a). When observed face-on, in optically thin radio frequencies, the magnetic field would appear to be uni-directional, as is normally observed, because one of the winding directions dominates over the other. The reversal in the halo could be revealed by analyzing observations in different radio frequencies, as has indeed been done for M51 (Berkhuijsen et al.~1997). \\nocite{Berkhuijsen+Horellou+Krause+;1997} Note that, in the model presented here, there is no creation of ``new'' magnetic flux---the growth of magnetic energy (Fig.~1b) is purely a result of compression and winding of an (approximately conserved) amount of magnetic flux, initially threading the disk plane over a range of radii.  With sufficient numerical resolution of the disk dynamics, there would of course be additional fine scale structure and enhancement of the magnetic energy, associated with turbulence in the disk.  The other common point of criticism against galactic magnetic fields originating from large scale, pre-galactic fields is the issue of pitch angle.  Here it should be remarked that conventional mean field dynamos invoke a large amount of magnetic diffusion, particularly in spiral arms, together with a rapid regeneration of the radial magnetic field component, to maintain the pitch angle.  The magnetic diffusion is assumed to arise, for example, from turbulence driven by star formation.  If such a mechanism was allowed for in the current experiments, it would be equally effective in maintaining the pitch angle.  Regeneration of the radial field component is not necessary; apart from vertical transport effects, its magnitude is determined by the (conserved) amount of wound-up magnetic flux and the pitch angle. It has also been argued that galactic winds, caused by the same SNe that drive the turbulence, could be a significant source of loss of magnetic field from galactic disks.  However, as has been demonstrated by Gudiksen (1999), \\nocite{Gudiksen;1999} the hot wind emanates from the disc in ``chimneys'', which are intermixed with much denser regions that are able to hold on to the magnetic field. Loss of magnetic flux is also counteracted by the incoming cooling flow.  In summary, the scenario adopted here is consistent both with current ideas about early AGNs as the origin of the observed, large scale magnetic fields in clusters, and with observed properties of galactic magnetic fields.  Since the results were obtained by simply tracing the magnetic field evolution in a cooling flow that results in a disk galaxy which is in other respect consistent with observations, one may even turn the argument around and conclude that the initial $P_{\\rm mag}/P_{\\rm gas}$ of the halo gas would have to be much smaller than the $P_{\\rm mag}/P_{\\rm gas}$ typical of large scale cluster magnetic fields, in order not to give rise to galactic magnetic fields similar to the observed ones."
    },
    "0010/astro-ph0010450_arXiv.txt": {
        "abstract": "We analyzed a deep {\\it Chandra} observation of M84, a bright elliptical galaxy in the core of the Virgo cluster. We find that the spatial distribution of the soft X-ray emission is defined by the radio structure of the galaxy.  In particular we find two low density regions associated with the radio lobes and surrounded by higher density X-ray filaments. In addition to a central AGN and a population of galactic sources, we find a diffuse hard source filling the central 10 kpc region. Since the morphology of the hard source appears round and is different from that seen in the radio or in soft X-rays, we propose that it is hot gas heated by the central AGN. Finally, we find that the central elemental abundance in the X-ray gas is comparable to that measured optically. ",
        "introduction": "M84 (NGC4374) is an E1 galaxy within the core of the Virgo cluster.  Radio observations at 1.4 and 4.9 GHz show two lobes and a jet (Laing \\& Bridle 1987).  In X-rays, the galaxy has a compact gas halo, rather low metallicity and an overabundance of alpha-process elements (e.g. Finoguenov \\& Jones 2000).  In this {\\it Letter} we discuss primarily the comparison of the X-ray and radio morphologies. We also determine the temperature and abundance structures in the galaxy center, which allows a direct comparison with optical abundance measurements. Finally, we resolve the nature of the hard X-ray component, found by ASCA in many ellipticals (Matsumoto \\etal 1997). ",
        "conclusions": "The high resolution Chandra image shows the remarkable interaction of the radio lobes and the diffuse soft X-ray emission. The expansion of the radio lobes appears to have created cavities in the soft emission that are surrounded by higher density shells. Based on the similarities in the temperature and abundance of the filaments around the radio lobes and the gas in the central region, we suggest that this gas has a common origin and circulates throughout the central region of the galaxy. Chandra's high resolution also has led to the detection of a population of luminous ($>10^{38}$ ergs/sec) galactic X-ray sources and the discovery of a hard, diffuse source in the central 10 kpc. We suggest that the AGN at the center of M84 heats the gas in this region. Future observations of M84 and other galaxies can test this prediction. Finally, Chandra's resolution allowed abundance measurements to be made on a spatial scale similar to those done optically, resolving the issue of low reported abundances for the X-ray gas, compared to optical studies.  \\vspace*{0.2cm}  The authors are thankful to Alexey Vikhlinin and Maxim Markevitch for their help in Chandra data reduction, and to Dan Harris, Yasushi Fukazawa, Brian McNamara and the referee for useful comments on the manuscript. This work was supported by NASA grants GO0-1045X and AG5-3064 and the Smithsonian Institution."
    },
    "0010/astro-ph0010121_arXiv.txt": {
        "abstract": "\\vskip-6.1in \\begin{flushright} UMN-TH-1924/00 \\\\ TPI-MINN-00/48 \\\\ CERN-TH/2000-288\\\\ astro-ph/0010121 \\\\ October 2000 \\end{flushright} \\vskip+5.0in  The recent observations of an approximately linear relationship between both Be and B and iron in  metal poor stars has led to a reassessment of the origin of the light elements in the early Galaxy. In addition to standard secondary production of BeB, it is necessary to introduce a production mechanism which is independent of the interstellar metallicity (primary), and in which freshly synthesized C,O and He are accelerated by supernova shock waves.  Primary mechanisms are expected to be dominant at low metallicity. At metallicities higher than O/H $\\ga -1.75$, existing data might indicate that secondary production is dominant. In this paper, we focus on the secondary process, related to the standard galactic cosmic rays, and we examine the cosmic ray energy requirements for both present and past epochs. We find the power input to maintain the present-day Galactic cosmic ray flux is about $\\ee{1.5}{41} \\ {\\rm erg/s} = \\ee{5}{50} \\ {\\rm erg/century}$; this estimate includes energy losses from both the escape of high-energy particle and ionization losses from low-energy particles. This implies that, if supernovae are the sites of cosmic ray acceleration, the fraction of explosion energy going to accelerated particles is about $\\sim 30\\%$, a value which we obtain consistently both from considering the present cosmic ray flux and confinement and from the present \\be9 and \\li6 abundances.  Using the abundances of \\be9 (and \\li6) in metal-poor halo stars, we extend the analysis to show the effect of the interstellar gas mass on the standard galactic cosmic ray energetic constraints on models of Li, Be, and B evolution. The efficiency of the beryllium production per erg may be enhanced in the past by a factor of about 10; thus the energetic requirement by itself cannot be used to rule out a secondary origin of light elements. Only a clear and undisputable observational determination of the O-Fe relation in the halo will discriminate between the two processes. ",
        "introduction": "Energetic considerations have historically provided a fundamental constraint on and probe of the origin of cosmic rays. For example, energetic constraints have been used by many authors (see, e.g., the monograph by Brerezinski\\u{\\i} \\etal \\pcite{bbdgp} and references therein) to point out that supernova explosions are a plausible cosmic ray power source, while other stars are not. An accounting of the energy budget have been particularly useful in the case of Li, Be, and B (LiBeB) production. Ryter, Reeves, Gradsztajn, \\& Audouze \\pcite{rrga} used this type of argument to rule out {\\it in situ} LiBeB production in young stars by flare particles (the ``autogenetic'' hypothesis), in favor of the present, ``galactogenetic,'' hypothesis of LiBeB production by galactic cosmic rays in the interstellar medium (Reeves, Fowler, \\& Hoyle \\cite{rfh}).  Recently, an energetics argument has been invoked to discriminate between two possibilities, namely, a primary or a secondary origin of light elements in the halo (Ramaty \\etal \\cite{rklr}). The appearance of a linear correlation between BeB and Fe in metal poor stars led to the proposal of primary mechanisms (different from standard GCRN) for LiBeB production, in which C, O nuclei ejected and accelerated by supernovae fragment on H and He at rest in the interstellar medium (e.g., Cass\\'e \\etal   \\cite{cac}, Ramaty \\etal \\cite{rkl} Vangioni-Flam et al.\\ \\cite{vroc}; Higdon \\etal \\cite{hlr}; Ramaty \\& Lingenfelter \\cite{rl}).  Given the approximately linear correlation between [Be/H] and [Fe/H], and a mean value of Be/Fe of about $10^{-6}$, one can conclude that if 0.1 \\msol\\ of iron is produced per supernova, then about 2 $\\times 10^{48}$ atoms of Be are produced per supernova.  Standard secondary models of Be production yield an energy requirement of about $\\sim 5\\times 10^{3}$ ergs per Be atom at $[{\\rm Fe/H}] = -2.5$ or about $10^{52}$ ergs per supernova going into the production of Be.  Of course this number is in excess of the total energy available in the supernova explosion. This is the basis of the energetics problem for secondary production and will be the main focus of this paper.  Note that the progenitor of BeB is not Fe but O (and C). Thus the pertinent correlation is not between BeB and Fe but between BeB and O. If [O/Fe] is constant then the linearity between BeB and Fe translates directly into a linear relation between BeB and O. In this case a primary mechanism is clearly required.  However, recent observations (Israelian \\etal \\cite{igr}, Boesgaard \\& King \\cite{bk}, Takeda \\etal \\cite{tak}),  indicate that [O/H] increases with decreasing [Fe/H]. In this case the linearity between  BeB and Fe leads to the correlation: log(Be/H) proportional to about 1.7 [O/H], and should be explained by a combination of primary and secondary mechanisms (Fields and Olive 1999a; Fields \\etal 2000, hereafter FOVC). While the new O/Fe results are increasing in number they are still controversial. An unambiguous determination of the true O/Fe trend in Pop II is crucial, and the results of the present study will only further underscore the need to resolve this issue.  It has been argued (Ramaty \\etal \\cite{rklr}, Ramaty, Scully, Lingenfelter, \\& Kozlovsky \\cite{rslk}, Parizot \\cite{par}) that energetic arguments can be used to clarify the primary vs secondary source for BeB. These authors have concluded that the secondary process in the early Galaxy meet with strong energetic difficulties. However, the energetics computation relies on many relatively uncertain factors and deserves a careful analysis (including the sensitivity to the injection energy spectrum to test the sensitivity of this criterion). Here, we will consider the energetics of {\\em standard Galactic} cosmic rays in both the early and present Galaxy.  As we will see, a careful treatment of energetics allows us (1) to clarify and make precise the manner in which standard cosmic rays are to be included into chemical evolution, and then (2) to examine the viability of proposed cosmic ray origins. We begin by establishing (in \\S \\ref{sect:gcr}) the present-day cosmic ray power budget, based on the observed properties of Galactic cosmic rays and of the Milky Way.  By explicitly identifying supernova remnants as the power sources for the cosmic rays (in \\S \\ref{sect:sne}), we can express the present day energetics in terms of a cosmic ray acceleration efficiency per supernova (in \\S \\ref{sect:cr_eff}). We find that this efficiency is subject to considerable uncertainties, due to uncertainties in the parameters describing cosmic rays, the Galaxy, and supernova blast energies. Nevertheless, the present efficiency must serve as a fiducial point against which we compare the past energetics as determined by Be (\\S \\ref{sect:be_eff}) and \\li6 (\\S \\ref{sect:li6}). In calculating the energetics of \\li6 and Be, we note the importance of the larger gas mass in the early Galaxy relative to today, which implies a larger total number of atoms of all kinds in the interstellar medium. This in turn implies a larger number of target atoms (for a given metallicity) and thus a larger Be production rate, than one would estimate using the present day Galactic gas mass; the upshot is that one finds a higher Be production efficiency by a factor of $M_{\\rm gas,init}/M_{\\rm gas,today} \\sim 10$ when one takes into account this gas-depletion effect (heretofore not considered in previous work). With this factor included, we find that the cosmic ray energetics implied by the Be data is in good agreement with present-day values.  Indeed, we find that the energetics constraint is completely correlated with the basic constraint of agreement with Be-Fe data:  models which fit the data also do not require more energy efficiency in the past. We discuss the implications of our results in \\S \\ref{sect:conclude}.  Thus, the existence of an energetics problem will ultimately rest with new observations of Be at low metallicity.  A dispersion of points above the secondary production predictions would indicate the need for a primary source of Be (perhaps occurring in spatially localized regions) to explain the high points. Barring that, the only true discriminant between primary and secondary production of Be at this time, will be a final and consistent determination of the O/Fe ratio at low metallicity. ",
        "conclusions": "\\label{sect:conclude}   In this paper, we have reviewed the past and present energy budget for standard Galactic cosmic rays We determine present-day cosmic ray power requirements from the observed cosmic ray flux and confinement. Comparing this to the fiducial $10^{51}$ erg of kinetic energy in a supernova blast, we find the efficiency of cosmic ray acceleration is about 30\\%. This value cannot be calculated to a precision better than a factor of about 3, given the uncertainties in the cosmic ray and Galactic input parameters. Nevertheless, we find that this value is consistent with estimates derived from present (solar) \\be9 and \\li6 values.  We have extended our analysis to the study the cosmic ray acceleration efficiency in the early Galaxy, as encoded in the Be abundances of metal-poor stars. We find that energy considerations are indeed powerful, and allow us to {\\em derive} the usual $\\phi \\propto \\psi$ scaling. We also find that in general, the requirement that models satisfy energetic requirements is equivalent to the requirement that the models produce LiBeB evolution which fits the abundances data, if SN are cosmic ray acceleration engines.  Furthermore, constructing a careful and consistent model for the energetics of cosmic ray acceleration and LiBeB production has an additional benefit: it brings to the surface key underlying physical issues and assumptions regarding early Galactic LiBeB production. Specifically, we have emphasized that the ISM gas content, and thus the total number of ISM atoms, was higher in the early Galaxy.  Thus, for a fixed metallicity, the total Galactic Be production is enhanced due to the increased total number of targets.  This effect has been neglected up to now but can play a crucial role in the energetics: due to this effect, the number of Be atoms produced per cosmic ray erg should be multiplied by about a factor 10 in the early Galaxy compared to the estimates lacking this effect.  Consequently, the energy test is less stringent than previously thought.  Note that one could invoke still more complications in the form of early Galactic evolution of cosmic ray confinement properties, or a lack of equilibrium between cosmic ray sources and sinks; further study of these issues is best done in a consistent model of cosmic ray acceleration and propagation in a Galaxy which evolves in time and space.  In the absence of such a model, the current ``leaky box'' picture that has been adopted for Be studies, with constant cosmic ray confinement parameters, leads to the factor of $\\sim 10$ increase in Be production efficiency we have described.  With this effect, the standard GCR model is able both to fit the Be-Fe data, and to maintain a reasonable (and roughly constant) acceleration efficiency over the history of the Galaxy at least down to [Fe/H] $\\simeq - 2.5$.  Thus we find that energetic arguments are extremely useful tools in analyzing LiBeB production in the early Galaxy, and thereby to examine questions of cosmic ray origin.  However, contrary to some analyses, we find that the standard GCR scenario does not suffer energetics problems, and thus that energetics alone cannot rule this model out. Therefore, the key of the origin of early Galactic cosmic rays and LiBeB (i.e., a primary or secondary process?) remains in the hands of the observers. Specifically, it is crucial to reconcile the different measurements of oxygen (and thus BeB-O and O-Fe) in halo stars. The determination of reliable oxygen abundances is to be ranked as a first priority of the abundance observation programs.  Indeed, looking towards the future, we can hope to turn the problem around. With reliable and accurate oxygen data, we can establish the origin of LiBeB  and cosmic rays in the early Galaxy. With this knowledge in hand, we can then use the detailed LiBeB abundance patterns and scatter to learn more about the physical conditions of gas and accelerated particles in the early Galaxy."
    },
    "0010/astro-ph0010317_arXiv.txt": {
        "abstract": "Near-Infrared high angular resolution speckle imaging and imaging spectroscopy of the nuclear region ($\\sim$ 10'' $\\approx$ 840~pc) of the Seyfert~1 galaxy NGC~3227 are presented. The images reveal an unresolved nuclear source in the $K$ band in addition to a nuclear stellar cluster which is slightly resolved in the $J$ and $H$ band. The contribution of this stellar cluster to the NIR continuum is increasing from the $K$ to the $J$ band. The stellar absorption lines are extended compared to the neighboring continuum suggesting a nuclear stellar cluster size of $\\sim$ 70~pc FWHM. Analysis of the stellar absorption lines suggests that the stars are contributing about 65\\% (40\\%) of the total continuum emission in the $H$ ($K$) band in a 3.6'' aperture. The dominant stellar type are cool M type stars. Population synthesis in conjunction with NIR spectral synthesis indicates that the age of the mapped nuclear stellar cluster is in the range of 25 to 50~Myr when red supergiants contribute most to the NIR light. This is supported by published optical data on the Mg I $b$ line and the CaII triplet. Although a higher age of $\\sim$ 0.5 Gyr where AGB stars dominate the NIR light can not be excluded, the observed parameters are at the limit of those expected for a cluster dominated by AGB stars. However, in either case the resolved stellar cluster contributes only about $\\sim$ 15 \\% of the total dynamical mass in the inner 300~pc. This immediately implies at least another much older stellar population which contributes to the mass but not the NIR luminosity. Pure constant star formation over the last 10 Gyr can be excluded based on the observational fact that in such a scenario the total observed (spatially unresolved and spectrally resolved) Br$\\gamma$ flux would be of stellar origin which is spatially extended. Therefore, at least two star formation/starburst events took place in the nucleus of NGC~3227. Since such sequences in the nuclear star formation history are also observed in the nuclei of other galaxies a link between the activity of the star formation and the AGN itself seems likely. ",
        "introduction": "INTRODUCTION}  The nuclear stellar properties of Seyfert galaxies are of great interest, since there might be a connection between the nuclear star formation and the AGN activity (Norman \\& Scoville 1988). Current scenarios for fueling the nuclear region involve nuclear bars (e.g. Shlosman, Frank \\& Begelman 1989). However, recent high-angular resolution observations of Seyfert galaxies do not find enough evidence for this model (Regan \\& Mulchaey 1999, Malkan, Gorjian \\& Tam 1998). The nuclear star formation activity and the AGN both use the same fuel -- the circumnuclear molecular gas. Detailed analysis of the nuclear stellar properties of the two Seyfert galaxies Circinus (Maiolino et al. 1998) and NGC~1068 (Thatte et al. 1997) have shown that the nuclear stellar clusters have age of $\\sim$ 10$^8$ yr. This suggests that at least some of the molecular gas reaching the inner few 100~pc is converted into stars rather than being funneled directly to the AGN. In addition there might be an interplay between the star formation and the AGN classification (Ohsuga \\& Umemura 1999). \\\\ \\\\ High angular resolution observations in the near-infrared (NIR) of the nuclear region of active galaxies are well-suited to investigate this problem via a quantitative analysis of the contributions of the AGN and the stars to the nuclear NIR continuum. The NIR offers two advantages over the optical wavelength ranges: (1) the extinction affecting the stellar light is much smaller (about 1/10 in the $K$ band compared to the $V$ band) and (2) the overall stellar contribution to the continuum emission is larger due to the spectral energy distribution (SED) of the AGN and the stars (e.g. Barvainis 1989). This allows for a more detailed analysis of the properties of the stellar contribution to the nuclear light. The combination of imaging spectroscopy with high angular resolution imaging is ideal to disentangle the contribution of the AGN from those of the stars using spectral and spatial information. The utilization of population synthesis and NIR spectral synthesis models allows to analyze the star formation history of the nuclear stellar component. NGC~3227, a nearby Seyfert~1 galaxy, was chosen for such a detailed study, since complementary data from other wavelengths are already available in the literature. \\\\ \\\\ NGC3227 (Arp~94b; type SAB pec, de Vaucouleours et al. 1993) is a Seyfert 1 galaxy located at a distance of 17.3~Mpc (group distance; Garcia 1993; 1'' $\\sim$ 84~pc) and interacting with its elliptical companion NGC~3226. Rubin \\& Ford (1968) studied the system NGC~3226/7 for the first time in the optical. In NGC~3227 they found indications for a nuclear outflow as well as a spiral arm that stretches toward the close elliptical companion. The nucleus of NGC~3227 exhibits clear signs of Seyfert~1 activity. The Narrow Line Region (NLR) is extended towards the northeast (Mundell et al. 1992a, Schmitt \\& Kinney 1996). The Broad Line Region (BLR) clearly shows variations in the optical continuum and line emission (Salamanca et al. (1994). Mundell et al. (1995b) detected a double radio source with a separation of 0.4'' at a PA of $\\sim$10$^o$. The cold molecular gas traced by its \\CO ~line emission shows a circumnuclear ring with a diameter of $\\sim$ 3'' plus weaker emission at the dynamical center. In the inner arcsecond strong non-circular motions are observed. Modeling of the kinematics indicates that a warped thin molecular gas disk can explain the observed kinematics quite well (Schinnerer, Eckart \\& Tacconi 1999, 2000). Optical and NIR observations (Mulchaey, Regan, Kundu 1997, De Robertis et al. 1998) indicate that the galaxy has a bar with R $\\approx$ 6.7 - 8.4~kpc and a position angle of about -20$^o$ relative to the major kinematic axis. \\\\ \\\\ In this paper we present the first quantitative analysis of the AGN and stellar contributions to the nuclear NIR continuum of NGC~3227 using high angular NIR imaging and integral field spectroscopy. We describe our observations in \\S \\ref{ww02}, and discuss the results from the integral field spectroscopy in \\S \\ref{ww04}. The nuclear stellar content is derived in \\S \\ref{xxs55}. We present an analysis of the high angular resolution imaging data in \\S \\ref{ww03}. Population synthesis and NIR spectral synthesis models are used in \\S \\ref{xxs7} to analyze the nuclear star formation history of NGC~3227. A description and discussion of these models is given in the appendix. A brief summary is presented in \\S \\ref{ww08}. ",
        "conclusions": ""
    },
    "0010/astro-ph0010584_arXiv.txt": {
        "abstract": "The failure of purely gravitational and gas dynamical models of \\xray cluster formation to reproduce basic observed properties of the local cluster population suggests the need for one or more additional physical processes operating on the intracluster medium (ICM).  We present results from 84 moderate resolution gas dynamic simulations designed to systematically investigate the effects of preheating --- an early elevated ICM adiabat --- on the resultant, local \\xray size--temperature, luminosity--temperature and ICM mass--temperature relations. Seven sets of twelve simulations are performed for a $\\Lambda$CDM cosmology, each set characterized by a different initial entropy level $S_i$. The slopes of the observable relations steepen monotonically as $S_i$ is increased.  Observed slopes for all three relations are reproduced by models with $S_i \\in 55-150 \\kevcmsq$, levels that compare favorably to empirical determinations of core ICM entropy by Lloyd-Davies, Ponman \\& Cannon. The redshift evolution for the case of a locally successful model with $S_i \\se 106 \\kevcmsq$ is presented.  At temperatures $kT\\gta 3$~keV, little or no evolution in physical isophotal sizes or bolometric luminosities is expected to $z \\lta 1$.  The ICM mass and total masses at fixed $T$ are lower at higher $z$ as expected from the virial theorm. ICM mass fractions show a mild $T$ dependence. Clusters with $T \\lta 3$~keV contain ICM mass fractions depressed by modest amounts ($\\lta 25\\%$) below the cosmic mean baryon fraction $\\Omega_b/\\Omega_m$; hot clusters subject to preheating remain good tracers of the cosmic mix of clustered mass components at redshifts $z \\lta 1$. ",
        "introduction": "The hot, \\xray emitting plasma known as the intracluster medium (ICM) represents a part of the baryonic matter of the universe that is not associated with individual galaxies but remains trapped in the deeper gravitational potential of galaxy clusters. Models of cluster formation in which the intergalactic gas simply falls into the dark matter-dominated gravitational well (so-called infall models) fail to reproduce all the structural properties of the local cluster population (\\eg Evrard \\& Henry 1991; Navarro, Frenk \\& White 1995; Mohr \\& Evrard 1997; Bryan \\& Norman 1998).  There appears to be additional physics driving ICM evolution.  As early \\textit{ROSAT} and \\textit{Einstein} data emerged, several researchers proposed that the missing element is the existence of a high entropy intergalactic gas prior to a cluster's collapse (David \\etal 1991; Evrard \\& Henry 1991; Kaiser 1991; White 1991).  The entropy floor preduced by the preheated gas breaks the self-similarity between the dark matter and ICM on different mass scales.  This occurs because the equivalent thermal energy of the entropy floor corresponds to a larger fraction of the virial temperature in lower mass clusters.  A physical scenario consistent with this picture is one in which the bulk of star formation in a proto-cluster region occurs early in its formation history.  Heating from supernova-driven galactic winds and AGN activity associated with the rapid star formation ultimately comes to exceed the cooling rate of ambient material, resulting in a period of net local heating.  In the low-density limit, for which subsequent cooling is unimportant, this feedback defines an initial adiabat $S_i$ that is conserved in the absence of shocks. Later shock heating can raise the adiabat above this ``entropy floor,'' but the entropy cannot decrease below this level unless cooling becomes important.   Observational lines of evidence provide at least partial support for such a picture.  Measurements of ICM metal--abundances and their gradients (\\eg Loewenstein \\& Mushotzky 1996) are likely explained by feedback from star formation.  Most of the stars in cluster elliptical galaxies are of a nearly uniform old age (Bower, Lucey \\& Ellis 1992; Kuntschner 2000, but see Trager etal 2000) and late-type galaxies within clusters are typically gas-poor and forming stars at a reduced rate compared to field galaxies (Cayette \\etal 1994, Chamaraux, Balkowski \\& G\\'{e}rard 1980). Even in MS1054, a very distant cluster at $z \\se 0.83$, much of the visible galaxy mass is contained in red, early-type galaxies (Van Dokkum \\etal 1999). An active period of star formation in at least some regions of the $z \\ssim 3$ universe is inferred from the ultraluminous sources detected by SCUBA (Barger, Cowie \\& Richards 2000), and the population of Lyman-break galaxies, the distant equivalent of the local normal population, is forming stars at a rate substantially higher than today (\\seg, Steidel \\etal 1999).  In addition to winds driven by star formation, the jets of radio galaxies present another potential source of ICM heating.  \\chandra\\ observations of low emission measure bubbles in the ICM, aligned with the radio halos of central cluster sources (McNamara \\etal 2000), implicate radio jets as another source of additional ICM entropy.  The fact that the radio galaxies appear more frequently in clusters at higher redshift (Owen 2000) suggests that heating from this population might be more important in the past.  Of course, these and other observations also indicate that some late star formation and feedback does occur in cluster environments and this fact ultimately limits the accuracy and applicability of the approach used here. Also, the statistics of apparent gaps in the Ly-$\\alpha$ forest of QSO spectra can be used to place limits on the volume fraction and epoch of heating to temperatures well above $10^4$ K (Theuns, Mo \\& Schaye 2000).  A principal benefit of a model where clusters form through the infall of preheated gas is that the physical treatment of ICM evolution through gravitationally-driven shock heating is familiar territory for cosmological gas dynamic codes.  Comparing the gravitational evolution of a single \\xray cluster derived from twelve nearly-independent codes, Frenk \\etal (1999) find agreement in structural properties at characteristic levels of $\\ssim 5\\%$ for dark matter and $\\ssim 10\\%$ for the ICM.  The price for operating in a regime where the numerical accuracy has been calibrated is an approximate physical treatment.  Neglect of radiative cooling is a particular concern, but the effect of radiative cooling and subsequent star formation on a cluster's thermal history is a complex problem.  Observations indicate that ICM cooling occurs frequently in clusters, but the phenomenon is restricted to a core region containing $\\ssim 1\\%$ of the collapsed (virial) cluster mass.  However, recent spectroscopic studies which resolve cooling flows (David \\etal 2000, Tamura \\etal 2000) indicate that the phenomenon is confined to an even smaller volume than previously thought.  At early times, cooling is an important ingredient of the galaxy formation process (Rees \\& Ostriker 1977; White \\& Rees 1978).  So important, in fact, that sources of heating must be introduced to stabilize its effects on small scales (White \\& Frenk 1991; Cole 1991; Blanchard, Valls-Gabaud \\& Mamon 1992).  Attempts to model the full problem of the galaxy formation process within clusters are limited by poor knowledge of the correct parameterizations for star formation and feedback at arbitrary epochs.  Semi-analytic methods (Kauffman, White \\& Guiderdoni 1993;  Baugh, Cole \\& Frenk 1996; Wu, Fabian \\& Nulsen 1998, 1999; Somerville \\& Primack 1999) are best able to explore the large parameter space associated with these processes, but the lack of a direct solution for the spatial distribution of material limits the ability of this approach to predict detailed structural observables.  Bryan (2000) has presented a model which shows that the elimination of low--entropy gas reproduces the luminosity--temperature scaling relation.   \\setcounter{footnote}{0}  An initial entropy excess prevents gas from falling into the dark matter--dominated potential well to the extent it would in the purely gravitational infall model, thus reducing the central gas density.  Core entropy\\footnote{In this paper, an equivalent entropy $S$ is expressed in units $\\kevcmsq$.  The usual thermodynamic definition of entropy is $s~=~c_V~ln(P/\\rho^\\gamma)$, where $\\gamma~=~5/3$ for a monotonic gas (appropriate for a fully ionized plasma), relates to the entropy of this paper by $S~=~e^{(s/c_V)}/R$, assuming an ideal gas equation of state $P~=~R~\\rho~T$.} is related to temperature and density by $S_{core} \\sim T/{\\rho_{core}}^{2/3}$.  If the temperature is set by the collisionless dark matter through the virial theorem $T \\spropto M_{200}^{2/3}$ ($M_{200}$ is the mass enclosed in a sphere defined as containing a density contrast of $200$ with respect to the critical density), then the core density of initially preheated material will satisfy \\begin{equation} \\rho_{core} ~ \\lta ~ M_{200} \\ / \\ S_i^{3/2} . \\label{rhocore} \\end{equation} Higher levels of preheating will produce lower core densities and, for a given preheating amplitude $S_i$, low mass clusters will have less dense cores than high mass clusters.  This differential effect results in a steepening of the slopes of the relations between density-dependent cluster observables (isophotal--size, luminosity and ICM mass) and temperature.   Previous simulations have found preheating to produce more realistic clusters in terms of the luminosity--temperature relation (\\eg Narvarro, Frenk \\& White 1995; Pierre, Bryan \\& Gastaud 1999). Metzler (1995) demonstrated a similar result using simulations that explicitly included the feedback of energy and mass from cluster galaxies.  We seek here to systematically investigate preheating's effect on the size--temperature, luminosity--temperature, and ICM mass--temperature relations.  Although not entirely independent observables, the functional dependence on the ICM density varies for these measures.  Each probes a differently-weighted moment of the radial density profile.  There is no guarantee, therefore, that a single preheating level $S_i$ will provide a simultaneous match to all three relations.   We use a total of 84 moderate resolution numerical experiments consisting of twelve initial configurations run at six different levels of preheating plus a set evolved without preheating for comparison.  Since we lack a detailed mechanism for the preheating, all simulations are initiated with the gas held at a fixed, elevated temperature that corresponds to adiabatically evolved temperatures of $\\sim 10^6 K$ at $z \\ssimeq 3$.   In Section \\ref{sec:tech}, the simulations are detailed.  The effect of preheating on the major cluster relations are presented in Section \\ref{sec:scal}.  In Section \\ref{sec:predict}, the range of entropies allowed by observation are determined and compared to other preheating studies.  The effects on evolution, the virial mass--temperature relation, and gas fractions are explored in Section \\ref{sec:consequence}. ",
        "conclusions": "We have used a set of 84 moderate resolution gas dynamic simulations of cluster evolution to systematically investigate the effects of preheating on the local \\st, \\lxt and \\mt relations.  We use a plasma emission model to estimate luminosities and determine spectral temperatures from fits to the overall cluster emission. We find that preheated models with initial entropy in the range $55-150 \\kevcmsq$ reproduce the slopes of the observed scaling relations.  There is agreement between the observations of Lloyd-Davies \\etal (2000) and these simulations in the allowed range of initial entropy.  Although the range is currently generous, the new generation of \\xray satellites should provide information to further constrain the allowed range or potentially falsify the model.  Better understanding of the ICM entropy aids in the deduction of its history and places restrictions on the mechanisms that may be responsible for preheating.  Examining evolutionary effects in a set with $S_i \\se 106 \\kevcmsq$, the models predict modest steepening in the three scaling relations at $z = 1.0$.  The predictions are currently in agreement with  \\footnotesize \\begin{center} {\\sc Table 5\\\\ Fits$^a$ to $h(z) M_{200}$ \\vs spectral temperature for set S3 } \\vskip 4pt \\begin{tabular}{ccc} \\hline \\hline z & m & b \\\\ \\hline 0.0 & 1.66(0.12) & 14.98(0.04) \\\\ 0.5 & 1.73(0.11) & 14.95(0.04) \\\\ 1.0 & 1.91(0.11) & 14.96(0.04) \\\\ \\hline \\end{tabular} \\centerline{$^a$ $\\log(h(z)M_{200})~=~m \\log(T_s/6 \\kev) + b$.} \\end{center} \\setcounter{table}{5} \\normalsize \\centerline{}  \\myputfigure{figure12.epsi}{2.8}{0.45}{-10}{0} \\figcaption{ Redshift-corrected total mass within $r_{200}$ \\vs spectral temperature for clusters in set S3. \\label{virplot}} \\vskip10pt  \\myputfigure{figure13.epsi}{2.8}{0.45}{-10}{0} \\figcaption{ICM mass as a fraction of total mass \\vs spectral temperature inside $r_{200}$ for clusters of model S3 at $z$ = 0.0, 0.5 and 1.0.  The global value is $\\Omega_b/\\Omega_m = 0.1$. \\label{gasplot}} \\vskip10pt  \\noindent the observed lack of evolution in the \\lxt relation to $z \\ssim 0.5$.  Deeper observations (\\eg, Stanford \\etal 2000) from Chandra and XMM-Newton should help reveal the epoch where the scalings depart from the simulation's expectations."
    },
    "0010/astro-ph0010067_arXiv.txt": {
        "abstract": "The present study of spectra of twelve metal-rich cool dwarf stars, carefully selected in order to cover a range of temperatures ($\\sim 4400 - 6000$ K), is a follow up on Feltzing \\& Gustafsson (1998)\\nocite{feltzing} with the aim to understand the apparent over-ionization and anomalous elemental abundances found by them for the  K dwarf stars in their sample.  Our method of analysis employs synthetic spectra of the full spectrum both to constrain the continuum level and to derive abundances. It is shown that by using this method and  imposing a strict excitation equilibrium (possible to do because of the  care in selection of observed Fe {\\sc i} lines) we are able to show that metal-rich K dwarf  stars do not show anomalous stellar abundances, as indicated in Feltzing \\& Gustafsson (1998), and can, with reasonable efforts, be analyzed in order to increase the number of metal-rich stars with useful chemical abundances.  With abundance analysis  by means of spectrum synthesis and assuming Local Thermodynamic Equilibrium (LTE) the abundances of Na, Si, Ca, Sc, Ti, V, Cr, Fe, Co, Ni, and Nd have been derived.  Also ionization balance is satisfied for Fe and Cr after correcting the stellar effective temperatures such that both ionization and excitation equilibrium were satisfied.  In addition, spectra from five cool dwarf stars of the Feltzing \\& Gustafsson (1998) sample have been analyzed with the methods used in this work. They show essentially the same  abundance patterns as the new stars in this sample. ",
        "introduction": "The atmospheres of dwarf stars retain, except for a small number of elements, the chemical composition of the gas they were formed out of. This makes them ideal as tracers of the galactic chemical evolution. Combining chemical and kinematical data  is a powerful method in studies of the galactic chemical evolution.  The last decade have seen several studies of stellar abundances of large numbers of dwarf stars with known kinematics and derived ages, eg.  Edvardsson et al. (1993) (hereafter EAGLNT93)\\nocite{eaglnt:93}, Feltzing \\& Gustafsson (1998) (FG98), Fuhrmann (1998)\\nocite{fuhrmann} and Chen et al. (2000)\\nocite{chen00}.  These studies have mainly been concerned with the warmer F and G dwarf stars and the general trends of the different elements are now well established and their gross features can be  reproduced, for most elements, by models of galactic chemical evolution, eg.  Samland (1998)\\nocite{samland} and Timmes et al. (1995)\\nocite{timmesetal}. Well defined abundance trends for stars more metal rich than the Sun can also provide vital clues to the production sites of several elements (see Samland, 1998).  The evolution above ${\\rm [Fe/H]}\\sim 0.1$ dex is still, however, fairly poorly sampled and several elements show large scatter. One reason for this is the relative rarity of the most metal-rich stars. FG98 studied a sample of 47 F, G and K dwarf stars with metallicities above solar. In general they confirmed the trends found in EAGLNT93, now also confirmed in Chen et al. (2000). However, for the cool K dwarf stars in their sample FG98 found both an apparent over-ionization as well as odd abundances for several elements. \\thorenletter\\nocite{thoren00calletter} \\ has shown that the odd Ca abundances were not, as originally thought, due to NLTE effects (see also Drake, 1991), but wrongly calculated damping parameters. The other effects, though, remain to be explained.  Sect. \\ref{sect:obs} describes the observations. Sect. \\ref{sect:red} reports on the data reduction and analysis. In sect. \\ref{sect:abund} the stellar abundances are presented. This is followed by a discussion in sect. \\ref{sect:disc} and conclusion in sect. \\ref{sect:conclusion}. ",
        "conclusions": "\\label{sect:conclusion}  The investigation by FG98 left several unexpected results and unanswered questions in particular about elemental abundances in K dwarf stars. Applying new methods, tailored to suite the K dwarf stars, in our abundance analysis we have been able to answer a number of these questions. In particular we find that:  - by imposing excitation balance for Fe {\\sc i} lines we naturally arrive close to LTE ionization balance for Fe and Cr  - stellar abundance results for cool K dwarf stars are more or less the same as for their warmer counter-parts  - we show that the stellar atmosphere models of the coolest stars need temperature corrections as compared to the calibrations presented by \\ols. This may indicate a systematic error in the calibration for the stars with colours $b-y>0.51$  - according to the photometric calibration for the metallicity \\cite{olsen:84}, HD 42182 is a Super-Metal-Rich (SMR, [Me/H]$>$0.20, Taylor, 1996\\nocite{taylor96}) candidate star. This is not confirmed by our data. Also, HD 21197, which has just above solar metallicity from the photometry is showed to be a SMR star, as well as HD 32147  - the two most metal-rich stars in our sample show extremely similar abundances, which makes it possible to draw strong conclusions on the stellar abundance trends  - [Si/Fe], [Ca/Fe], [Ti/Fe] and [Cr/Fe] all vary as [Fe/H] also for stars more metal-rich than 0.2 dex  - [Na/Fe] (and possibly [V/Fe], [Co/Fe] and [Ni/Fe]) increases with increasing [Fe/H].  \\begin{acknowledgement} The authors thank Docent Bengt Edvardsson for valuable discussions and comments on the manuscript. The referee is thanked for suggestions that improved the quality of the article. PT was supported by the Swedish Natural Science Research Council, NFR. \\end{acknowledgement}"
    },
    "0010/astro-ph0010298_arXiv.txt": {
        "abstract": "I cross-correlate the \\OG\\ database of Cepheids in the Large and Small Magellanic Clouds  (MCs) with the second incremental release of the \\M\\ $JHK_{\\rm s}$ survey, and the {\\sc denis} $IJK_{\\rm s}$ point source catalog towards the MCs.  Of the 3384 \\C\\ in \\OG\\ in both Clouds, 1745 have a counterpart in the \\M\\ survey within an 1\\arcsec\\ radius and good-photometry in all three bands. Only 173 have a counterpart in the \\DE\\ survey within an 1\\arcsec\\ radius and good-photometry in all three bands. The reason for this difference is that the limiting magnitudes of the \\M\\ survey are considerably fainter than for the \\DE\\ survey.  The standard stars of the Carter IR photometric system are also correlated with the \\M\\ survey to derive transformation equations from the natural \\M\\ system to the Carter-system.  In order to describe to first order the geometrical configuration of the MCs a plane is fitted to the reddening-free Wesenheit index ($W = I - 1.55 \\times (V-I)$) and the inclination and position angle of the line-of-nodes are determined. For the LMC an inclination angle of 18 $\\pm$ 3\\degr\\ is derived, somewhat smaller that previous determinations. For the SMC a value of $i = 68 \\pm 2$\\degr\\ is derived in good agreement with previous determinations. These results are being used to take out the effect that some \\C\\ are closer to the observer than others. For the LMC this effect is small over the area covered by the \\OG\\ survey, but for the SMC $PL$-relations with a reduced scatter are derived in this way. Nevertheless, the dispersion in the $PL$-relation in $W$ for the SMC is larger than that for the LMC, indicating a significant intrinsic depth of the SMC of about 14 kpc (front-to-back). The depth of the LMC is found to be small compared to the intrinsic scatter in the $PL$-relation.  Using single-epoch \\M\\ data for the \\OG\\ \\C\\, $PL$-relations are fitted in $JHK$ (on the Carter system) for both Fundamental mode (FU) and First-overtone (FO) pulsators, taking out the effect of the inclined disk.  Because of the large number of stars available, this results in the most accurately determined slopes in the infrared up to now, and slopes for FO pulsators derived for the first time. This provides additional constraints for theoretical models.  The 16 derived $PL$-relations (in $WJHK$ for FU and FO pulsators for LMC and SMC) are used to derive a relative distance modulus ${\\Delta}_{\\rm SMC-LMC} = 0.50 \\pm 0.02$ assuming no metallicity correction. Two recent theoretical models predict different metallicity corrections, indicating a systematic uncertainty of up to 0.1 mag.  Using the database of Galactic \\C\\ observed with \\HP\\ and the method outlined in Groenewegen \\& Oudmaijer (2000), the zero points of Galactic $PL$-relations for FU pulsators in $W$ and $K$ are derived, using the newly derived slopes for LMC and SMC pulsators.  Combining the zero points of the Galactic and MCs $PL$-relations distance moduli of 18.60 $\\pm$ 0.11 (based on $W$) and 18.55 $\\pm$ 0.17 (based on $K$) to the LMC, and of 19.11 $\\pm$ 0.11 ($W$) and 19.04 $\\pm$ 0.17 ($K$) to the SMC are derived (without taking into account possible metallicity corrections). ",
        "introduction": "Cepheids are important standard candles in determining the extra-galactic distance scale. The results of the \\HP\\ mission allowed a calibration of the Galactic Period-Luminosity ($PL-$) relation in the $V$-band (Feast \\& Catchpole 1997), $V$ and $I$-bands (Lanoix et al. 1999a) and $V$, $I$, $K$ and the reddening-free Wesenheit index (Groenewegen \\& Oudmaijer 2000, hereafter GO00). Feast \\& Catchpole (1997) and GO00 combined these data with available data on LMC Cepheids to determine the distance to the LMC.  It is well know that the infrared has some considerable advantages over the optical region for deriving distances. First of all the problem of reddening is less important, and second, as summarised in GO00, the effect of metallicity on the $PL$-relation seems to be less.  Of the 236 \\C\\ in the sample considered in GO00, only 63 have sufficiently well determined infrared light curves to determine accurate intensity-mean magnitudes (also see Groenewegen 1999). In addition, the number of Magellanic Clouds (MCs) \\C\\ with well determined infrared lightcurves is also rather small, of order 20 in each Cloud (Laney \\& Stobie, 1994, 1986a).  The micro lensing surveys have increased the number of known \\C\\ in the MCs tremendously and they provide optical photometry for many of them. In particularly, Udalski et al. (1999b,c; hereafter U99b,c\\footnote{In April 2000 the \\OG\\ team released a re-reduced dataset changing the photometry slightly, and hence the $PL$-relations with respect to those in the papers published in 1999. The new data and corresponding $PL$-relations can be found on the \\OG\\ Homepage. When appropriate I will refer to these new values.})  recently published the results on Cepheids in the Small and Large Magellanic Cloud from the \\OG\\ survey.  Complementary, in the infrared, the \\M\\ survey is an ongoing all-sky survey in the $JHK_{\\rm s}$ near-infrared bands, that, when completed, will contain of order 300 million stars (Beichman et al. 1998). The \\M\\ team released their second incremental data set in March 2000, that includes the MC area. In parallel, the {\\sc denis} survey is a survey of the southern hemisphere in $IJK_{\\rm s}$ (Epchtein et al. 1999), that recently released a Point Source Catalog (PSC) of sources in the direction of the MCs (Cioni et al. 2000a).  In this paper the \\OG\\ database is correlated with the \\M\\ second incremental database and the \\DE\\ PSC of MC stars (Sect.~2), transformation formulae from the natural \\M\\ system to the Carter (1990) system are derived (Sect.~3), the spatial structure of the MCs is investigated (Sect.~4), and $PL$-relations in $JHK$ for fundamental mode and first overtone pulsators derived (Sect.~5). The results are discussed in Sect.~6. ",
        "conclusions": "\\subsection{The slopes of the $PL-$relations}  One of the main results of this paper is the derivation of the slopes of $PL$-relation in the Wesenheit-index ($W$), and in $JHK$, for both FU and FO pulsators. In this section the slopes are intercompared, compared to previous determinations, and to predictions of theoretical models.  The first thing to remark is that in the LMC the slope of the $PL$-relation of the FU and FO pulsators disagree at the 1.6-2.9$\\sigma$ level. For all 4 bands the slope of the $PL$-relation of the FO pulsators is steeper than that of the FU pulsators. In the SMC, the slope of the $PL$-relation of the FO pulsators is steeper than that of the FU pulsators in $WJH$, but in all 4 bands the difference is less than 1$\\sigma$, largely influenced by the fact that the error in the slope of the $PL$-relation for the SMC FO pulsators is relatively large.  The second thing to remark is that for the FU pulsators the slope for the $PL$-relation of the LMC and SMC Cepheids agree within $1\\sigma$ in $WK$. In $JH$ the slope of the $PL$-relation for the LMC pulsators is steeper than that for the SMC pulsators at about the 1.7-2.2$\\sigma$ level. For the FO pulsators the slope of the $PL$-relation for the LMC \\C\\ is steeper than that of the SMC \\C\\ in all 4 bands but the difference is 1.5$\\sigma$ in $HK$ and less in $WJ$.   Table~5 collects slopes quoted in the literature both from observations and theoretical models, and the final values derived in the present paper.  Regarding the Wesenheit-index, the only previous determination was by Tanvir (1999), who used a different definition for this index and so can not be compared directly to the present result. The results by U99a are essentially based on the same sample considered here and hence are very close to the present results.  Regarding the IR relations, the slopes derived in the present work are the most accurate available for the moment. The slopes agree with in the errors with all previous determinations. The previously most accurate slopes were by Gieren et al. (1998)\\footnote{I have calculated the error in the slope using the data they kindly provided (their Table~9), and made a division on period as the sample they consider almost exclusively contains \\C\\ with long periods, unlike the \\OG\\ sample.}. Also here the agreement is excellent. The reliability of the result is enhanced if one considers that the Gieren et al. sample contains mainly long-period \\C\\ while the \\OG\\ sample is dominated by short-period \\C.  The agreement with theoretically predicted slopes is less satisfactory. For the Wesenheit-index the predicted slopes by Bono \\& Marconi (2000, private communicataion) are too shallow compared to the observations at the 2-4$\\sigma$ level. In $J$ the models by Alibert et al. (1999) are clearly too steep. In $K$ the agreement with the Bono et al. (1999) models is within the errors, but the Alibert et al. (1999) models predict slopes that are too steep.  Bono et al. (1999) are the only ones to predict the slopes for FO-pulsators (averaged over $Z$ = 0.02, 0.008 and 0.004). Their prediction is in good agreement with the observed slope in the LMC and within 2$\\sigma$ of the observed slope in the SMC.  Clearly, the present results can be used as tight constraint for theoretical modelling as for the first time accurate slopes in the infrared for both FU and FO pulsators are presented.    \\begin{table*} \\caption[]{Slopes of the $PL$-relations.}  \\small  \\begin{tabular}{ccll} \\hline Slope         & Colour & System & Reference \\\\ \\hline  \\multicolumn{4}{c}{Emperical results, FU pulsators}  \\\\  $-3.337 \\pm 0.013$ & Wesenheit & LMC & this work \\\\ $-3.411 \\pm 0.036$ & Wesenheit & LMC & Tanvir (1999); for $W = I - 1.45\\;(V-I)$ \\\\ $-3.300 \\pm 0.011$ & Wesenheit & LMC & Udalski et al. (1999a) \\\\ $-3.328 \\pm 0.023$ & Wesenheit & SMC & this work \\\\ $-3.310 \\pm 0.020$ & Wesenheit & SMC & Udalski et al. (1999a) \\\\   $-3.144 \\pm 0.035$ & J & LMC & this work \\\\ $-3.129 \\pm 0.052$ & J & LMC & Gieren et al. (1998) \\\\ $-3.147 \\pm 0.065$ & J & LMC & this work; 54 stars with $\\log P > 0.833$ from Gieren et al. (1998)\\\\ $-2.99 \\pm 0.59$  & J & LMC & this work;  5 stars with $\\log P < 0.833$ from Gieren et al. (1998)\\\\ $-3.31 \\pm 0.10$  & J & LMC & Madore \\& Freedman (1991) \\\\ $-3.31 \\pm 0.12$  & J & LMC & Laney \\& Stobie (1986a; their solution 1b)  \\\\  $-3.037 \\pm 0.034$ & J & SMC & this work \\\\ $-3.22 \\pm 0.17$  & J & SMC & Laney \\& Stobie (1986a; their solution 10b)  \\\\   $-3.236 \\pm 0.033$ & H & LMC & this work \\\\ $-3.249 \\pm 0.044$ & H & LMC & Gieren et al. (1998) \\\\ $-3.246 \\pm 0.056$ & H & LMC & this work; 54 stars with $\\log P > 0.833$ from Gieren et al. (1998)\\\\ $-3.49 \\pm 0.26$  & H & LMC & this work;  5 stars with $\\log P < 0.833$ from Gieren et al. (1998)\\\\ $-3.72 \\pm 0.07$  & H & LMC & Caldwell \\& Laney (1991)  \\\\ $-3.40 \\pm 0.10$  & H & LMC & Laney \\& Stobie (1986a; their solution 2b)  \\\\ $-3.37 \\pm 0.10$  & H & LMC & Madore \\& Freedman (1991) \\\\  $-3.160 \\pm 0.032$ & H & SMC & this work \\\\ $-3.25 \\pm 0.07$  & H & SMC & Caldwell \\& Laney (1991)  \\\\ $-3.36 \\pm 0.15$  & H & SMC & Laney \\& Stobie (1986a; their solution 11b)  \\\\   $-3.246 \\pm 0.036$ & K & LMC & this work \\\\ $-3.267 \\pm 0.041$ & K & LMC & Gieren et al. (1998) \\\\ $-3.304 \\pm 0.052$ & K & LMC & GO00 [54 stars with $\\log P > 0.833$ from Gieren et al. (1998)]\\\\ $-3.37 \\pm 0.39$  & K & LMC & GO00 [ 5 stars with $\\log P < 0.833$ from Gieren et al. (1998)]\\\\ $-3.42 \\pm 0.09$  & K & LMC & Madore \\& Freedman (1991) \\\\ $-3.41 \\pm 0.10$  & K & LMC & Laney \\& Stobie (1986a; their solution 3b)  \\\\  $-3.212 \\pm 0.033$ & K & SMC & this work \\\\ $-3.38 \\pm 0.15$  & K & SMC & Laney \\& Stobie (1986a; their solution 12b)  \\\\  \\multicolumn{4}{c}{Theoretical results, FU pulsators}  \\\\  $-3.02 \\pm 0.04$   & Wesenheit & 0.02  & Bono \\& Marconi (2000, private communication) \\\\ $-3.17 \\pm 0.04$   & Wesenheit & 0.008 & Bono \\& Marconi (2000, private communication) \\\\ $-3.21 \\pm 0.05$   & Wesenheit & 0.004 & Bono \\& Marconi (2000, private communication) \\\\  $-3.286$          & J & 0.01  & Alibert et al. (1999)\\\\ $-3.262$          & J & 0.004 & Alibert et al. (1999)\\\\  $-3.19 \\pm 0.09$  & K & 0.008 & Bono et al. (1999) \\\\ $-3.395$          & K & 0.01  & Alibert et al. (1999) \\\\ $-3.27 \\pm 0.09$  & K & 0.004 & Bono et al. (1999)  \\\\ $-3.369$          & K & 0.004 & Alibert et al. (1999) \\\\  \\multicolumn{4}{c}{Emperical and theoretical results, FO pulsators}  \\\\  $-3.381 \\pm 0.076$ & K & LMC & this work, FO-pulsators \\\\ $-3.102 \\pm 0.155$ & K & SMC & this work, FO-pulsators \\\\ $-3.44 \\pm 0.05$  & K & all $Z$ & Bono et al. (1999), FO-pulsators \\\\  \\hline \\end{tabular} \\end{table*}    \\subsection{The relative distance SMC-LMC}  These results allow a determination of the relative distance SMC to LMC, based on the various $PL$-relations. Since the slopes of the $PL$-relations in SMC and LMC are not exactly the same, the absolute magnitudes are calculated for $\\log P = 0.5$ when using FU pulsators, and $\\log P = 0.3$ for FO pulsators for each Cloud and then subtracted, using the $PL$-relations where the distance effect has been taken out. The results are listed in Table~6. The errors are based on the errors in the slope and zero point. It seems that the difference in distance modulus is somewhat less using the IR $PL$-relations than from the Wesenheit-index. A mean using all available 8 determinations yields ${\\Delta}_{\\rm SMC-LMC} = 0.50 \\pm 0.02$ mag.  This is in good agreement with the original determination in U99c based primarily on the Wesenheit index for the FU pulsators, and the 0.50 $\\pm$ 0.03 determined by Cioni et al. (2000b) as the average of the difference in $JHK$ and $m_{\\rm bol}$ of the TRGB (Tip of the Red Giant Branch) magnitudes based on \\DE\\ data for the two Clouds.  All this assumes no explicit metallicity correction. As recapitulated in GO00 there is no agreement between theory and empirical evidence, and amongst different theoretical models, about the extent of a metallicity dependence of the $PL$-relation, if any. In the infrared and Wesenheit-index there are no empirical estimates for the metallicity dependence. Table~7 summarises the predictions of $\\Delta M = M$(LMC) $-M$(SMC) in $W,J,K$ for 2 sets of recent models at $\\log P = 0.5$. Again the disagreement is obvious and at a level of upto 0.1 mag which is not negligible.   \\begin{table} \\caption[]{The relative distance modulus SMC-LMC} \\begin{tabular}{cc} \\hline $\\Delta$        & Remarks \\\\ \\hline 0.51 $\\pm$ 0.03 & U99c; W; FU \\\\  0.510 $\\pm$ 0.023 & present work; W; FU \\\\ 0.522 $\\pm$ 0.042 & present work; W; FO \\\\  0.479 $\\pm$ 0.044 & present work; K; FU \\\\ 0.488 $\\pm$ 0.091 & present work; K; FO \\\\ 0.465 $\\pm$ 0.041 & present work; H; FU \\\\ 0.511 $\\pm$ 0.081 & present work; H; FO \\\\ 0.469 $\\pm$ 0.044 & present work; J; FU \\\\ 0.534 $\\pm$ 0.092 & present work; J; FO \\\\  0.475 $\\pm$ 0.022 & mean of the 6 IR determinations \\\\ 0.504 $\\pm$ 0.015 & mean of all 8 determinations \\\\  \\hline \\end{tabular} \\end{table}   \\begin{table} \\caption[]{Metallicity dependence of the absolute magnitude between LMC and SMC, $\\Delta M = M$(LMC) $-M$(SMC) at $\\log P = 0.5$ for FU pulsators.} \\begin{tabular}{rrrl} \\hline ${\\Delta M}_{\\rm W}$ & ${\\Delta M}_{\\rm J}$ & ${\\Delta M}_{\\rm K}$ & Reference \\\\ \\hline $+0.030$ & $-$      & $+0.020$ & Bono et al. (1999)\\\\ $-$      & $-0.093$ & $-0.091$ & Alibert et al. (1999)\\\\ \\hline \\end{tabular} \\end{table}"
    },
    "0010/astro-ph0010251_arXiv.txt": {
        "abstract": "We have undertaken a systematic search for diffuse radio halos and relics in all of the Abell clusters that are visible in the Westerbork Northern Sky Survey (WENSS). In this survey we found 18 candidates, 11 of which are already known from the literature, and 7 for which we provide the first evidence of diffuse radio emission. All the clusters in this sample show other evidence for a recent or ongoing merger. We also investigate the correlation between cluster X-ray luminosity and radio power of halos. We develop a very simple model for merger shocks that reproduces the sense of this correlation, although it is probably not as steep as the correlation in the data. We discuss the implications of X-ray--radio correlations for future detections of radio halos. ",
        "introduction": "\\label{sec:intro}  A number of clusters of galaxies are known to contain large-scale diffuse radio sources which have no obvious connection to the clusters' population of galaxies. These sources are referred to as radio halos when they appear projected on the center of the cluster, and are called relics when they are found on the cluster periphery. Because of projection effects, the distinction between halos and relics is physically debatable, but does provide a convenient classification for their discussion. It is also possible that halos and relics in fact have different physical origins, as we discuss below. Halos (and relics; we will use ``halos'' hereafter for brevity) are believed to be produced by synchrotron emission from a population of relativistic electrons which have been accelerated or reaccelerated, possibly by shocks in the intracluster medium (e.g., Jaffe 1977; Roland 1981; Schlickeiser et al.\\ 1987). These shocks may be the product of cluster mergers. In fact, all known halos are found in clusters which show other signs of being in some stage of a merger (e.g., Feretti 1999; Schuecker \\& B\\\"{o}hringer 1999). In the early stages of mergers, halos are often found on the border between the subclusters, where the cluster gas is first being shocked (e.g., Abell 85; Slee \\& Reynolds 1984). In more advanced mergers, more conventional centrally located halos (e.g., Coma) and peripheral relics (e.g., Abell 3667) are found.  The most common explanation of the physical origin of halos and relics is that they originate from particle acceleration in merger shocks, but other theories have been suggested. En{\\ss}lin et al.\\ (1998) suggest that relics may trace shocks created in the initial structure formation of the Universe. Liang et al.\\ (2000) posits that turbulent reacceleration may maintain the population of cosmic rays necessary to produce a halo after they have been accelerated initially by merger shocks. There is also disagreement over whether the cosmic ray electrons are accelerated directly from the thermal population or are re-accelerated cosmic rays previously produced in starbursts and AGN. Alternately, the secondary electron model (Dennison 1980; Blasi \\& Colafrancesco 1999 and references therein) proposes that the necessary electrons are created as a result of interactions between cosmic ray protons and the intracluster gas. The secondary electron model is unique in that it does not require a merger to create a radio halo.  Halos and relics are rare phenomena, with only about 10--20 having been known until quite recently. With the completion of the NRAO VLA Sky Survey (NVSS: Condon et al.\\ 1998) and Westerbork Northern Sky Survey (WENSS: Rengelink et al.\\ 1997), systematic searches for additional sources have been possible, yielding about 25 new halos and relics. Evidence on the exact origin of halos and relics is still unclear, mostly because of their small numbers. With a larger sample, however we can begin to determine if the merger shock picture of their formation is correct, and perhaps even disentangle the different mechanisms for halo and relic formation, if they are in fact different phenomena.  We assume $H_0 = 50$ km s$^{-1}$ Mpc$^{-1}$ and $q_0 = 0.5$ throughout this paper. ",
        "conclusions": "\\label{sec:concl}  We have discovered 7 new candidate radio halos and relics in a search of Abell clusters in the WENSS. We also confirm the presence of diffuse emission in 7 clusters and find further evidence of such emission in 2 more clusters where the presence of a halo has been posited but remains uncertain. Our search technique detected all but one previously known source that fell within our sample. More detailed radio observations of the new radio halos and relics would be very useful to accurately determine their structure.  We argue that radio halos or relics are only found in clusters which are currently undergoing a cluster merger. All the clusters in our sample show evidence for a merger from either their X-ray surface brightness distribution or their galaxy distribution. This can explain the relative rarity of diffuse radio emission in clusters.  We also find weak evidence for the observed correlation between monochromatic power of radio halos and cluster X-ray luminosity. We present a very simple model for the correlation of radio power with X-ray luminosity or temperature in clusters which are currently merging, on the assumption that the radio-emitting electrons are accelerated by merger shocks. We consider two cases for this model: one in which the magnetic field is the same for all clusters and one in which the field varies as a fixed proportion of the gas pressure. The latter model is marginally more successful at fitting the data. This argument predicts a strong radio--X-ray luminosity correlation, although not as steep as the one observed at 20~cm.  Our survey is the first to look for halos in a large sample of clusters with low X-ray luminosity. In general, radio halos and relics are not found in low X-ray luminosity clusters. We argue that this is the result of the steep radio power vs.\\ X-ray luminosity correlation. If this is true, many more low luminosity clusters could be detected as diffuse radio sources if the sensitivities of the surveys could be greatly increased. Clearly, deeper imaging with high sensitivity at short ($u,v$) spacings is necessary to  test whether the $P-L_X$ correlation holds  at lower X-ray luminosities."
    },
    "0010/astro-ph0010192_arXiv.txt": {
        "abstract": "We present an exhaustive statistical analysis of the associated ($\\Delta~v_{\\mathrm{abs}}~<~5000$~\\kms), high ionization (\\CIV, \\NV, \\OVI) narrow absorption line (NAL) systems in a sample of 59 QSOs defined from the {\\it Hubble Space Telescope} QSO Absorption Line Key Project. The goals of the research were twofold: (1) to determine the frequency of associated NALs at low redshift and in low luminosity QSOs; and (2) to address the question of what QSO properties either encourage or inhibit the presence of associated NAL gas. To that end, we have compiled the QSO rest--frame luminosities at {2500~\\AA}, {5~GHz}, and {2~keV}, spectral indices at {2500~\\AA} and {5~GHz}, the {H$\\beta$} emission line {\\fwhm}, and the radio core fraction at observed {5~GHz}. In addition, we have measured the {\\CIV} emission line {\\fwhm}. We find 17 associated NALs (16 selected by {\\CIV} and 1 selected by {\\OVI}) toward 15 QSOs, of which {$\\sim10$} are statistically expected to be intrinsic. From a multivariate clustering analysis, we find that the QSOs group together (in parameter space) based primarily on radio luminosity, followed (in order of importance) by radio spectral index, {\\CIV} emission line {\\fwhm}, and soft X--ray luminosity. We find that radio--loud QSOs which have compact radio morphologies, flat radio spectra [$\\alpha(\\mathrm{5~GHz}) > -0.5$], {\\it and} mediocre {\\CIV~\\fwhm} ($\\lesssim6000$~\\kms) do {\\it not} have detectable associated NALs, down to {$\\rew($\\CIV$)=0.35$~\\AA}. We also find that BALQSOs have an enhanced probability of hosting detectable NAL gas. In addition, we find that the velocity distribution of associated NALs are peaked around the {\\it emission} redshifts, rather than the systemic redshifts of the QSOs. Finally, we find only one strong NAL [$\\rew($\\CIV$)\\gtrsim 1.5$~\\AA] in our low redshift sample. A comparison with previous higher redshift surveys reveals evolution in the number of strong NAL systems with redshift. We interpret these results in the context of an accretion--disk model. We propose that NAL gas hugs the streamlines of the faster, denser, low latitude wind, which has been associated with {\\it broad} absorption lines. In the framework of this scenario, we can explain the observational clues as resulting from differences in orientation and wind properties, the latter presumably associated with the QSO radio properties. ",
        "introduction": "\\pagestyle{myheadings} \\markboth{\\small~~Ganguly et al. \\hfill QSO--Intrinsic Absorbers~~} {\\small~~Ganguly et al. \\hfill QSO--Intrinsic Absorbers~~}  Ultraviolet resonant narrow (i.e., with velocity widths {$\\lesssim500$~\\kms}) absorption lines (NALs) observed in the spectra of QSOs arise from an eclectic assortment of objects ranging from intervening galaxies to intracluster gas to the intergalactic medium. Additionally, the well known broad absorption line (BAL) phenomenon arises from high--velocity dispersion ($\\Delta~v\\sim10,000$~\\kms) gas intrinsic to the QSO. In the last few years, some high--redshift NALs have been shown to exhibit the signature of partial coverage, implying that the gas is close to the central engine \\citep{bs97,bhs97,ham97a,gan99,sp00}. Intrinsic NAL gas has also been identified through variability of the absorption profiles on relatively short timescales ($\\sim$~months) in the host QSO rest--frame \\citep{ham95,bhs97,ham97b,ham97c,ald97}. For clarity, we introduce the acronym NALQSO as refering to QSOs with truly intrinsic NALs, distinguishing them from QSOs in which the intervening gas is entirely responsible from the appearance of NALs, and from BALQSOs in which the absorbing gas is intrinsic to the QSO, but has a much larger velocity dispersion.  Through variability studies and through high--resolution spectroscopy, intrinsic narrow absorption lines show great promise in providing an understanding of the physical conditions of the environment in which QSOs reside. However, before asking questions about what can be learned about NALQSOs through these types of ventures, one must address where intrinsic NAL gas is located, what the geometry of the gas is, and why some QSOs have intrinsic NALs while others do not. That is, what is the physical origin of NALQSOs?  Preceding the revolution in the QSO absorption line field brought on by high--resolution spectroscopy, progress toward understanding the origin of NALQSOs was made through the statistical assessment of the number of intrinsic systems in subsamples of large surveys of intermediate to high redshift {$1.4 \\lesssim \\zem \\lesssim 4.1$} QSOs (where the UV resonant doublets, like {\\CIVdblt}, are shifted into the optical). Specifically, the subsamples were restricted to {\\it associated} NALs (those that appear within {$\\sim$5000~\\kms} of the QSO emission redshift). The first of these surveys began with a sample of QSOs from the Third Cambridge Catalogue of Radio Sources \\citep{wwpt,foltz86,and87}. These studies found that there was a statistical excess of associated systems selected by the {\\CIVdblt} doublet over what would be expected if NALs were randomly distributed in space. (Strictly speaking, the velocity range was established {\\it after the fact} using the excess.) Moreover, these studies concluded that high equivalent width systems ({$\\gtrsim1.5$~\\AA} in the rest--frame) were preferentially hosted by radio--loud QSOs. Later studies employing optically--selected samples of QSOs found no such excess, not even with radio--selected subsamples \\citep{ysb,sbs88}. The primary difference in the radio properties of the two types of samples was the radio spectral index. The QSOs from the radio--selected samples primarily had steep radio spectra while those from the optically--selected, radio--loud subsample had mostly flat radio spectra. A later survey of {\\MgII} absorbers in a sample of steep radio spectrum, radio--loud QSOs by \\cite{ald94} rediscovered the excess of associated systems.  Also in recent history, there has been mounting evidence that the ``warm absorption'' gas seen through soft X--rays in AGN may be connected to the intrinsic gas detected through UV absorption (e.g., Mathur, Elvis, \\& Wilkes 1995\\nocite{mew95}). In an HST/FOS archival survey of Seyfert 1 galaxies, Crenshaw {\\etal} (1999\\nocite{cren99}) find a one--to--one correspondence between the objects with X--ray warm absorbers and those with intrinsic UV absorption. Moreover, Brandt, Laor, \\& Wills (2000\\nocite{brandt}, hereafter, BLW) find a striking relation between the rest--frame equivalent width of {\\CIV} absorption and the optical/X--ray two--point spectral index, $\\aox$, in {$\\zem\\lesssim0.5$} Palomar-Green QSOs. They have argued that this relation is driven by absorption, with absorbed objects having both large {\\CIV} equivalent widths and large negative values of {$\\aox$} (corresponding to apparently weak soft X--ray emission relative to the optical emission).  Motivated by these recent developments, we consider the connection between intrinsic narrow absorption line gas and the properties of {\\it low redshift} QSOs to determine which properties either encourage or inhibit the detectable presence of intrinsic NALs. As a by--product of using a low redshift sample, we extend the analysis to lower luminosity QSOs -- something difficult to do at high redshift -- and we investigate trends with redshift.  Similar to the Crenshaw {\\etal} (1999\\nocite{cren99}) analysis of Seyfert 1 galaxies, we start with the archived HST/FOS data set provided by the HST QSO Absorption Line Key Project \\citep{kpi,kpvii,kpxiii}. This QSO sample is unbiased toward the detection of associated absorption and has uniformly good {$S/N$} spectra. We define the sample in \\S2. In \\S3, we present the results of our efforts to detect associated absorption in these QSOs. In \\S4, we discuss the optical, radio, and X--ray properties compiled for this study. We perform a multivariate tree--clustering analysis in \\S5 and summarize our results in \\S6. Finally, we consider and discuss the constraints the results place on scenarios for the narrow absorption line gas in \\S7. ",
        "conclusions": "\\subsection{Model Requirements from the Properties of NALQSOs and the Disk-Wind Model}  From the preceding sections, there are four major attributes that any model attempting to explain NALQSOs, or QSOs in general, must exhibit. First, and foremost, QSOs that have properties placing them in group three (i.e., radio--loud, flat radio spectrum, radio core dominated, mediocre {\\CIV~\\fwhm}) should be noticeably devoid of intrinsic NALs with {$\\rew($\\CIV$)\\gtrsim0.35$~\\AA}. This can be accomplished either through the absence of intrinsic NAL gas or through an observational selection effect (e.g., orientation or small equivalent width).  Second, the velocity field of intrinsic NAL gas should be such that, when projected onto the line of sight, intrinsic NALs may appear either blueward or redward of the UV emission line peak. But the distribution of line of sight velocities should be roughly centered about the QSO emission redshift. Third, intrinsic NAL gas should evolve such that strong [$\\rew($\\CIV$)>1.5$~\\AA] systems are conspicuously rare at low redshift ($\\zem\\lesssim1$). Finally, there should be a connection between soft X--ray weak QSOs and the strength of UV absorption and an enhanced probability of detecting intrinsic NAL gas in BALQSOs.  To explain the above properties, we consider the disk--wind model of QSOs. We begin with a brief summary of the disk--wind model that has been used to explain the existence of BALQSOs \\citep[hereafter MCGV]{mur95}. In this model, a wind is launched from the entire surface of the accretion disk. The dynamics of the wind are effectively determined by radiation pressure. Local radiation pressure elevates the atmosphere of the disk over a range of radii, while UV photons originating from the innermost parts of the disk provide the radiation pressure needed to accelerate this gas radially outwards until it reaches a terminal speed.  As the wind expands from the center it conserves angular momentum and therefore continues to rotate out to a large distance, although at an ever decreasing rate.  The inner part of the wind is the source of the broad UV emission lines. When viewed from large inclination angles, absorption by this high-velocity dispersion wind produces deep, broad troughs, resulting in a BALQSO. The detailed dynamics have been worked out analytically by MCGV and simulated numerically by Proga, Stone, \\& Kallman (2000\\nocite{psk00}; hereafter, PSK).  An example of the resulting geometry is shown in the cartoon of Fig.~\\ref{fig:diskwind}. With these theoretical ideas in mind, we consider how the observed intrinsic NAL properties can be explained in the context of wind models.  \\subsection{One Possible Explanation}  If we postulate that the gas producing intrinsic NALs exists in all QSOs (in the spirit of perpetuating unification schemes), a disk--wind scenario, much like the one described above, offers a framework within which the properties of intrinsic NALs can be understood. In Fig.~\\ref{fig:diskwind}, we show a cartoon of the disk--wind model. In this scheme, intrinsic NAL gas exists in a region hugging the wind streamlines at relatively large distances from the continuum source. The simulations of PSK show that the region in the polar direction is also occupied by gas launched from the disk, but this gas is highly ionized and hence not an effective absorber. Furthermore, we hypothesize that there is a difference in the properties of the wind (density and mass loss rate) between radio--loud and radio--quiet QSOs. In particular, we propose that, in comparison to radio--loud QSOs, radio--quiet QSOs have a higher mass loss rate (relative to the mass accretion rate), a higher wind velocity, and therefore, a higher density of matter in their winds.  Our proposal is based on the observation that the UV--to--X--ray spectra of radio--quiet QSOs are steeper than those of their radio--loud counterparts, meaning that radio--quiet QSOs have stronger UV bumps than radio--loud ones (see, for example, Green {\\etal} 1995\\nocite{green95}).  The relative UV--to--X--ray luminosity of a QSO is a central parameter in the models of MCGV since it determines how effectively the wind can be accelerated by UV photon pressure and how vulnerable it is to over--ionization by X--rays.  The reason for this difference in spectral energy distributions is not important for the purposes of our argument. However, we are tempted to speculate that in radio--quiet QSOs, the accretion rate onto the central black hole is much closer to the Eddington limit than in radio--loud QSOs (in other words, the ratio $\\dot M/\\dot M_{\\mathrm{Edd}}$ is closer to unity in the former case and considerably lower than unity in the latter).  In the context of the picture described above, we may interpret the five groups from our clustering analysis as the result of a different radio luminosity (radio--quiet vs. radio--loud) and a different viewing angle of the central engine by the observer. We illustrate this interpretation in Fig.~\\ref{fig:diskwind} by using arrows to indicate the direction of the observer corresponding to each group. An observer at a given orientation is never guaranteed to observe a NALQSO because the intrinsic NAL gas may be patchy and not cover all lines of sight in the same direction uniformly.  We elaborate on the proposed scheme below, starting with groups three, four, and five, which contain radio--loud QSOs for which the radio morphology can serve as an orientation indicator, then continuing with the radio--quiet QSOs.  \\begin{itemize}  \\item QSOs in group five are double--lobed radio sources, which suggests that the viewing angle from the axis of the accretion disk is large. The observer's line of sight is likely to pass through parcels of intrinsic NAL gas, with the result that intrinsic NALs appear in the spectrum. Although at this inclination the observer's line of sight is likely to pass through the faster, denser parts of the wind, the density of the fast wind in radio--loud QSOs is low enough, as we have postulated above, that BALs do not appear. This contention is supported by the fact that QSOs in this group do not show the signature of heavy absorption in their spectral energy distributions, unlike QSOs in group two, which we discuss later on.  \\item Group three represents the opposite extreme in inclination angles. The compact radio morphology and flat radio spectra of objects in this group, suggest that the observer's line of sight is very close to the axis of the radio jet. As a result, the likelihood of photons from the UV continuum source or broad--emission line region to pass through the NAL region on their way to the observer is extremely small. Hence, this group includes no NALQSO candidates.  \\item Group four comprises a mixture of lobe--dominated and core--dominated radio sources. Thus, we interpret this as a group of QSOs viewed at intermediate orientations between pole on and edge on with a finite probability that the line of sight intercepts NAL gas. It is noteworthy that this group contains the objects with the broadest {\\CIV} emission lines in our entire collection. If we adopt the above prescription for the dynamics of the wind, then the large width of the {\\CIV} emission lines in group four suggests that the combination of projected rotation and outflow velocities is maximum at intermediate inclinations.  \\item Since radio--quiet QSOs are associated with compact radio sources the radio morphology cannot serve as an indicator of orientation. Nevertheless, we notice that BALQSOs are preferentially found in group two and accordingly we identify this group with highly inclined QSOs so that the observer's line of sight passes through the dense, fast wind. These are, in a sense, the radio--quiet analogs of the QSOs in group five. Group two QSOs are UV weak and soft X--ray weak, which suggests that they are are highly absorbed.  The high absorption fits in very well with our proposed picture for the wind properties: it appears only in radio--quiet QSOs where the wind is hypothesized to be dense, but not in radio--loud QSOs, where it is not.  \\item Finally, group one can be taken to represent intermediate--to low inclination cases among radio--quiet QSOs. This group contains the radio--quiet analogs of group three and group four QSOs. The relatively small number of QSOs in this group is likely the reason that this group does not significantly break up via the {\\CIV~\\fwhm}. As a result of the lack of an orientation indicator in radio--quiet QSOs we cannot pick out a subset of such objects without intrinsic NALs as could be done with their radio--loud counterparts. Moreover, if the winds of radio--quiet QSOs are indeed denser than those of radio--loud ones, there may be no viewing direction that can avoid NAL gas completely.  \\end{itemize}  The ability of intrinsic NALs to appear either blueward or redward of the broad emission line (BEL) redshifts can be understood as a consequence of the velocity field of the wind. The simulations of PSK show that gas returning from a high altitude above the disk can be entrained by the fast outflowing stream and flow inwards towards the center. Thus it can produce absorption lines that are redshifted relative to the peak of the broad emission lines. Alternatively, the rotation of the wind can also result in redshifted NALs since the projected velocity of the NAL gas can be either positive or negative, depending on its azimuthal position.  A particularly interesting issue is raised by the fact that we find only one strong associated NAL ($\\rew($\\CIV$) \\gtrsim 1.5$~\\AA) in our low redshift sample. In comparison, previous surveys of higher--redshift QSOs has revealed a considerably larger frequency of such strong associated NALs in objects of comparable luminosity (see \\S6.1, above). This may be a consequence of an evolution of the fueling rate of QSOs with redshift. If the QSO fueling rate is higher at higher redshifts, then one might expect the mass loss rate, hence the density of the wind, to be higher as well. Therefore, the simple picture that we have examined offers the tantalizing prospect of tracing the fueling history of QSOs through the strength of their intrinsic NALs."
    },
    "0010/astro-ph0010471_arXiv.txt": {
        "abstract": "The Becklin-Neugebauer (BN) object in the Orion Nebula Cluster (ONC) is a well-studied optically invisible, infrared-bright young stellar object, thought to be an intermediate-mass protostar. We report here that BN exhibited nearly-sinusoidal periodic variability at the near-infrared H- and K$_s$-bands during a one month observing campaign in 2000 March/April. The period was 8.28 days and the peak-to-peak amplitude $\\sim$0.2 mag.  Plausible mechanisms for producing the observed variability characteristics are explored. ",
        "introduction": "The BN object was revealed during early raster scanning of the Orion Nebula region at 2$\\micron$, and immediately recognized as a candidate protostellar object (Becklin \\& Neugebauer 1967). BN is the brightest at near-infrared wavelengths of a group of $\\sim$20 intermediate- and high-mass young stars and protostars in OMC-1. These objects were discovered at mid-infrared (Rieke, Low, \\& Kleinman 1973; Downes et al. 1981; Lonsdale et al. 1982; Dougados et al. 1993; Gezari, Backman, \\& Werner 1998), radio (Churchwell et al. 1987; Felli et al. 1993; Menten \\& Reid 1995), and x-ray (Garmire et al. 2000) wavelengths. The total luminosity emanating from the embedded cluster is $\\sim$10$^5 L_{\\odot}$.  In addition to its luminous point sources, the BN region is also notable as the source of the spectacular H$_2$ ``fingers'' or ``bullets'' (Allen \\& Burton 1993) which extend several arcminutes to the northwest and southeast.  BN itself is extinguished by A$_V$ = 17 mag and has luminosity 2500 L$_{\\odot}$ (Gezari et al.) corresponding to a main sequence B3-B4 star.  It was the first of a still small class of young, mostly luminous, stars with the 2$\\micron$ CO bandheads in emission (Scoville et al. 1979, 1983).  These $\\Delta\\nu$=2 transitions are thought to arise from collisional or shock excitation in a hot, dense region, perhaps the inner part of a circumstellar disk or wind. BN also has relatively strong H$_2$ (Scoville et al. 1983) and weak Hu, Pf, and Br hydrogen recombination lines (e.g. Bunn, Hoare, \\& Drew 1995). Broad absorptions at 3.3$\\micron$ and 10$\\micron$ are due to circumstellar ice and dust (Gillett \\& Forrest 1973). ",
        "conclusions": "Given the protostellar nature of BN, it is worth considering the physical origin of the 1.6-2.2$\\mu$m flux (the shortest wavelengths at which BN has been detected). Assuming the B3-B4 spectral type and 17 mag of visual extinction from Gezari et al., the H-band magnitude matches within a few tenths that predicted for a reddened stellar photosphere. We take this as minor evidence that the H-band photometric variability may arise close to the photosphere.  At K$_s$-band, however, the (de-reddened) magnitude is almost 4 mag above the predicted photosphere.  The hot dust and/or gas producing the 2.2$\\mu$m excess must subtend an area larger than that predicted by a standard blackbody disk or shell model since all close-in grains are destroyed by stellar heating. Accretional heating is one way to do this.  That the K$_s$-band magnitude is dominated by circumstellar flux suggests, alternately, that the variability may occur in the dust envelope.  Despite these differences between H and K$_s$ in the ratio of non-photospheric to photospheric flux, the observed periods are the same with the ratio of period amplitudes consistent with reddening.  Periodic variable stars have been studied optically in the ONC region by Herbst et al. (2000, and references therein), Stassun et al. (1999), and Rebull (2001).  Almost all stars identified as periodic in these unbiased studies have been low mass, $<2 M_\\odot$.  Only one periodic variable earlier than mid-G has been found, JW660 with a mid-B spectral type and a mass $\\sim 6 M_\\odot$ (Hillenbrand 1997). The period of 6.15 days reported by Mandel \\& Herbst (1991) has, perhaps notably, not been found again in subsequent observing seasons (Herbst et al. 2000). BN, by contrast, is a B3-B4 star with a mass $\\sim 6-8 M_\\odot$ (and a maximum mass of 20 M$_\\odot$ if the B0 spectral type inferred from HII region characteristics is adopted).  It is the most massive periodic variable detected thus far in the Orion region. The next brightest stars having significant periods in Carpenter et al. are $>$2.2 mag fainter at K$_s$ with spectral types K0 and later.  BN's periodic variability could be due to a number of well-recognized phenomena, although none of the following explanations seems totally satisfactory.  Periodic variability in young stars is usually interpreted in terms of long-lived nonuniformities in photospheric structure, i.e. spots that are either cooler or hotter the stellar effective temperature.  These spots rotate with the star and modulate the lightcurve as they pass through the line of sight of the observer. Interpreted as stellar rotation, the 8.28 day period implies an equatorial velocity of $\\sim$30 kms$^{-1}$ which is on the slow tail for rotation of intermediate- and high-mass young stars in the ONC (Wolff, Strom, \\& Hillenbrand 2001).  It is not generally accepted that massive stars like BN have the surface magnetic structures required for production of cool spots.  However, given that BN is a protostar, accretion may produce surface shocks perhaps also requiring magnetic fields in the form of ordered magnetic dipoles which lead to hot spots as material from the circumstellar environment falls in along them.  Considered independently, the variability amplitudes at H and K$_s$ can be well-modelled by spots with $\\Delta$T$\\approx$15,000 K from the photosphere and $\\sim$15-20\\% coverage, or $\\Delta$T$\\approx$5,000 K and $\\sim$50\\% coverage, as examples. However, neither cool nor hot spots are capable of producing the small H-K$_s$ color amplitude since the effect of adding a spot is essentially colorless in the near-infrared given the early B photosphere ($<$0.02 mag for $\\Delta$T$<$20,000 K and coverage $<$50\\%). If spot-modulated rotation is the cause of the observed photometric periodicity, the phase and the amplitude both should change on timescales of months to years as the spot structure varies. Although the 2000 February data point phases well with the period derived for the 2000 March/April time series, the 1998 March data point does not; however, this could be due simply to accumulation of period error over the longer time baseline.   Pulsating behavior leads to short periods (0.1-0.3 day) in radial modes and to only slightly longer periods ($<$1 day) in non-radial modes (e.g. $\\beta$Cep stars in the early B range and 53 Per type stars and others at late B types). Both the amplitude and the near-sinusoidal shape of BN's lightcurve are consistent with certain types of pulsational behavior, but the period is too long. Longer period variability ($\\sim$2-30 days) in massive stars is often explained in terms of winds, with some mechanisms requiring binary systems. The radio spectral index of BN is $\\alpha=0.8\\pm0.2$ from 2-6 cm (Felli et al. 1993), consistent with thermal emission from an ionized stellar wind.  The observed period and mass estimate for BN imply an orbital radius of $\\sim$0.15 AU for any hypothetical low-mass companion. If an outflow/wind from a companion was colliding with the outflow/wind from BN, pulsating behavior within the interaction region might be identified via x-rays (e.g. Ishibashi et al. 2000 for Eta Car). BN was seen in heavily absorbed x-rays by Chandra (Garmire et al. 2000).  However, the Chandra position was 1.1\" northwest ($\\sim$500 AU) of the near-infrared position, whereas all other x-ray/optical-infrared matches were within 0.5\". Garmire et al.  attribute the offset, if real, to physics associated with outflow phenomena interacting with stationary, dense cloud material; no binary system is required.  Another scenario to consider is an eclipsing binary. BN's lightcurve does not seem consistent with a fully eclipsing system given its low amplitude and nearly sinusoidal nature. A partial eclipse situation with a near-equal mass/size companion, a small orbital separation ($\\sim$10 R$_*$), and a reasonable inclination ($\\sim40^\\circ$) would match the gross shape, period, and amplitude of the phased lightcurves. In the near-equal mass situation, the true orbital period would be double that derived naively from observations since a single revolution produces two minima and two maxima.  However, eclipses should exhibit the same amplitude in all bands (modulo limb darkening), which is not what is observed for BN unless the companion has the same radius but the colors of a much cooler star.  Interestingly, Scoville et al. (1983) and others have suggested the possibility of wide binarity for BN to explain its large radial velocity relative to other ONC stars and to the ambient cloud.  A further possibility is periodic occultation by asymmetry in a circumstellar disk at the 0.15 AU orbital radius implied by the period. High column density, partially grey, orbiting material could produce the shape of the lightcurve as well as the color and magnitude amplitudes. It is interesting in this context to note that Biscaya et al. (1997) claimed time variability in BN's 2$\\mu$m CO bandhead emission lines, and that no emission is present in the spectrum of Penston et al. (1971).  The CO lines are formed in the dense, hot circumstellar environment and their time variability may have the same physical origin as the continuum variability found by us.   Finally, we note that there may be historical precedent for photometric variability in BN, as detailed in Table 2.  One must be cautious in interpreting this ensemble of early near-infrared measurements as indicative of large scale flux variations, however.  The observational complexities of working in the Orion region combined with a variety of aperture sizes may explain entirely the apparent differences in photometry."
    },
    "0010/astro-ph0010653_arXiv.txt": {
        "abstract": "The evolution of a neutron-star $r$-mode driven unstable by gravitational radiation (GR) is studied here using numerical solutions of the full nonlinear fluid equations.  The dimensionless amplitude of the mode grows to order unity before strong shocks develop which quickly damp the mode.  In this simulation the star loses about 40\\% of its initial angular momentum and 50\\% of its rotational kinetic energy before the mode is damped.  The nonlinear evolution causes the fluid to develop strong differential rotation which is concentrated near the surface and poles of the star. \\pacs{PACS Numbers: 04.40.Dg, 97.60.Jd, 04.30.Db} ",
        "introduction": " ",
        "conclusions": ""
    },
    "0010/astro-ph0010379_arXiv.txt": {
        "abstract": "We consider long-slit emission-line spectra of galactic nuclei when the slit is wider than the instrumental PSF, and the target has large velocity gradients. The finite width of the slit generates complex distributions of brightness at a given spatial point in the measured spectrum, which can be misinterpreted as coming from additional physically distinct nuclear components.  We illustrate this phenomenon for the case of a thin disc in circular motion around a nuclear black hole (BH). We develop a new method for estimating the mass of the BH that exploits a feature in the spectrum at the outer edge of the BH's sphere of influence, and therefore gives higher sensitivity to BH detection than traditional methods. Moreover, with this method we can determine the black hole mass and the inclination of the surrounding disc separately, whereas the traditional approach to black-hole estimation requires two long-slit spectra to be taken. We show that with a given spectrograph, the detectability of a BH depends on the sense of rotation of the nuclear disc. We apply our method to estimate the BH mass in M84 from a publicly available spectrum, and recover a value 4 times lower than that published previously from the same data. ",
        "introduction": "Long-slit spectroscopy has been for decades a fundamental tool of astronomy, and although integral-field spectrographs are now beginning to be important for ground-based observations of galactic dynamics (e.g., SAURON, Copin \\etal 2000), the Hubble Space Telescope (HST) has never been more reliant on long-slit spectroscopy.  Observations with long-slit spectrographs often use a slit that is wider than the FWHM of the instrumental point-spread-function (PSF). This practice is conventionally considered to enhance the signal-to-noise ratio (S/N) of the data at the price of what may be an insignificant loss in velocity resolution. Here we show that when the target has steep velocity gradients, the use of a wide slit can have more subtle effects, because the position and velocity information becomes entangled along the dispersion direction. If these effects are not recognized in the data, a misleading impression of the structure of the target can be inferred.  When the effects of a wide slit are recognized, however, they can increase the diagnostic power of the spectrum over that of a narrow-slit spectrum of equal S/N.  Since our study has been motivated by a programme of spectroscopy of galactic nuclei using the STIS instrument on HST (Marconi \\etal 2000; Axon \\etal 2000), we focus on the specific case of STIS spectra of gas swirling around a black hole in a galactic nucleus. However, the principles we elucidate have a wider applicability: whenever an object with steep velocity gradients is studied with a slit that is wider than the FWHM of the PSF. Large velocity gradients occur in shocks and contact discontinuities, as well as in accretion discs around black holes.  In Section 2 we analyze the appearance of galactic-disc spectra that are taken with a slit that is much wider than the PSF and has been laid along the disc's line of nodes. In Section 3 we use this analysis to develop a new estimator for the mass of the nuclear BH. In Section 4 we generalize the analysis of Sections 2 and 3 to a slit that is inclined to the line of nodes. We find that, paradoxically, the detectability of a BH with a given spectrograph depends on the sense of rotation of its accretion disc. We also investigate the case of a slit that does not pass through the nucleus. In Section 5 we consider the case in which the FWHM of the PSF is comparable to the slit width. In Section 6 we reanalyse STIS observations of M84 taken by Bower et al (1998), and show that the results of Section 4 enable us to understand these data with a much simpler model of the galaxy than that developed by Bower et al., and yield a mass for the black hole in M84 that is smaller than that of Bower et al.\\ by a factor $\\sim4$. ",
        "conclusions": "We have analysed the consequences of using a long-slit that is significantly wider than the PSF of the primary telescope optics. When the object under study contains large velocity gradients, such that the object's velocity can change appreciably across the slit, the pattern of light that will be measured on the detector differs qualitatively from what one would observe with an ideal narrow slit. There is more information in a spectrum taken with a wide slit that in a spectrum of equal S/N that is taken with a narrow slit. This happy state of affairs in natural intuitively, since a wide slit contains many narrow slits within it.  Competition between real velocity differences within the object, and offsets in velocity that arise because light from different points across the slit enter the spectrograph at different angles, frequently cause the intensity of light on the detector to have more than one peak at a given spatial location, even for a trivial velocity field in the object. Such multiple peaks in spectra of galactic nuclear discs have already been observed from the ground (\\eg Rubin, Kenney \\& Young 1997), and are evident in STIS spectra (Bower \\etal 1998). In this last observation, they were interpreted as evidence for multiple components along a single line of sight. Bertola \\etal (1998) showed that this does not need to be the case, and serious physical problems are likely to be encountered when a three-dimensional model is sought that displays such components. Here we confirm this finding, and in addition conclude that, when correctly interpreted, the multiple peaks have considerable diagnostic power: they enable us to detect nuclear black holes by exploiting data for a given galaxy at larger distances from the nucleus than in the traditional approach to BH mass-estimation. Our BH-detection method is therefore reliable down to lower black-hole masses ($M_\\bullet$) for which detection of Keplerian rise of velocity inwards is beyond the resolution limit. Moreover, our estimator recovers $M_\\bullet$ and the inclination $i$ of the disc independently from a single long-slit spectrum, whereas from the same spectrum the traditional approach determines only the value of the product $M_\\bullet\\sin^2i$ and requires an additional spectrum from an off-set slit to extract $M_\\bullet$ and $i$ separately from the product. Nevertheless we ought to point out that our method serves only as a rough indicator of a central mass concentration, and tells us nothing about the mass distribution inside the caustic. The traditional method that follows the Keplerian rise in velocities will secure the BH detection once data with sufficient resolution become available.  When the slit is inclined at some angle to the line of nodes, the distribution of light on the spectrograph's detector differs qualitatively depending on the sense in which the disc rotates. Consequently, with a given spectrograph setup, the detectability of a black hole depends on the sense in which the disc surrounding it rotates.  We applied our mass estimator to a published spectrum of M84 (Bower \\etal 1998), and find that in this galaxy $M_\\bullet$ is 4 times smaller than previously estimated. We argue that interpretation of data taken with a wide slit in terms of an infinitely thin slit will cause $M_\\bullet$ to be systematically overestimated.  Currently we are estimating the masses of nuclear BHs in a sample of $\\sim50$ nearby galaxies for which we have STIS spectra (Marconi \\etal 2000, Axon \\etal 2000).  \\begin{figure*} \\centerline{\\psfig{file=fig8.ps,width=\\hsize}} \\caption{Left: an $H\\alpha$ emission-line spectrum of M84  taken with STIS using the G750M grating (from Bower et al.\\ 1998). The dispersion direction is vertical. Centre: a modeled light distribution on the spectrograph's detector (convolved with instrumental PSF) for $i=74\\degr$, $\\hbox{PA}=21\\degr$, $M_\\bullet=4 \\times 10^8\\msun$, and the centre of the slit shifted by 0.03 arcsec from the nucleus. The distance to M84 has been taken to be $ 17\\Mpc$ and the surface-brightness of the emission has been assumed to be proportional to $R^{-1.5}$. Right: light distribution from the central panel integrated over detector pixels. \\label{bower}} \\end{figure*}"
    },
    "0010/astro-ph0010465_arXiv.txt": {
        "abstract": "We present the timing analysis results for our observations of the x-ray dip source X1916-053 conducted with RXTE between February and October of 1996. Our goal was to finally measure the binary period - as either  the x-ray dip period or $\\sim$1\\% longer optical modulation period, thereby establishing if the binary has a precessing disk (SU UMa model) or a third star (triple model).  Combined with historical data (1979-96), the x-ray dip period is measured to be 3000.6508 $\\pm$ 0.0009 sec with a 2$\\sigma$ upper limit $|\\dot P| \\leq 2.06 \\times 10^{-11}$. From our quasi-simultaneous optical observations (May 14-23, 1996) and historical data (1987-96), we measure the optical modulation period to be 3027.5510 $\\pm$ 0.0052 sec with a 2$\\sigma$ upper limit $|\\dot P| \\leq 2.28 \\times 10^{-10}$. The two periods are therefore each stable (over all recorded data) and require a $3.9087 \\pm 0.0008$d beat period. This beat period, and several of its harmonics is also observed as variations in the dip shape. Phase modulation of x-ray dips, observed in a 10 consecutive day observation, is highly correlated with the $\\sim$3.9d dip shape modulation. The 1987-1996 optical observations show that the optical phase fluctuations are a factor of 3 larger than those in the x-ray. We discuss SU UMa vs. triple models to describe the X1916-053 light curve behavior and conclude that the x-ray dip period, with smaller phase jitter, is probably the binary period so that the required precession is most likely similar to that observed in SU UMa and x-ray nova systems.  However the ``precession'' period stability and especially the fact that the times of x-ray bursts may partially cluster to occur just after x-ray dips, continue to suggest that this system may be a hierarchical triple. ",
        "introduction": "The discovery of periodic ($\\sim$50 min) dips in the x-ray flux from the moderately faint source 4U1915-05 (X1916-053) provided the first direct evidence for a binary periodicity in an x-ray burst source (Walter et al 1982, White and Swank 1982). Combined with the general success of the thermonuclear flash model for (Type I) x-ray bursts (e.g. Joss and Li 1980), this confirmed that bursters are indeed neutron stars accreting from binary companions. The x-ray dips are believed to be caused by partial occultation by vertical structure (or clouds) in the accretion disk. The short dip period ($\\sim$3000 sec), if identified as the binary period, provided the the first example of an ultra-compact x-ray binary with a very low mass degenerate companion.  The x-ray period has been reported with values from 2985 to 3015 sec (Walter et al. 1982; White and Swank 1982; Small et al. 1989; Yoshida et al. 1995 and Church et al. 1997). Smale et al. (1989) discovered a possible second candidate period of 2984.6 $\\pm$ 6.8 sec from the Ginga 1987 observations.  In addition to the primary dips, secondary dips with phase difference $\\sim 180^{\\circ}$ are sometimes observed.  The varieties of dip width, depth, phase, and association with the period show the complex behavior of X1916-053.  In addition to the 3000sec dip period, a 199d long-term x-ray modulation period was also reported by Priedhorsky and Terrell (1984).  The optical counterpart of X1916-053 was discovered (Grindlay et al. 1988) as a V=21 mag optical star with a modulation period of 3027.5 sec, $\\sim1\\%$ longer than the X-ray dip period.  The difference between  the x-ray and optical periods leads to the question of which represents the orbital period of the system. The optical period was found to be stable over (at least) 7 years (Callanan, Grindlay and Cool, 1995, hereafter CGC95).    In this paper, we briefly introduce SU UMa vs. triple system models previously suggested (Grindlay, 1992) for explaining the x-ray/optical period difference (section 2), describe our 1996 RXTE observations  of X1916-053 (section 3) and report (section 4) the timing analysis of the data, including the x-ray dip periodicity, $\\sim$ 3.9d period dip shape modulation, the dip phase jitter phenomenon, and the long term x-ray dip period stability from the combination of the RXTE and previous x-ray observations (Einstein (1979-1981), EXOSAT (1983) and Ginga (1987-1990)). Analysis of our optical May 14-23, 1996, optical observations quasi-simultaneous with RXTE and historical (1987-96) optical data are used to derive the optical ephemeris and phase jitter for comparison with the x-ray values. We use the x-ray dip ephemeris, which is able to predict dip times within \\about0.1 phase (300sec) back to 1979, to study the relation of the times of x-ray bursts to dips since a possible correlation was previously noted (Grindlay 1992). In section 5, we discuss SU UMa vs. triple models for the complex light curve behavior of X1916-053, concluding that either model would require interesting new phenomena to explain the present data but that the triple model would be required if the dip-burst correlation is real. ",
        "conclusions": "Table 3 shows all the x-ray dip periods reported since 1982.  The side bands from  the RXTE analysis of X1916-053 may explain the many different reports of the x-ray dip periodicity of X1916-053 .  Owing to the insufficient time span of the observations, both uncertainties of the 3000 sec period peak and its first neighbor side bands are large such that the peaks merge to affect the centroid of the 3000 sec peak.  The centroid shifts left or right depending on which (first left or right) side band is more significant.   The analysis in this paper suggests that there is a $\\sim$3.9d period modulation in the dip width and depth, and the phases may also have a $\\sim$3.9d period jitter with an amplitude of 0.1 (peak-to-peak).  The 6.5$\\pm$1.1d period for dip phase reported by Yoshida et al. (1995) is probably the sub-harmonic of 3.9d period.  The optical modulation also has a $\\sim$4d period in amplitude, and possibly phase modulation (Grindlay 1992), which is consistent with our RXTE x-ray observations.  The fact that both the x-ray and the optical light curves have the $\\sim$ 3.9d beat period implies that there are 3 periodicities in the X1916-053 system, with periods of approximately 3000 sec, 3027 sec and $\\sim$3.9d. Of these three periodicities, at most two of them are independent because of the beat relation among them. One of the two short (3000 sec and 3027 sec) periods should represent the binary orbital period and it should be the more stable one.  Both the optical and the x-ray period have long-term stabilities over the total historical data span of $\\sim$9 years and $\\sim$17 years respectively.  The period derivative of the optical modulation could be $\\sim$14 times larger than that for the x-ray dips, although both $\\dot{P}$ values are statistically insignificant.  The phase jitter statistics show that the x-ray dips have smaller fluctuations than the optical minima.  Thus the x-ray dip period may be the better candidate for the orbital period, and thus X1916-053 contains a precessing disk as in SU UMa systems.  However, the larger optical phase jitter could simply be due to the shallower modulation in the optical vs. x-ray (compare Figure~\\ref{tyoplc} and Figure~\\ref{tenfef}).  As outlined in section 2, SU UMa and triple models have been proposed to explain the $\\sim$1\\% difference of x-ray dip and optical variation periods.  We consider additional implications of each model.  \\subsection{SU UMa Model Implications}  Our result that the x-ray dips have less phase jitter, and are at least (if not more) stable over the historical database, provides strong support for the SU UMa model. Applying the SU UMa model to the X1916 system, the 3000 sec x-ray dip period is the orbital period which is due to the shadow of accretion stream which crosses our line of sight $\\sim$0.1 in binary phase before the secondary star. The 3027 sec optical period (the ``superhump'' period) is due to the beat with the 3.9d precession period of accretion disk.  The asymmetric disk and its $\\sim$3.9d retrograde precession that changes the vertical structure (cloudlets) at the outer edge with $\\sim$3.9d period, provide an explanation for the 3.9d modulation of x-ray dip shape and phase.  Hirose and Osaki (1990) performed hydrodynamic simulations of accretion disks for the superhump phenomenon in SU UMa stars.  They showed that the disk precession angular frequency can be written as  $$  {{{\\omega}_p} \\over{{\\omega}_{orb}}} = a(r) {q \\over {(1+q)^{1/2}}}\\eqno(12)$$  \\noindent where $$a(r)={{1\\over 4} {1\\over {r^{1/2}}} {d\\over {dr}}  {({r^2} {{db^{(0)}_{1/2}(r)} \\over {dr}})}},\\eqno(13)$$  \\noindent q is the mass ratio, r is the ratio of disk radius and binary separation and ${b^{(0)}_{1/2}(r)}$ is the Laplace coefficient of order 0 in celestial mechanics (see Brouwer and Clemence 1961, equation 42, Chapter 15). Hirose and Osaki also demonstrated that the tidal instability of accretion disks is caused by the parametric resonance between particle orbits in the disk and the orbiting secondary star with 1:3 period ratio.  Then, from Kepler's third law, $r=(1/3)^{2/3}=0.48$ for the tidal instability radius.  From the Laplace coefficients in Brouwer and Clemence (1961), $a(r=0.48)=0.41$.  Applying this model to the X1916-053 system, for which $w_p / w_{orb}$= 0.0089, the mass ratio is q = 0.022, or the mass of the secondary star $M_2$=0.03$M_{\\odot}$ for the assumed 1.4 $M_{\\odot}$ neutron star. This compares favorably with the value $M_2$=0.016$M_{\\odot}$ derived from the requirement that Roche lobe-filling secondary has $R_{WD}\\sim R_{RL}$.  Thus the secondary is indeed a helium white dwarf, as expected from its binary period.   The analysis in this paper indicates that the x-ray dip period is more likely to be the true orbital period, which supports the SU UMa model.  The only problem may be the stability of the superhump period. The optical period derivative  obtained by the 1987-1996 dataset is $|\\dot{P}|<2.28 \\times 10^{-10}$, whereas the SU UMa stars typically have $\\dot{P} \\sim -(3-10) \\times 10^{-5}$ (Patterson et al. 1993a). However, Patterson et al.(1993b) showed that the range of superhump period derivatives is $|\\dot{P}| \\sim 10^{-3} - 10^{-9}$ (for V795 Her, $|\\dot{P}| <1.7 \\times 10^{-9}$, see Zhang et al. 1991).  AM CVn(=HZ 29), a binary with a white dwarf primary star and low mass helium white dwarf secondary star, is a compact binary system which has behavior similar to X1916-053.  It has two\\footnote{In fact, AM CVn also has a 1011 sec period which perhaps comes from the beat period of the 1028 sec period and a 16.96 hr nodal regression period (Patterson et al. 1993c, Skillman et al. 1999).} periods, 1028 sec and 1050 sec.  Patterson et al. (1993c) discovered the 13.38 hr period, the beat period of the 1028 sec and 1050, in the helium absorption-line profile and suggested that its accretion disk is elliptical and slowly precesses around the white dwarf.  Provencal et al. (1995) and Solheim et al. (1998) analyzed the Whole Earth Telescope (WET) data and discovered the 1051 sec period is very stable with period derivative of $\\dot P(1051s) = +(1.71 \\pm 0.04) \\times 10^{-11}$($\\sim$ 10 years baseline)\\footnote { The 1051 sec period wanders with $\\dot{P}= \\pm 2 \\times 10^{-8}$ (time scale of 4-12 months) but the mean period has not wandered more than 0.1 sec in 35 years, which leads to $|\\dot{P}| < 10^{-10}$ (Skillman et al. 1999).}. They concluded that 1051 sec is the orbital period and 1028 sec is the superhump period which would be opposite to the normal superhump system and not supported by the tidal instability model.  On the other hand, Skillman et al. (1999) found that the period derivative of 1028 sec period is even more stable with $|\\dot P(1028s)|< 2 \\times 10^{-12}$ (over 7 years baseline). Thus, they concluded the shorter period is the orbital period and the longer one is the superhump period, which is comparable with our results for X1916-053.  Both systems thus have very stable long-term superhump periods (${\\dot P}_{sh}$(AM CVn)=$+1.7 \\times 10^{-11}$ and ${\\dot P}_{sh}$(X1916-053) $= P_{opt} \\times {{\\dot{P}}_{opt}} /{P_{opt}} \\sim +8.6 \\times 10^{-11}$, $\\sim10^6$ times smaller than ones from typical superhumps of dwarf novae). Therefore, at least some of the ``superhump CVs'' (e.g. AM CVn, V795 Her\\footnote{These are not true SU UMa (superhump) systems since neither system has been observed to have outbursts.}) have very stable superhump periods comparable to X1916-053.   The very stable superhump period implies a corresponding stable mean mass transfer rate, and thus x-ray luminosity.  This is because conservative mass transfer (assumed) at rate ${\\dot{M}}_2$ implies $\\dot{P}$/$P$ =${\\dot{M}}_2$/$M_2$ (van der Klis et al. 1993) where $M_2$ is the secondary star mass.  Our limit (eq. (8)) implies that ${\\dot{M}}_2 \\leq 6.4 \\times 10^{-9} M_\\odot yr^{-1}$, or $L_x \\sim 0.15 \\dot{M}_2c^2 \\leq 5.4 \\times 10^{37}$ erg/sec, which is a factor of $\\sim$10 larger than the typical system luminosity measured from our RXTE data (Bloser et al. 2000).  Thus, measurement of $\\dot{P}/P\\sim 10^{-10}$, combined with measured variations in $L_x$ on the $\\sim$190d period (see below), might allow a conclusive test of the SU UMa model.   \\subsection{Triple Model Implications}  Although the SU UMa (or superhump) model is the conservative choice, we cannot exclude the triple model (Grindlay et al 1988, Grindlay 1989) since it could also explain the reported (Priedhorsky and Terrell 1984; hereafter PT84) 199d long-term modulation of X1916-053 (cf. section 2).  We have searched for the $\\sim$199d period by an epoch-folding analysis.  We find comparably significant peaks in a $\\chi^2$ vs. period plot at $\\sim$160d and 190d but not at the 198.6$\\pm$1.72d value found by PT84.  Longer epoch analysis for the 199d period is needed, since the $\\sim$4 yr RXTE/ASM data span is short enough that a ``local'' period departure of the Mazeh-Shaham period is possible (Mazeh, private communication). Extended x-ray and optical coverage is also needed to reduce the limits on $\\dot{P}$ for both the x-ray dip and optical periods. Although the predicted (Bailyn 1987) phase glitches for the x-ray dips, and the implied phase clustering of x-ray dips vs. optical (binary) minima, are not found in our phase analysis (except possibly for secondary dips), more simultaneous x-ray and optical observations are needed.  The strongest remaining evidence for the triple model is the surprising clustering of x-ray burst times vs. the times of x-ray dips (cf. Figure ~\\ref{burststat}). This would be expected under the triple model if the enhanced matter in the accretion ring can be accreted very rapidly onto the NS: for bursts to occur within the $\\sim$0.2 phase bin following a dip, the accretion time must be $\\leq$600 sec. This itself would be almost as interesting as the retrograde triple system nature (and its origin; possibly in a globular cluster), since such a surprisingly short accretion timescales implies accretion via an ADAF-like quasi-spherical flow, rather than a dense accretion disk, from the ring down to the NS.  Although the dip vs. burst association appears to be significant, more data are (as usual) needed. Quasi stable burst repetition times have been seen in one other source, GS1826-238 (Cocchi et al 2000), but without the remarkable stability apparent in X1916-053. \\\\   If the bursts are indeed correlated with dips (i.e. follow by $\\sim$0.0-0.2 phase), then under the simple binary and SU UMa model this would imply bursts are preferentially detected when the $\\sim$0.03$M_{\\odot}$ WD secondary is approximately aligned between us and the NS (i.e. when the NS would be eclipsed if the system  were at higher inclination).  We consider two very unlikely possibilities: lensing and beaming.  The WD-NS alignment might suggest gravitational lensing or focusing small bursts on the NS otherwise not detectable.  However, this would not preferentially amplify bursts since the accretion luminosity would be similarly lensed.  The near-eclipse WD-NS alignment could imply burst emission is somehow beamed, perhaps by the WD-NS accretion stream-tube which might confine the burst at its footprint on NS by magnetic pressure.  The lensing or beaming scenarios are both so implausible that confirmation of the burst-dip association would (virtually) require the triple model.   A final test of the triple vs SU UMa models could be made by irrefutably establishing the burst correlations (triple model) or spectroscopically showing that the dip period is the binary period (SU UMa model)."
    },
    "0010/astro-ph0010186_arXiv.txt": {
        "abstract": "A high density of massive dark objects (MDOs), probably supermassive black holes, in the centres of nearby galaxies has been inferred from recent observations. There are various indications that much of the accretion responsible for producing these objects took place in dust-enshrouded active galactic nuclei (AGNs). If so, then measurements of the intensity of background radiation and the source counts in the far-infrared and submillimetre wavebands constrain the temperature of dust in these AGNs.  An additional constraint comes from the hard X-ray background, if this is produced by accretion. One possibility is that the dust shrouds surrounding the accreting AGNs are cold, about $30\\,{\\rm K}$. In this event, the dusty AGNs could be some subset of the population of luminous distant sources discovered at $850\\,\\mu$m using the SCUBA array on the JCMT, as proposed by Almaini, Lawrence and Boyle (1999). An alternative is that the dust shrouds surrounding the accreting AGNs are much hotter ($> 60\\,{\\rm K}$). These values are closer to the dust temperatures of a number of well-studied low-redshift ultraluminous galaxies that are thought to derive their power from accretion.  If the local MDO density is close to the maximum permitted, then cold sources cannot produce this density without the submillimetre background being overproduced if they accrete at high radiative efficiency, and thus a hot population is required. If the dust-enshrouded accretion occured at similar redshifts to that taking place in unobscured optical quasars, then a significant fraction of the far-infrared background radiation measured by $COBE$ at $140\\,\\mu$m, but very little of the submilllimetre background at $850\\,\\mu$m, may have been produced by hot dust-enshrouded AGNs which may have already been seen in recent {\\it Chandra} X-ray surveys. ",
        "introduction": "There is increasing evidence (Magorrian et al.\\ 1998; van der Marel 1999, Ferrarese \\& Merritt 2000, Gebhardt et al.~2000) for massive dark objects (MDOs), probably supermassive black holes (Kormendy \\& Bender 1999), in the centers of most large nearby galaxies. The comoving mass density of these MDOs is higher by at least a factor of two (Salucci et al.~1999), and possibly more (Phinney 1997; Haehnelt, Natarajan \\& Rees 1998), than can be explained by accretion onto optically-luminous AGNs.  Haehnelt et al.\\ (1998) suggested one possible solution to this problem: that much of the accretion is obscured by dust, and that this population of dust-obscured AGNs is missing from optically selected samples (see also Fabian 1999 for a specific model). Dust-enshrouded AGNs are certainly known at high redshift: see Table 1 for some examples, and McMahon et al.\\ (1999) and Benford et al.\\ (1999). However, note that the fraction of their bolometric luminosity that originates from an AGN as opposed to a starburst has not been well established in most cases. In an important recent development, analysis of the hard X-ray background (Fabian \\& Iwasawa 1999; Salucci et al.\\ 1999) has revealed that if the sources responsible for producing the X-ray background are obscured (type-2) quasars, then they may also be the class of sources that are responsible for producing the high local MDO density, as dust and the associated density of molecular and neutral gas is transparent to hard X-rays.  \\begin{table*} \\caption{Examples of dust-enshrouded high-redshift AGNs.} {\\vskip 0.75mm} {$$\\vbox{ \\halign {\\hfil #\\hfil && \\quad \\hfil #\\hfil \\cr \\noalign{\\hrule \\medskip} Object  & Redshift  & Reference  &\\cr \\noalign{\\smallskip \\hrule \\smallskip} \\cr IRAS F10214+4724 & 2.3 & Lacy et al.\\ (1998) and references therein &\\cr QSO H1413+117       & 2.6 & Kneib, Alloin \\& Pello (1998) and references therein  &\\cr SMM J02399$-$0136 & 2.8 & Ivison et al.\\ (1998), Frayer et al.\\ (1998) &\\cr APM 08279+5255 & 3.9 & Irwin et al.\\ (1998), Lewis et al.\\ (1998) &\\cr BR 1202$-$0725 & 4.7 & Isaak et al.\\ (1998), Kawabe et al.\\ (1999) &\\cr SMM J02399$-$0134 & 1.1 & Soucail et al.~(1999), Bautz et al.~(2000) &\\cr \\noalign{\\smallskip \\hrule} \\noalign{\\smallskip}\\cr}}$$} \\end{table*}  Fabian \\& Iwasawa (1999) suggest that the bulk of the energy absorbed at optical, ultraviolet and soft X-ray wavelengths is reradiated at far-infrared and submillimetre wavelengths by obscured AGNs.  We investigate this scenario using constraints from far-infrared and submillimetre source counts and backgrounds. If we assume that interstellar dust is responsible for the absorption and reradiation, then the question becomes: what is the dust temperature of these sources?  Almaini et al.\\ (1999) and Gunn \\& Shanks (2000) recently proposed that the obscured quasars contributing to the X-ray background have cold ($\\leq 40$\\,K) dust shrouds, and that they account for about ten to twenty per cent of the mJy sources detected at 850\\,$\\mu$m using the SCUBA camera at the JCMT (Blain et al.\\ 2000 and references therein). Their motivation for suggesting this temperature comes from observations of high-redshift radio-quiet AGNs (Benford et al.\\ 1999). Interestingly this is the most favoured temperature for the SCUBA sources derived from current measurements of the far-infrared and submillimetre background, source counts and redshift distribution (Blain et al.\\ 1999a,b; Trentham, Blain \\& Goldader 1999).  This idea is very attractive. The dust in type-2 AGNs, which are required to produce the hard X-ray background, is observed directly by SCUBA reradiating away the absorbed ultraviolet and soft X-ray energy in the submillimetre waveband. Two of these SCUBA sources, in the field of the lensing cluster Abell 370 (Smail, Ivison \\& Blain 1997), which are known from optical spectroscopy to contain AGNs, have in fact been seen at X-ray wavelengths with {\\it Chandra} by Bautz et al.~(2000). On the other hand, much of the accretion may take place surrounded by hotter ($ \\geq 60$\\,K) dust shrouds, as would be required by radiative transfer models of the spectral energy distributions (SEDs) of luminous infrared galaxies (Rowan-Robinson 2000 and references therein). These objects would then on the whole have SEDs skewed to wavelengths too short to be seen easily with SCUBA, yet too long to have been seen in ground-based near-infrared surveys.  They may be present in substantial numbers but until recently have been very difficult to detect at all. The first such objects to be seen in a systematic survey were recently described by Wilman, Fabian \\& Gandhi, who found $ISO$ 6.7-$\\mu${\\rm m} and 15-$\\mu${\\rm m} counterparts of two hard X-ray sources detected by {\\it Chandra}. Many local infrared-luminous AGNs, for example the `Warm ULIGs' of Sanders \\& Mirabel (1996), have temperatures close to 60\\,K. Higher temperatures in excess of 100\\,K are observed for a few, rare high-redshift ULIGs, such as P09104+4109 (Kleinmann et al.\\ 1988) and APM 08279+5255 (Lewis et al.\\ 1998), with temperatures in excess of 100\\,K. The fortunate circumstances under which APM 08279+5255 (Irwin et al.~1998) was discovered (the infrared source happens to lie close to the caustic of a lensing galaxy so is highly magnified) are perhaps suggestive that many such objects (without such a large magnification) exist but are as yet unidentified. Even higher dust temperatures (150--200\\,K) are implied by the observed bumps in the SEDs of the most luminous Seyfert galaxies at about 25\\,$\\mu$m (Sanders \\& Mirabel 1996), although the bolometric luminosities of these galaxies are not dominated by these hot components.  Therefore it is worth investigating the constraints on the temperatures of dust-enshrouded AGNs, the observations of which are described in Section\\,2. In Section\\,3 we demonstrate the effect of the dust temperature on both the observable source counts and the efficiency of accretion. We conduct more detailed calculations in Section\\,4. In Section\\,5 we discuss our results and indicate the most promising lines of enquiry for the future. Throughout this paper we assume that $H_0 = 50$\\,km\\,s$^{-1}$\\,Mpc$^{-1}$ and $\\Omega_0 = 1$. ",
        "conclusions": "It is not yet possible to determine a unique description of the SEDs of the dust-enshrouded AGNs that are hypothesized to be responsible for both the hard X-ray background radiation intensity and the density of massive dark objects (MDOs) in the cores of low-redshift galaxies. If the density of MDOs is at the bottom of the current observational range, then the low-temperature dust-enshrouded AGNs described by Almaini et al.\\ (1999), a subset of the high-redshift SCUBA galaxies, can account for all the accretion. However, a hotter population of dust-enshrouded AGNs is required if the MDO density is towards the top of the range permitted by current observations. This second scenario has the advantage of accounting for most of the far-infrared background radiation. New clues have been provided by the lack of strong cross-correlation between SCUBA and $Chandra$ sources, implying that most cold sources, if they are AGNs, are Compton-thick.  There are more hot than cold Compton-thin sources, and so one might expect there to be significant numbers of hot Compton-thick sources seen by neither $Chandra$ nor SCUBA. A hotter population of dust-enshrouded AGNs does indeed seem to be a realistic proposition. Deeper observations at higher X-ray energies with {\\it XMM - Newton} will help to uncover such a population, if it exists.   \\begin{figure*} \\begin{minipage}{170mm} {\\vskip-3.5cm} \\begin{center} \\epsfig{file=fig3.ps, width=18.65cm} \\end{center} {\\vskip-5.7cm} \\caption{The contribution to the submillimetre and far-infrared extragalactic background light at various wavelengths from reprocessed starlight as a function of the mean flux-weighted temperature $T$ of the dust (solid lines). The amount of absorbed starlight is chosen to match the internal extinction corrections to the star-formation history derived by Steidel et al.\\ (1999, Fig.\\,9); these corrections also reproduce the observed value of the baryonic density in stars $\\Omega_* \\approx 0.004$ (Pettini 1999), as we require if the SCUBA sources are powered by AGNs rather than star formation. We assume that the shape of their luminosity function (Fig.\\,5 of that paper) does not change with redshift, that the contribution to the star-formation rate from galaxies with $M_{\\rm AB}$(1700 \\AA) $> -20$ or $M_{\\rm AB}$(1700 \\AA) $< -23$ is negligible, and that the dust emissivity index is 1.5. The dashed lines and errorbars represent observed backgrounds: top left -- Fixsen et al.~(1998); top right -- Schlegel et al.\\ (1998; dot-dashed), Hauser et al.\\ (1998; long-dashed), Lagache et al.\\ (1999; short-dashed); bottom left -- as top right; bottom right -- Finkbeiner et al.~(2000).} \\end{minipage} \\end{figure*}"
    },
    "0010/astro-ph0010523_arXiv.txt": {
        "abstract": "Sharp dips observed in the pulse profiles of three X-ray pulsars (GX~1+4, RX~J0812.4$-$3114 and A~0535+26) have previously been suggested to arise from partial eclipses of the emission region by the accretion column occurring once each rotation period.  We present pulse-phase spectroscopy from {\\it Rossi X-ray Timing Explorer\\/} satellite observations of GX~1+4 and RX~J0812.4$-$3114 which for the first time confirms this interpretation. The dip phase corresponds to the closest approach of the column axis to the line of sight, and the additional optical depth for photons escaping from the column in this direction gives rise to both the decrease in flux and increase in the fitted optical depth measured at this phase.  Analysis of the arrival time of individual dips in GX~1+4 provides the first measurement of azimuthal wandering of a neutron star accretion column. The column longitude varies stochastically with standard deviation 2--6$\\arcdeg$ depending on the source luminosity.  Measurements of the phase width of the dip both from mean pulse profiles and individual eclipses demonstrates that the dip width is proportional to the flux.  The variation is consistent with that expected if the azimuthal extent of the accretion column depends only upon the Keplerian velocity at the inner disc radius, which varies as a consequence of the accretion rate $\\dot{M}$. ",
        "introduction": "Binary X-ray pulsars are neutron stars which accrete stellar material from a companion star \\nocite{white95}(e.g. {White}, {Nagase} \\& {Parmar} 1995). The accretion flow is typically mediated through an accretion disc and accretion columns delineated by the magnetic field of the neutron star ($B\\sim10^{12}$~G at the surface) which terminate at strongly-emitting regions near the magnetic poles \\nocite{gl79a,gl79b}(e.g. {Ghosh} \\& {Lamb} 1979a,b).  The plasma flows along the columns from the inner edge of the accretion disc, where the magnetic field begins to dominate the accretion flow, to the magnetic poles (Fig. \\ref{schem}). The base of the columns are expected to cover an arc-shaped region on the neutron star surface as a consequence of the dipolar magnetic field and the limited set of magnetic field lines along which accretion is energetically favourable.  To date, observational verification of this scenario has not been possible due to instrumental limitations and a lack of theoretical models describing the anisotropy of X-ray emission from the column.  Optical spectroscopic and polarimetric measurements of accreting white dwarf stars in cataclysmic variables \\nocite{warner95}(e.g. {Warner} 1995) provide corroboration, but such techniques are not available for X-ray pulsars where the accretion columns radiate in X-rays rather than the optical and UV bands.  In general there is little observational data available which allow detailed studies of the dynamics of accretion flows.  \\begin{figure} \\epsfxsize=9.0cm % \\epsfbox{nature1.eps} \\caption[nature1.eps]{ Schematic showing the accretion geometry of X-ray pulsars (not to scale). The accreting gas from the companion is thought to form an accretion disc far from the neutron star, which is disrupted at $r_{\\rm A,disc}$ due to the strong (dipolar) magnetic field. Within this radius the accreting matter may only flow along the magnetic field lines, forming two accretion columns terminating at the magnetic poles. Thermal X-ray emission originates from the shaded region at each pole and is Compton scattered by the accretion column plasma above it.  Here the magnetic axis is aligned relative to the neutron star rotation axis by $\\beta$ (the magnetic colatitude). The inclination angle $i$ is such that at phase $\\phi=0.0$ one of the accretion columns will be closely aligned with the line of sight, giving rise to a partial `eclipse' of the shaded emission region by the accretion column. \\label{schem} } \\end{figure}  Sharp (phase width $\\Delta\\phi\\approx0.1$) dips are usually observed in the pulse profiles of the X-ray pulsars A~0535+26, RX~J0812.4$-$3114 and GX~1+4.  These dips have previously been attributed to interactions of the accretion column and the emission region \\nocite{cem98,rei99,gil00}({\\v{C}emelji\\'c} \\& {Bulik} 1998; {Reig} \\& {Roche} 1999; {Giles} {et~al.} 2000) although supporting arguments have been scarce.  A~0535+26 is observed only in outburst, when the X-ray emission is modulated on a period of $P_{\\rm spin}\\approx103.5$~s \\nocite{neg00}(e.g. {Negueruela} {et~al.} 2000). RX J0812.4$-$3114, the X-ray counterpart to the B0-1 III-IV star LS~992 was identified from cross-correlation of {\\it ROSAT} galactic plane survey data with SIMBAD OB star catalogues \\nocite{motch97}({Motch} {et~al.} 1997).  Hourly flux variations, as well as X-ray pulsations at $P\\approx31.9$~s were discovered during a {\\it Rossi X-ray Timing Explorer} ({\\it RXTE\\/}) observation of the source on 1998 February \\nocite{rei99}({Reig} \\& {Roche} 1999).  The X-ray behaviour of these Be/X-ray binaries is strongly dependent upon the orbital parameters, and the neutron star spin periods $P_{\\rm spin}$ show a strong correlation with the orbital period $P_{\\rm orb}$ \\nocite{corbet86}({Corbet} 1986).  The $\\approx81$~d periodicity suggested by {\\it RXTE\\/} All-Sky Monitor (ASM) data obtained between 1998--1999 \\nocite{corbet00}({Corbet} \\& {Peele} 2000) is consistent with this correlation and probably represents the binary period.  The distance was estimated at 9~kpc based on the companion spectral type and interstellar reddening, with a likely uncertainty of 50 per cent \\nocite{motch97}({Motch} {et~al.} 1997).  The pulse profile from the source is strongly asymmetric, with a sharp dip forming the primary minimum.  The energy spectrum measured by {\\it RXTE}\\/ was fairly typical for X-ray pulsars, and could be adequately fit using a power law model with exponential cutoff \\nocite{rei99}({Reig} \\& {Roche} 1999).  GX~1+4 is the most persistently bright of the three, and posesses the slowest period.  Measurements by {\\it RXTE\\/} and the Burst and Transient Source Experiment (BATSE) aboard the {\\it Compton Gamma-Ray Observatory} \\nocite{zhang95}({Zhang} {et~al.} 1995) between 1996--7 found $P_{\\rm spin}=123.5$--125.5~s (Galloway \\& Greenhill 2001, in preparation).  The companion to the X-ray source is the 17th magnitude M6 giant V2116 Ophiuchus \\nocite{dav77}({Davidsen}, {Malina} \\& {Bowyer} 1977), the only known symbiotic binary with a neutron star as the energy source for the emission line nebula \\nocite{symcat00}({Belczy{\\'n}ski} {et~al.} 2000). Regular episodes of relatively rapid spin-up measured from BATSE monitoring of the source suggest an orbital period of $\\approx 304$~d \\nocite{per99}({Pereira}, {Braga} \\& {Jablonski} 1999). The distance to the source is thought to be 3--15~kpc \\nocite{chak97:opt}({Chakrabarty} \\& {Roche} 1997).  Several estimates of the magnetic field strength are in the range 2--3$\\times 10^{13}$~G \\nocite{beu84,dot89,mony91,gre93,cui97}({Beurle} {et~al.} 1984; {Dotani} {et~al.} 1989; {Mony} {et~al.} 1991; {Greenhill} {et~al.} 1993; {Cui} 1997).  The dips in this extremely variable source are clearly visible in the X-ray lightcurve when the source is bright and the time resolution is sufficient, and are present in the folded pulse profiles even when the luminosity falls almost to the detection limit \\nocite{gil00}({Giles} {et~al.} 2000). When the source is bright the dips are broad and generally do not reach zero, while at lower luminosities they are narrower and emission at the dip phase is consistent with the background level.  The phase-averaged X-ray spectra from the source are consistent with an approximately thermal spectrum (with temperature $T_0\\approx1.3$~keV) modified by Compton scattering within hotter material ($T_{\\rm e}=6$--10~keV) with characteristic optical depth $\\tau=2$--6 \\nocite{myphd:a}({Galloway} 2000b).  The thermal spectrum is thought to arise from the region of the polar cap at the base of the accretion column, with Compton scattering occurring in the hotter plasma above. ",
        "conclusions": "RX~J0812.4$-$3114 provides another example of an X-ray pulsar after GX~1+4 whose mean spectra are broadly consistent with Comptonisation continuum components.  In contrast to GX~1+4, the properties of this source are probably typical of many others, since Be/X-ray binaries make up a significant proportion of all known X-ray pulsars.  The phase-averaged spectral fit parameters fall within the ranges expected for neutron star accretors \\nocite{myphd:a}({Galloway} 2000b). We suggest that Comptonisation continuum components are excellent candidates for fits to spectra from X-ray pulsars.  The sharp dip in the pulse profiles is clearly associated with an increase in the scattering optical depth $\\tau$ both for GX~1+4 (at distinctly different fluxes and phase-averaged spectral conditions) and also for RX~J0812.4$-$3114. It is true that the estimated significance is only convincing for the brightest observation of GX~1+4.  We note however that the confidence interval limits for the pulse phase spectral fit parameters were determined by projecting the full $p$-dimensional confidence region (where $p$ is the number of free parameters in the model) onto the axis corresponding to each parameter. This provides a conservative estimate of the confidence intervals for individual parameters \\nocite{lmb76}({Lampton}, {Margon} \\& {Bowyer} 1976) and so the true significance of the spectral variation may be much greater.  Variation in $\\tau$ with pulse phase is possible if the scattering region lacks rotational symmetry, e.g. is extended in one direction preferentially. In that case the maximum possible optical depth will be obtained when the line of sight to the observer coincides with the longest dimension of the scattering region.  For the accretion column, the longest dimension is clearly along the column axis. Thus, we identify the sharp dips as `eclipses' resulting from an approximate alignment of the accretion column with the line of sight, occuring once each rotation period ($\\phi=0.0$ in Fig. \\ref{schem}).  Absorption dips in the orbital lightcurves of some polars, which are white dwarf stars also accreting through a magnetically confined column \\nocite{grei98}(e.g. RX~J1802.1+1804; {Greiner}, {Remillard} \\& {Motch} 1998), have previously been associated with a similar effect. The spectral variation in the latter source during the eclipse suggests a warm absorber, in good agreement with the results described here.  Clearly such interactions require a fortuitous combination of the magnetic colatitude $\\beta$ (the angle between the neutron star spin axis and the magnetic axis) and the inclination angle $i$ (between the spin axis and the line of sight).  Since the base of the accretion column is offset from the magnetic pole, and the column meets the surface at an angle with respect to the local normal, $i$ must substantially greater than $\\beta$ for eclipses to occur. The column orientation also depends upon the inner disc radius $r_{\\rm A,disc}$ since this defines which field lines connect the boundary region to the neutron star.  The small number of sources exhibiting such eclipses (3 of more than 70 known pulsars) appears to be consistent with the relative unlikelihood of such an agreement; no reliable estimates of $i$ and $\\beta$ have been made for the three pulsars described which might rule out this possibility.  While numerical simulations based on unmagnetized Compton scattering in a simplified emission region support this interpretation \\nocite{gal00:beam}({Galloway} \\& {Wu} 2001) it is important to note that the strong magnetic field present in the accretion column will significantly affect the scattering cross section \\nocite{dh86}({Daugherty} \\& {Harding} 1986).  The cross section $\\sigma(\\nu,\\theta)$ will depend on both the photon energy $\\nu$ and the angle between the propagation direction and the local magnetic field vector $\\theta$.  For GX~1+4, the estimated magnetic field strength implies an energy for the fundamental cyclotron resonance of $h\\nu_{\\rm cyc}\\ga200$~keV, much greater than the PCA energy band.  Thus for a particular photon energy, $\\sigma_{\\nu,\\theta}$ will reach a minimum for photons propagating along the magnetic field lines ($\\theta=0$). In the absence of anisotropies in the distribution of matter this would result in enhanced emission along the magnetic axis.  This effect will clearly compete with the decrease in emission due to the additional optical depth experienced by photons propagating along the accretion column.  However $\\sigma_{\\nu,\\theta}$ reaches a minimum value (which depends upon the magnetic field strength and the photon energy) for photons propagating exactly along the magnetic field lines.  The increase in optical depth is instead limited only by the geometry and the curvature of the accretion column.  Despite magnetic anisotropy effects it should still be possible to get a significant drop in emission due to accretion column eclipses.  The observation of accretion column eclipses in GX~1+4 provides a unique opportunity to study accretion dynamics under the influence of a strong magnetic field.  The variability in the observed phase of individual dips suggests that the column is wandering stochastically in longitude by 2--6$^{\\circ}$, depending upon the source luminosity.  It is possible that the dip timing residuals are a consequence of variations in the $P_{\\rm spin}$ itself, although we consider this unlikely.  Accreting material typically posesses a large specific angular momentum with respect to the neutron star, and in combination with magnetic torques transmitted from the accretion disc via the magnetic field may induce systematic period evolution of more than 2 per cent per year.  However the variations seen in the {\\it RXTE\\/} observations are typically of the same order in just one pulse period, and in addition there appears to be no cumulative drift in the pulse period on time--scales of tens of minutes or more as a consequence of the timing residuals.  The accretion column is delineated by the `bundle' of magnetic field lines which bound it, and it seems more likely that the position of the column is varying perhaps as a consequence of accretion along a varying set of field lines, resulting in a change in its apparent position.  The dip phase width $w_{\\rm d}$ measured from both the mean pulse profiles and the average of the individual dip fits from each observation show a significant correlation with X-ray luminosity $L_{\\rm X}$.  This parameter is a measure of the width of the accretion column, which ultimately depends upon the azimuthal extent of the accretion stream at the inner accretion disc boundary (see Fig. \\ref{schem}).  The detailed physics of this boundary region are poorly understood, and so we assume that the entrainment of disc material to the column takes place at a rate which is independent of the plasma density in the disc. The column will then be wider at higher $\\dot{M}$ due to the increased velocity of disc plasma as the inner disc edge moves closer to the neutron star. The inner disc radius is determined by a balance of the magnetic stress and the ram pressure \\begin{eqnarray} r_{\\rm A,disc}&\\approx & 0.5\\times2^{-3/7}\\mu^{4/7}(GM_*)^{- 1/7}\\dot{M}^{-2/7} \\nonumber \\\\ & = & 2.4\\times10^8 B_{12}^{4/7} \\dot{M}_{16}^{-2/7}\\ {\\rm cm} \\label{alfven} \\end{eqnarray} where $\\mu$ is the magnetic dipole moment, $B_{12}$ is the corresponding surface magnetic field strength in units of $10^{12}$~G, $M_*=1.5\\ {\\rm M}_{\\sun}$, $R_*=10$~km the mass and radius of the neutron star and $\\dot{M}=\\dot{M}_{16}\\times10^{16}\\ {\\rm g\\,s^{-1}}$ the accretion rate \\nocite{gl79a,gl79b}({Ghosh} \\& {Lamb} 1979a,b).  Assuming that the plasma in the boundary region is still flowing tangentially at approximately the Keplerian velocity $v_{\\rm K}=\\sqrt{GM_*/r}$ corresponding to stable circular orbits, we have \\begin{equation} w_{\\rm d} \\propto \\frac{v_{\\rm K}}{r_{\\rm A,disc}} \\propto L_{\\rm X}^{3/7} \\label{wdprop}. \\end{equation} The agreement between this relationship and the $w_{\\rm d}$ measured from the individual dips is encouraging (Fig. \\ref{column}b), although the error bars on several points are large indicating significant variation of dip width for individual observations.  We note that the dip width measured for RX~J0812.4$-$3114 on 1999 Mar 25 was $w_{\\rm d}\\approx0.05$, comparable to that measured in GX~1+4 at similar luminosity levels. If the conditions in the disc--magnetosphere boundary layer are similar in both pulsars, and relative bolometric corrections are negligible, this implies that the magnetic moments are similar and that the field strength for RX~J0812.4$-$3114 is also around 2--3$\\times10^{13}$~G. That these two pulsars, both with atypically strong magnetic fields, also posess an unusual alignment permitting accretion column eclipses seems rather unlikely however.  Since the dips are associated with interactions with the accretion column, they are expected to always be present so long as substantial amounts of material are accreting (i.e. $\\dot{M}\\neq0$). The {\\it RXTE\\/} observations of GX~1+4 span more than two orders of magnitude in $L_{\\rm X}$, and dips were observed in the mean pulse profiles of every observation except one.  During a short {\\it RXTE\\/} observation on 1997 May 17 , the dips were weak or absent in the lightcurve even though the phase-averaged countrate from the source was $\\approx 150\\ {\\rm count\\,s}^{-1}$, still well above the background which is typically $100\\ {\\rm count\\,s}^{-1}$.  The X-ray spectrum measured at this time was unusually hard, with Compton scattering optical depth $\\tau\\approx19$; the pulse-phase spectral variation was also substantially different from that measured at other times \\nocite{myphd:b}({Galloway} 2000a). Both the disappearance of the dip and the spectral measurements indicate a distinctly different accretion regime than is normal for this source, in which the column is no longer present in the usual sense.  This work has revealed some important new phenomena likely to be common to other X-ray pulsars.  Further observational and theoretical development appears promising, in particular where the unusual properties of sources like GX~1+4 and RX~J0812.4$-$3114 may be exploited.  The influence of the longitudinal variation in column position on the net torque transmitted to the neutron star may be non-negligible. This may affect the long-term period evolution in persistent pulsars, which is generally rather poorly understood \\nocite{bil97}(e.g. {Bildsten} {et~al.} 1997).  Measurements of the dip (and hence column) profile may provide more detailed information regarding the plasma uptake to the accretion column at the inner disc radius. Finally, the suggestion of disruption of the accretion column coupled with distinct spectral states at high and low luminosities \\nocite{myphd:a}({Galloway} 2000b) provides evidence for significantly more complex flow dynamics than are predicted by current theory."
    },
    "0010/astro-ph0010009_arXiv.txt": {
        "abstract": "We present observational evidences that dust in the circumnuclear region of AGNs has different properties than in the Galactic diffuse interstellar medium. By comparing the reddening of  optical and infrared broad lines and the X-ray absorbing column density we find that the $\\rm E_{B-V}/N_H$ ratio is nearly always lower than Galactic by a factor ranging from $\\sim$3 up to $\\sim$100. Other observational results indicate that the $\\rm A_V/N_H$ ratio is significantly lower than Galactic in various classes of AGNs including intermediate type 1.8--1.9 Seyferts, hard X--ray selected and radio selected quasars, broad absorption line QSOs and grism selected QSOs. The lack of prominent absorption features at 9.7$\\mu$m (silicates) and at 2175\\AA \\ (carbon dip) in the spectra of Seyfert 2s and of reddened Seyfert 1s, respectively, add further evidence  for dust in the circumnuclear region of AGNs being different from Galactic.  These observational results indicate that the dust composition in the circumnuclear region of AGNs could be dominated by large grains, which make the extinction curve flatter, featureless and are responsible for the reduction of the $\\rm E_{B-V}/N_H$ and $\\rm A_V/N_H$ ratios.  Regardless of the physical origin of these phenomena, the reduced dust absorption with respect to what expected from the gaseous column density should warn about a mismatch between the optical and the X-ray classification of the active galactic nuclei in terms of their obscuration. ",
        "introduction": "The properties of the circumnuclear gas are a key issue to understand the physics of Active Galactic Nuclei (AGNs). In particular, gas obscuration has important consequences on the classification of AGNs (Antonucci 1993), on their infrared emission (Granato et al. 1997, Pier \\& Krolik 1993) and also on the X-ray background (Setti \\& Woltjer 1989). AGNs are, optically-wise, divided in two main classes: type 1 AGNs, showing broad permitted emission lines, and type 2 AGNs, which only show narrow emission lines. The Unified Model assumes that AGNs of both classes host the same kind of nuclear engine and ascribes their differences solely to orientation effects with respect to an obscuring gaseous medium, possibly arranged in a torus-like geometry. For those lines of sights intercepting the obscuring torus, both the Broad Line Region (BLR, $< 1$ pc in size) and the nuclear engine are obscured and only the much more extended Narrow Line Region (NLR) can be observed. This model has gained success from a large number of observational tests (see Antonucci 1993 for a review). In particular, X-ray observations have supported the unified scenario by discovering large columns of absorbing gas in type 2 AGNs (eg. Awaki et al. 1991, Turner et al. 1997, Maiolino et al. 1998, Risaliti et al. 1999). Also, spectroscopic observations in the infrared, where dust absorption is greatly reduced, detected broad permitted lines in several AGNs which are classified as type 2 in the optical.  However, the properties of the absorbing medium result to be more complex with a more quantitative analysis. Maccacaro et al. (1982) first noted, in a few AGNs, that the dust reddening affecting the BLR is significantly lower than expected from the N$_H$ measured in the X-rays, assuming a Galactic standard extinction curve and dust-to-gas ratio. Indications for a low value of $\\rm A_V/N_H$ in the obscuring torus were also found by Granato et al. (1997) by modelling the IR emission of AGNs; they ascribed the low $\\rm A_V/N_H$ ratio to the sublimation of dust at the inner face of the torus.  The goal of this paper is to observationally verify the low $\\rm E_{B-V}/N_H$ and low $\\rm A_V/N_H$ phenomena with a higher confidence (sects. 2 and 3). We also investigate dust spectral signatures which directly probe the properties of dust grains in the circumnuclear region of AGNs. Most of the interpretation of these observational effects is addressed in a companion paper (Maiolino et al. 2000b, paper II). Finally, note that in this paper we will often distinguish the $\\rm E_{B-V}/N_H$ and the $\\rm A_V/N_H$ ratios since, as discussed in paper II, in the circumnuclear region of AGNs dust reddening and obscuration are not necessary tied by the Galactic standard relation ($\\rm A_V/E_{B-V}=3.1$).  Throughout this paper we will assume a cosmology with $\\rm H_0=65$ and $\\rm q_0=0.5$. ",
        "conclusions": "We have reported various observational evidences indicating that the dust reddening and absorption of the nuclear region of AGNs is generally much lower than the values expected from the gaseous column density measured in the X-rays, if a standard Galactic dust--to--gas ratio and extinction curve are assumed.  Quantitatively, the most convincing argument supporting a low $\\rm E_{B-V}/N_H$ ratio is the comparison between the reddening inferred from the optical and infrared broad line ratios and the N$_H$ derived from the hard X-rays for a sample of 20 objects. With the exception of three low luminosity AGNs ($\\rm L_X < 10^{42} erg~s^{-1}$), whose physics might differ from Seyfert-- and QSO--like luminosity systems, all the objects appear characterized by an $\\rm E_{B-V}/N_H$ ratio sistematically lower than the Galactic standard value by a factor ranging from $\\sim$3 up to $\\sim$100.  Also, we have presented additional evidences suggesting a reduced $\\rm A_V/N_H$ ratio in AGNs. The presence of substantial gaseous absorbing column ($\\rm N_H \\sim$ a few times $10^{22}$) in intermediate type 1.8--1.9 Seyferts contrasts with their optical appearance, but can be reconciled by a low $\\rm A_V/N_H$ ratio. A more extreme case of this effect is observed in a number of type 1 AGNs (mostly QSOs) whose X-ray spectrum shows evidence for substantial gaseous absorption despite their optical unabsorbed appearance. Two classes of type 1 AGNs which seems to be nearly systematically affected by this phenomenon (i.e. significant X-ray absorption but little, or no, optical absorption) are the Broad Absorption Line QSOs and the grism selected QSOs.  The samples used are probably affected by selection effects, which are discussed in the body of the paper, and therefore cannot be considered as representative of the whole population of AGNs. However, we can certainly state that at least a sub-population of the AGNs is characterized by a low $\\rm E_{B-V}/N_H$ or $\\rm A_V/N_H$ with respect to the Galactic value. Also, we could {\\it not} find evidence for a Galactic standard $\\rm E_{B-V}/N_H$ or $\\rm A_V/N_H$ ratio in nearly any object.  We presented additional evidences indicating that the properties of dust grains in the circumnuclear region of AGNs are different with respect to the Galactic diffuse interstellar medium:\\\\ -- although type 2 Seyfert nuclei appear significantly absorbed by dust in the mid--IR, their average ISO spectrum does not show evidence for the silicate absorption feature at 9.7$\\mu$m which, instead, is expected to be very deep in case of heavy absorption;\\\\ -- some type 1 Seyferts which appear affected by some reddening do not show evidence for the carbon dip at 2175\\AA , while according to the Galactic standard extinction curve this absorption feature should be prominent in their UV spectra.\\\\ Both these observational evidences suggest that dust in the circumnuclear region of AGNs is depleted of the small grains ($<3\\mu$m and 100-200\\AA \\ respectively) which are responsible for these absorption features. A dust grain distribution biased in favor of large grains would also make the extinction curve flatter (Laor \\& Draine 1993). If the bias for large grains is due to coagulation this would also explain the reduced $\\rm E_{B-V}/N_H$ and $\\rm A_V/N_H$ (Kim \\& Martin 1996). A more detailed discussion on the interpretation of these results is given in a companion paper (Maiolino et al. 2000b, paper II).  Regardless of the intepretation of these observational phenomena, these results have important consequences on the unified theories of AGNs. The finding that the dust absorption in generally significanlty lower with respect to what expected from the gaseous column density implies that for several AGNs the optical classification might be de-coupled from the X-ray classification. In particular, AGNs which appear obscured (type 2) in the X-rays might appear as relatively unobscured (type 1) in the optical."
    },
    "0010/gr-qc0010087_arXiv.txt": {
        "abstract": "The important studies of Peebles, and Bond and Efstathiou have led to the formula $C_{\\ell} = const/[ \\ell(\\ell +1) ]$ aimed at describing the lower order multipoles of the CMBR temperature variations caused by density perturbations with the flat spectrum. Clearly, this formula requires amendments, as it predicts an infinitely large monopole $C_{0}$, and a dipole moment $C_{1}$ only $6/2$ times larger than the quadrupole $C_{2}$, both predictions in conflict with observations. We restore the terms omitted in the course of the derivation of this formula, and arrive at a new expression. According to the corrected formula, the monopole moment is finite and small, while the dipole moment is sensitive to short-wavelength perturbations, and numerically much larger than the quadrupole, as one would expect on physical grounds. At the same time, the function $\\ell (\\ell +1)C_{\\ell}$ deviates from a horizontal line and grows with $\\ell$, for $\\ell \\geq 2$. We show that the inclusion of the modulating (transfer) function terminates the growth and forms the first peak, recently observed. We fit the theoretical curves to the position and height of the first peak, as well as to the observed dipole, varying three parameters: red-shift at decoupling, red-shift at matter-radiation equality, and slope of the primordial spectrum. It appears that there is always a deficit, as compared with the COBE observations, at small multipoles, $\\ell \\sim 10$. We demonstrate that a reasonable and theoretically expected amount of gravitational waves bridges this gap at small multipoles, leaving the other fits as good as before. We show that the observationally acceptable models permit somewhat `blue' primordial spectra. This allows one to avoid the infra-red divergence of cosmological perturbations, which is otherwise present. ",
        "introduction": "The important studies of Peebles \\cite{Peeb,Pebook} and Bond and Efstathiou \\cite{BondEf,Efst} have led to the formula for the multipole moments of the CMB radiation, $C_{\\ell}$, caused by small density perturbations with flat spectrum (also known as Harrison--Zel'dovich--Peebles, or scale-invariant spectrum; spectral index ${\\mathrm{n}}=1$): \\begin{equation} C_{\\ell} = \\frac{const}{\\ell (\\ell +1)}. \\label{eq:one} \\end{equation} This formula aims to describe the lower order multipoles of the CMBR variations in a spatially flat universe. The general acceptance of this result is reflected in the fact that the $C_{\\ell}$ distributions are usually plotted in terms of the function $\\ell(\\ell +1)C_{\\ell}$ versus $\\ell$. In these units, formula (\\ref{eq:one}) is conveniently plotted as a horizontal line.  Clearly, formula (\\ref{eq:one}) was derived under some assumptions, and this derivation requires amendments. Indeed, let us start from the monopole moment $C_{0}$, setting $\\ell =0$ in this formula. Expression (\\ref{eq:one}) predicts infinity for $C_{0}$, in conflict with observations. Surely, it would be difficult to observationally distinguish a monopole component caused by perturbations from the isotropic unperturbed temperature $T_{0}$, but it is strange that a theoretical formula gives infinity for an effect produced by small perturbations. Let us now consider the dipole, $\\ell =1$, and the quadrupole, $\\ell=2$, moments. According to (\\ref{eq:one}), the quantities $C_{1}$ and $C_{2}$ are numerically of the same order of magnitude, $C_{1} = (6/2) C_{2}$. However, we know from observations that the observed dipole variation of the temperature is about 100 times larger than the observed quadrupole variation.  The predictions of formula (\\ref{eq:one}) also seem to disagree with independent theoretical studies. According to Grishchuk \\& Zel'dovich \\shortcite{GriZe}, in the limit of long-wavelength perturbations, which normally provide dominant contributions to the $\\ell =0$ and $\\ell= 2$ multipoles, the monopole and quadrupole variations of the CMBR temperature should be small and approximately equal. At the same time, the dipole variation should be suppressed, as compared with the monopole and quadrupole, in proportion to the ratio $l_{{\\mathrm{H}}}/ \\lambda$, where $l_{{\\mathrm{H}}}$ is the Hubble radius and $\\lambda$ is the perturbation's wavelength, $\\lambda \\gg l_{{\\mathrm{H}}} $. The multipole moments $C_{\\ell}$, being quadratic in the temperature variations, contain this factor in the second power. The factor $l_{{\\mathrm{H}}}/\\lambda$ becomes much larger than 1 in the opposite limit of short-wavelength perturbations, $\\lambda \\ll l_{{\\mathrm{H}}}$. The dipole component is, thus, expected to be very sensitive to short waves, and its value should eventually exceed the values of the monopole and quadrupole components, as soon as the contribution of short-wavelength perturbations is also included. However, formula (\\ref{eq:one}) suggests results differing from these conclusions. The discrepancy is especially troublesome, as the derivation of (\\ref{eq:one}) seems to have taken into account the perturbations of all wavelengths. Indeed, this formula arises from the table integral (discussed below in Section \\ref{sec-THIRD}) \\[ \\int_{0}^{\\infty}[j_{\\ell}(n\\xi)]^2 \\frac{{\\mathrm{d}}n}{n} = \\frac{1}{2\\ell (\\ell +1)},\\] where the integration over wave-numbers $n$ ranges formally from zero to infinity. Therefore, according to this integral, the very short-wavelength perturbations have also been taken into account in the calculation of all $C_{\\ell}$'s, including the dipole $C_{1}$.  The demonstrated deficiencies of formula (\\ref{eq:one}) for $\\ell =0$ and $\\ell =1$ indicate that there could also be significant deviations from the values suggested by (\\ref{eq:one}) in the case of $\\ell=2$ and higher order multipoles. This consideration has prompted us to reanalyse the derivation of formula (\\ref{eq:one}). In Section \\ref{sec-FIRST} we introduce the necessary notations and remind the reader the basics of the Sachs--Wolfe result \\cite{SaWo}. For the purpose of later comparison with observations, we consider primordial spectra with arbitrary spectral indices, so that the flat spectrum ${\\mathrm{n}} = 1$ is simply a particular case. Section \\ref{sec-SECOND} summarises the statistical assumptions and defines the multipole moments. In Section \\ref{sec-THIRD} we identify the terms that were dropped out in the course of the original derivation of formula (\\ref{eq:one}). We show the importance of these terms and restore them. After this correction, the numerical values of the monopole and quadrupole moments become comparable to each other, while the dipole moment becomes sensitive to short-wavelength perturbations, as it should. At the same time, the correct formula for $C_{\\ell}$ requires the function $\\ell (\\ell+1)C_{\\ell}$ to deviate from the horizontal line and grow with $\\ell$ for $\\ell \\geq 2$. Formally, this growth would be unlimited if one were allowed to use the primordial power-law spectrum for arbitrarily large values of the wavenumber $n$. However, it is known that the primordial power spectrum of density perturbations should experience a turnover, caused by the transition of the cosmological expansion from the radiation-dominated to the matter-dominated regime \\cite{ZelNo,Pebook}. By the time of decoupling, the primordial spectrum gets modified. Mathematically, the original power spectrum gets multiplied by the modulating function (transfer function). The modulating function is practically unity in the spectral region of long-wavelength perturbations, but it strongly suppresses the power in short-wavelength perturbations. As a result, the growth of the function $\\ell (\\ell+1)C_{\\ell}$ terminates and is being followed by its decline at multipoles, roughly, around $\\ell=200$. This causes the function $\\ell (\\ell+1) C_{\\ell}$ to form the first peak, recently observed \\cite{Hana,deBer,Maus}. The following (less prominent) peaks of the modulating function are responsible for the further peaks of the function $\\ell (\\ell +1)C_{\\ell}$. In Section \\ref{sec-FOURTH} we analyse in detail the form of the modulating function and derive the possible (theoretical) positions and heights of the first peak. We think that the origin of the subsequent peaks in the $\\ell$-space is due to the modulation of the metric power spectrum of density perturbations (oscillations in the $n$-space). This modulation, in its turn, entirely arises due to the standing-wave character of the initial density perturbations. Travelling sound waves in the radiation-dominated era would not have produced Sakharov oscillations at the time of decoupling, as was explained long ago (see Section 11.6 in Zel'dovich \\& Novikov \\shortcite{ZelNo}). The recent paper \\cite{Miller} may be providing support to this interpretation of the peaks. However, the origin and position of secondary peaks is not the subject of this paper, so we leave the door open for alternative interpretations. The full computation of all multipoles is presented in Section \\ref{sec-FIFTH}. We replace the existing uncertainties about the matter content of the Universe by the two phenomenological parameters: the decoupling red-shift $z_{{\\mathrm{dec}}}$ and the matter-radiation equality red-shift $z_{{\\mathrm{eq}}}$. We also vary the third parameter, $\\beta$, which characterises the spectral slope of the primordial perturbations. (The flat spectrum ${\\mathrm{n}} = 1$ corresponds to $\\beta = -2$.) We show how the distributions of multipole moments depend on the choice of these parameters. Since we concentrate on anisotropies of gravitational origin, the chemical composition of matter in the Universe is unimportant, as long as the perturbations can be described by solutions to the perturbed Einstein equations at the radiation-dominated stage (equation of state $p = \\epsilon/3$) followed by the matter-dominated stage (equation of state $p=0$). Section \\ref{sec-SIXTH} gives a preliminary comparison of the calculated multipole moments (including the dipole) with the available CMBR observations. We require the theoretical (statistical) dipole, as well as the location and the height of the first peak, to be as close as possible to their observed values. It appears, then, that the values of the COBE-observed \\cite{Smoot} multipoles, around $\\ell =10$, are somewhat larger than the theoretical curves for $\\ell (\\ell +1)C_{\\ell}$ would require. The, so far neglected, intrinsic anisotropies (temperature variations at the last scattering surface) can only increase the height of the first peak making this discrepancy larger. The most natural way of removing this discrepancy is to take into account gravitational waves. They mostly contribute to the multipoles from $\\ell=2$ to approximately $\\ell=30$, without changing the dipole, and not seriously affecting the location and height of the peak. Section \\ref{sec-SEVENTH} describes the contribution of gravitational waves. In effect, we calculate the amount of gravitational waves needed in order to raise the `plateau' at lower $\\ell$'s to the actually observed level. In Section \\ref{sec-EIGHTH} we again compare the theory with observations, this time including gravitational waves, and discuss the values of the parameters $z_{{\\mathrm{dec}}}$, $z_{{\\mathrm{eq}}}$, $\\beta$ that produce distributions $\\ell (\\ell +1) C_{\\ell}$ in better agreement with CMBR observations. We argue that a slightly larger amount of gravitational waves allows the primordial spectrum to be somewhat `blue' (${\\mathrm{n}} > 1$ or $\\beta > -2$) instead of being flat. The $C_{\\ell}$'s produced by such `blue' spectra are still consistent with the discussed observations. We consider this fact as a satisfactory conclusion, since, from the theoretical standpoint, the flat spectrum and `red' spectra (${\\mathrm{n}} < 1$ or $\\beta < -2$) are undesirable. Indeed, the mean square values of the metric (gravitational field) fluctuations are logarithmically divergent for the ${\\mathrm{n}} = 1$ case and power-law divergent for all ${\\mathrm{n}} < 1$, in the limit of long waves. The removal of these divergences would require extra assumptions. Appendix A contains some technical details of the derivation of the corrected formula for $C_{\\ell}$. ",
        "conclusions": "\\label{sec-EIGHTH}  We denote by $C^{{\\mathrm{dp}}}_{\\ell}$ and $C^{{\\mathrm{gw}}}_{\\ell}$ the multipole moments induced by density perturbations and gravitational waves, respectively. $C^{{\\mathrm{dp}}}_{\\ell}$ is given by (\\ref{eq:Clfinal}), whereas $C^{{\\mathrm{gw}}}_{\\ell}$ is given by (\\ref{eq:Clgw}). We introduce a numerical parameter $\\varpi$ in such a way that the total $\\ell(\\ell+1)C_{\\ell}$ is given by \\begin{equation} \\ell(\\ell+1)C_{\\ell} = \\ell(\\ell+1)[C^{{\\mathrm{gw}}}_{\\ell} + \\varpi \\cdot C^{{\\mathrm{dp}}}_{\\ell}] . \\label{eq:defvarpi} \\end{equation} The value of $\\varpi$ would be strictly 1, if one could guarantee that the idealised model that has led to formulae (\\ref{eq:Clfinal}), (\\ref{eq:Clgw}) has accurately taken into account all physical processes that might have affected density perturbations and gravitational waves throughout their entire evolution, from the moment they existed in a vacuum quantum state and up to the era of decoupling of CMB radiation from matter. Since one may have legitimate doubts in such accuracy, the correction parameter $\\varpi$ may be thought of as a measure of a necessary modification (but only in the form of an overall numerical coefficient) to our theoretical framework.  As was shown in Section \\ref{sec-SEVENTH}, gravitational waves do not contribute to the dipole, and do not practically contribute to the peak. Therefore, we require the quantity $\\varpi \\cdot \\ell(\\ell+1)C^{{\\mathrm{dp}}}_{\\ell}|_{{\\mathrm{peak}}}$ to be equal to the characteristic value (\\ref{eq:obspeak}). At the same time, we require $\\ell(\\ell+1)C_{\\ell}|_{\\ell=10}$, defined by setting $\\ell =10$ in (\\ref{eq:defvarpi}), to be equal to the characteristic value (\\ref{eq:obs10}). This comparison revises the ratio $\\Lambda= (l_{{\\mathrm{Pl}}}/l_{0})^2$ and determines the parameter $\\varpi$. We then construct the ratio $\\ell(\\ell+1)C^{{\\mathrm{gw}}}_{\\ell}|_{\\ell=10}/(\\varpi \\cdot \\ell(\\ell+1)C_{\\ell}^{{\\mathrm{dp}}}|_{\\ell=10})= C^{{\\mathrm{gw}}}_{10}/(\\varpi \\cdot C^{{\\mathrm{dp}}}_{10})$, which shows the relative contributions of gravitational waves and density perturbations to the small multipoles. Finally, we compare the value $\\varpi \\cdot C_{1}$ with the observed value (\\ref{eq:obs1}) of the dipole. The results of this analysis are shown in table \\ref{fullcomparison}, while a visual representation is given in figure \\ref{compinclgw}.  \\begin{figure}% \\centerline{\\epsfxsize=15cm \\epsfbox{fig14.eps}} \\caption{Comparison of the theoretical model, including density perturbations and gravitational waves, with observations. The dotted line shows the contribution of the corrected density perturbations alone, the dashed-dotted line shows the contribution of gravitational waves alone, and the solid line is the total $\\ell(\\ell+1)C_{\\ell}$. The top graph corresponds to the set of parameters $(5000,1000,-2)$, the middle graph to $(4500,1000,-1.95)$, and the bottom graph to $(4000,1000,-1.9)$. The crosses, again, correspond to the observed values of $\\ell(\\ell+1){\\mathbf{C}}|_{\\ell=10}$ and $\\ell(\\ell+1){\\mathbf{C}}|_{{\\mathrm{peak}}}$. In all cases, the contribution of gravitational waves to the plateau is comparable to that of density perturbations, and for larger values of $\\beta$ the contribution of gravitational waves is numerically greater than that of density perturbations.} \\label{compinclgw} \\end{figure} \\begin{table} \\caption{Comparison of the theoretical model with observations, for the sets of parameters $(5000,1000,-2)$, $(4500,1000,-1.95)$ and $(4000,1000,-1.9)$. Primordial gravitational waves have now been taken into account. The position and the height of the first peak, as well as the dipole moment and the plateau, are in an excellent fit to the observations.} \\label{fullcomparison} \\begin{small} \\begin{tabular}{@{}lccccccccc} $z_{{\\mathrm{eq}}}$&$z_{{\\mathrm{dec}}}$&$\\beta$&$\\Lambda$&$\\varpi$&$\\varpi \\cdot \\ell(\\ell+1)C^{{\\mathrm{dp}}}_{\\ell}|_{\\ell=10}$&$\\ell(\\ell+1)C^{{\\mathrm{gw}}}_{\\ell}|_{\\ell=10}$&$C^{{\\mathrm{gw}}}_{10}/(\\varpi \\cdot C^{{\\mathrm{dp}}}_{10})$&$\\varpi \\cdot C^{{\\mathrm{dp}}}_{1}$&${\\mathbf{C}}_{1}/(\\varpi \\cdot C^{{\\mathrm{dp}}}_{1})$ \\\\  \\hline 5000&1000&-2.00&$2.6\\cdot 10^{-10}$&29.2&$4.1\\cdot 10^{-10}$&$2.2\\cdot 10^{-10}$&0.5&$6.1\\cdot 10^{-6}$&1.04 \\\\  % 4500&1000&-1.95&$2.9\\cdot 10^{-10}$&17.2&$3.4\\cdot 10^{-10}$&$2.9\\cdot 10^{-10}$&0.8&$6.1\\cdot 10^{-6}$&1.03 \\\\  % 4000&1000&-1.90&$2.8\\cdot 10^{-10}$&11.6&$2.9\\cdot 10^{-10}$&$3.4\\cdot 10^{-10}$&1.2&$6.1\\cdot 10^{-6}$&1.03 \\\\  \\hline \\end{tabular} \\end{small}  \\end{table}  To summarise, we have revised the formula (\\ref{eq:one}) for $C_{\\ell}$'s caused by density perturbations and derived the corrected expression (\\ref{eq:Clfinal}). We have shown that the monopole moment $C_{0}$ is actually finite and small, while the dipole moment is very sensitive to short wavelengths. Because of that, the dipole moment is much larger than the $\\ell =0$ and $\\ell \\geq 2$ multipoles. We have then proceeded to the multipoles $\\ell \\geq 2$. The correct formula predicts growth of the function $\\ell(\\ell+1) C_{\\ell}$ with $\\ell$. We have illustrated how the presence of the modulating function $M^{2}\\!(n\\xi_{2})$ gives rise to the first peak in the CMBR anisotropy pattern produced by density perturbations. (It would be nice to see further peaks discovered in the positions and with the heights suggested by the modulating function, but this is the domain of wavelengths where many other physical processes may intervene.) A first comparison with observations has indicated that the observed multipoles can hardly be explained by density perturbations alone, because of the deficit in the small $\\ell$ plateau. We took a step further and included primordial gravitational waves. This inclusion gives a picture in better agreement with observations. We explored the cases $\\beta > -2$ which make the primordial spectrum a little bit `blue', thus allowing to avoid the infrared divergence of the primordial perturbations that occurs for $\\beta \\leq -2$ (see (\\ref{eq:powerB}) and (\\ref{eq:powerspectrumform})). We conclude that models with $\\beta > -2$, and therefore with larger contributions of gravitational waves, are better than a model without gravitational waves and with the flat spectrum $\\beta =-2$. This analysis is complementary to other studies of the CMBR data \\cite{Albr,Balb1,Balb2,DuNo,Huetal,Kinn,Lang,McGa,MaRiS,PaSe,PeSeH,Pogo,RaSha,ScLaG,TeZa,TeZaH1,TeZaH2}, not all of which, though, seem to agree with each other."
    },
    "0010/astro-ph0010073_arXiv.txt": {
        "abstract": "{ Recent spectropolarimetry of QSOs has revealed a surprising rise in polarization in the Lyman continuum of several objects. We discuss several recent attempts to interpret this feature, including the role of Lyman limit absorption in PG 1222+228. We present new theoretical results involving electron scattering in a hot corona or wind above an accretion disk, and polarization resulting from the relativistic returning radiation. These mechanisms can lead to potentially observable polarization at short wavelengths, but neither has quantitative success in fitting the observed Lyman continuum polarization rises. A renewed capability for ultraviolet spectropolarimetry from space is urgently needed to provide further clues to the nature of this phenomenon. } ",
        "introduction": "\\label{sec:Introduction}  Accretion onto black holes is widely believed to power active galactic nuclei (AGN).  The accretion flow likely takes the form of an orbiting disk, which gives efficient energy production and defines a natural axis for jets and double radio sources. Specific observational confirmation of the presence of disks in AGN has, however, been elusive.  Disks in AGN are defined by the black hole's mass, $M$, its dimensionless angular momentum, $a_*$, and the accretion rate, $\\dot M$.  For much of the relevant range of parameters, the disk should be geometrically thin  and optically thick. In this case, the disk may emit much of its power as thermal emission in the optical and ultraviolet.  Observations of QSOs indeed show a strong, broad continuum component, called the ``Big Blue Bump'', that is suggestive of disk emission (e.g., Shields 1978; Malkan 1983). Much effort has gone into observations at infrared to X-ray wavelengths, polarization studies, and related theory, in an attempt to confirm that this emission does indeed come from a thermally emitting disk (see review by Koratkar \\& Blaes 1999).  The case today remains inconclusive.  One reason for the difficulty of proving the presence of disks in AGN is the high orbital velocity of the material in the emitting zone.  This tends to mask spectral features of a disk that might have diagnostic value.  The bulk of the thermal continuum should come from dimensionless radii $r_* \\equiv R/R_g \\approx 10,$ where $R_g = GM/c^2$ is the gravitational radius.  The corresponding orbital velocity is $\\sim 0.3c$.  Detection of highly broadened Fe K$\\alpha$ emission has been cited as an argument for a relativistic disk (Tanaka \\etal\\ 1995), but spectral features in the optical and ultraviolet have not been identified.  One potential feature is the Lyman edge of hydrogen, at rest wavelength 912~\\AA.  Depending on the parameters, this feature may be in emission or absorption or absent. It seems likely to be present in some fashion at least in some objects, albeit highly broadened by Doppler broadening and relativistic effects (cf. Koratkar \\& Blaes 1999).  In fact, only a small fraction of QSOs, the so-called ``Lyman edge candidate QSOs'', show indications of a broadened Lyman edge in absorption (\\eg, Antonucci, Kinney, \\& Ford 1989; Koratkar, Kinney, \\& Bohlin 1992).  The hot, low density gases in the atmosphere of an AGN disk should be highly ionized.  Electron scattering should should then give significant plane polarization.  A drop in polarization in the Lyman continuum was predicted by Laor, Netzer, \\& Piran (1990) because of the increased absorption opacity from photoionization of hydrogen.  With this motivation, spectropolarimetry was obtained with the  {\\em Faint Object Spectrograph} on the {\\em Hubble Space Telescope} (\\HST) for a number of QSOs (Impey \\etal\\ 1995; Koratkar \\etal\\ 1995).  For several candidate Lyman edge QSOs, a surprising rise in polarization was found.  From values of only $\\sim 1\\%$, the polarization rises rather steeply around rest wavelength $\\sim 750$ \\AA, reaching $\\sim 5\\%$ in PG1222+228 and $\\sim 20\\%$ in PG 1630+377.  Weaker polarization rises at a similar wavelength were found in several other QSOs.  Until recently, these polarization rises appeared to be associated only with candidate Lyman edge QSOs (Koratkar \\etal\\ 1998).  Lyman continuum polarization rises in QSOs have generated considerable theoretical interest.  An interesting stellar atmosphere effect was proposed by Blaes \\& Agol (1996), in which the combined effects of electron scattering and photoionization opacity give a polarization rise somewhat to the short wavelength side of the Lyman edge.  This occurs for just the range of disk effective temperatures, \\teff, that should give a Lyman edge in absorption in the total flux.  This process did not give polarizations as high as 20\\%, however.  Moreover, Shields, Wobus, and Husfeld (1998, SWH) found that relativistic effects caused an additional blueshift of the polarization rise, relative to the wavelength in the rest frame of the orbiting gas.  The observed polarization rise then would occur at a wavelength too short to fit the observations.  SWH noted that if the polarization rises abruptly at the Lyman edge in the rest frame of the gas, relativistic effects give a good fit to the observed wavelength dependence of the polarization rise in PG1630+377.  The fit provides a possible measure of the black hole spin.  However, no physical reason for the postulated polarization rise is known.   Recently, Beloborodov \\& Poutanen (1999, BP) have offered an explanation for the polarization rises in terms of electron scattering in a corona or wind above the disk. In a different approach,  Lee \\& Blandford (1997) discussed scattering by resonance lines of heavy elements as a possible cause of the Lyman continuum polarization rises.  However, they did not give a quantitative model of the observed polarization rises.  We report here theoretical investigations of several aspects of Lyman continuum polarization rises in QSOs.  In \\S 2, we summarize recent results on PG1222+228, concerning the apparent coincidence of its polarization rise with an intervening Lyman limit system (LLS).  In \\S 3, we present new results for electron scattering winds and coronae.  In \\S 4, we discuss the relevance of relativistic returning radiation to the Lyman continuum polarization rises.  Our conclusions are summarized in \\S 5. ",
        "conclusions": "\\label{sec:conclusions}  We have explored several models to explain the Lyman continuum polarization in QSOs.  Neither Compton scattering nor returning radiation appear to be capable of explaining the strong, rapid rise in polarized flux observed for PG 1630+377. The same seems likely to be true for PG 1222+228, especially after correction for the LLS absorption.  Quite apart from the issues discussed above, there is the observed polarization of the \\lalpha\\ line in PG 1630+277. It is polarized in a similar positional angle to the 800 \\AA\\ continuum polarization, to a degree of about 3\\% (Koratkar \\etal\\ 1995).  The wavelength of the feature in polarized flux actually is on the red wing of the emission line in total flux, roughly at the expected wavelength of the N V $\\lambda$1240 resonance line (\\cf\\ Shields 2000). Certainly, the emission line in the polarized flux does not share the large blueshift of the polarization rise in the continuum, if it is indeed associated with the Lyman edge of hydrogen.  Discussions of the Lyman continuum polarization rise have tended to ignore the ``\\lalpha'' feature, although Koratkar \\etal\\ (1995) did suggest scattering of optically thin hydrogen emission followed by passage through an absorbing layer of modest optical depth at the Lyman limit.  From another point of view, however, there are only two natural position angles in the context of a disk geometry (parallel and perpendicular to the disk axis).  The polarized Lyman continuum and Lyman alpha could come from two different sources, which might coincidentally have the same position angle.  The polarized \\lalpha\\ feature has a normal broad emission-line width, and presumably comes from material at a radius $\\sim10^4 R_g$.  The polarized Lyman continuum may come from a smaller radius, as in the models of SWH and Blaes and Agol (1996).  Here, for a thermal intrinsic line width, thermalization may suppress the \\lalpha\\ emission, suggested by simple estimates based on the black body limit.  From a broader perspective,  most proposals for the Lyman continuum polarization rise seem contrived to explain this particular observation, rather than following naturally from a more comprehensive theory of AGN.  Thus, they do not address the issue of the statistical incidence of the polarization rise phenomenon. Unfortunately, there is currently no observational capability to confirm and extend the measurements of Lyman continuum polarization in QSOs.  The \\lalpha\\ forest is an increasing problem at higher redshifts, and there is an urgent need for a renewed capability to do ultraviolet spectropolarimetry from space."
    },
    "0010/astro-ph0010590_arXiv.txt": {
        "abstract": "A summary is given of the calibration of the Cepheid scale required in the derivation of distances from observations at $V$ and $I$, as in the HST work on extragalactic Cepheids. There is some evidence that a small metallicity correction may be necessary in deriving distances in this way. The evidence for this comes both from observations of Cepheids of different metallicities and from a comparison of the LMC distance modulus derived from Cepheids and that derived by other methods. ",
        "introduction": "At the present time the value adopted for $H_{0}$ depends directly on the calibration of local distance indicators. These local calibrators should have a number of rather obvious characteristics.\\\\ 1. Their absolute magnitudes should be derivable with good accuracy. The methods used to obtain these absolute magnitudes should be based on sound empirical results and be as free as possible from assumptions based on theory.\\\\ 2. The absolute magnitudes of the calibrators should ideally be free from a dependence on either the age or the metallicity of the objects involved. But if there are age and metallicity effects, these must be known empirically and measurable in both the calibrating and programme stars. \\\\ 3. The interstellar reddenings and absorptions of both calibrating and programme stars must be measurable with good accuracy.\\\\ Considerable emphasis has been put on point 1 in much of the current literature. However, points 2 and 3 are of equal importance, especially if one is hoping to establish scales to 5 or 10 percent. In fact, problems relating to these two latter points seem to account for some of the conflicting scales that have been recently proposed.\\\\ ",
        "conclusions": "Galactic calibration of the Cepheid PL($V$) and PC($V-I$) relations indicate that the scale used by HST workers (at least prior to mid-2000) needs increasing by about seven percent. Various tests suggests that there is a small metallicity effect when using $V$ and $I$ data in the manner adopted by the HST workers. The size of this effect is still rather uncertain. It is possibly $\\rm \\sim 0.1$ or $\\rm 0.2 \\, mag \\, [Fe/H]^{-1}$ in the sense that the corrected distances of metal-poor Cepheids are smaller than the uncorrected ones."
    },
    "0010/astro-ph0010629_arXiv.txt": {
        "abstract": "Four examples are given of problems that arise in the study of emission-line stars, each of which deserves further quantitative study. (1) Archival slit spectra of $\\eta$ Car show that epochs of low excitation have occurred regularly in the 2020 day period of Damineli et al. since at least 1948. They last about 10 percent of the period. Earlier slit spectra (1899 -1919) suggest that the the object was always in a low excitation state at that time. This result may be connected with the gradual brightening of $\\eta$ Car through most of the 20th century. (2) FeII and [FeII] emission in Mira variables is probably chromospheric. It would be important to understand how this is excited and its relation to mass loss and dust formation in Miras. (3) Declines of RCB stars are due to eclipses by dust clouds. At such times a rich emission line spectrum from the outer atmosphere of the star is seen. The quantitative study of the changes in this spectrum as more and more of the star is eclipsed should yield important information on the extended atmosphere of these objects. (4) In deep minima when the entire star is eclipsed, RCB stars show a broad-line emission spectrum. The excitation of this spectrum, particularly HeI 3888A, constitutes a puzzle. Two possible excitation mechanisms are discussed. ",
        "introduction": "The study of emission-line stars presents many opportunities to probe the physical nature of stellar atmospheres and stellar environments and also to investigate little understood modes of stellar evolution. This has been the theme of much of this meeting. In an attempt to show that we have as yet only begun to unlock this rich source of information, this paper begins with a discussion of results obtained using archival spectra of $\\eta$ Carinae and then examines two classes of emission-line stars which have not so far been mentioned at this meeting but which would repay further quantitative study. ",
        "conclusions": ""
    },
    "0010/astro-ph0010303_arXiv.txt": {
        "abstract": "We present CCD $BVI$ photometry of 85 variable stars from the field of the globular cluster NGC~6934.  The photometry was obtained with the image subtraction package ISIS. 35 variables are new identifications: 24 RRab stars, 5 RRc stars, 2 eclipsing binaries of W UMa-type, one SX Phe star, and 3 variables of other types.  Both detected contact binaries are foreground stars.  The SX Phe variable belongs most likely to the group of cluster blue stragglers.  Large number of newly found RR~Lyr variables in this cluster, as well as in other clusters recently observed by us, indicates that total RR~Lyr population identified up to date in nearby galactic globular clusters is significantly ($>30\\%$) incomplete.  Fourier decomposition of the light curves of RR Lyr variables was used to estimate the basic properties of these stars.  From the analysis of RRc variables we obtain a mean mass of $M=0.63\\;M_\\odot$, luminosity $\\log L/L_\\odot=1.72$, effective temperature $T_{\\rm eff}=7300$ and helium abundance $Y=0.27$.  The mean values of the absolute magnitude, metallicity (on Zinn's scale) and effective temperature for RRab variables are: $M_V=0.81$, ${\\rm [Fe/H]}=-1.53$ and $T_{\\rm eff}=6450$, respectively.  From the $B-V$ color at minimum light of the RRab variables we obtained the color excess to NGC 6934 equal to $E(B-V)=0.09\\pm 0.01$.  Different calibrations of absolute magnitudes of RRab and RRc available in literature were used to estimate apparent distance modulus of the cluster: $(m-M)_{\\rm V}=16.09\\pm 0.06$. We note a likely error in the zero point of the {\\em HST}-based $V$-band photometry of NGC~6934 recently presented by Piotto et al. Among analyzed sample of RR Lyr stars we have detected a short period and low amplitude variable which possibly belongs to the group of second overtone pulsators (RRe subtype variables).  The $BVI$ photometry of all variables is available electronically via {\\tt anonymous ftp}.  The complete set of the CCD frames is available upon request. ",
        "introduction": "NGC 6934 (${\\rm RA}=20^h34^m$, ${\\rm Decl.}=+7^\\circ24'$, J2000.0) is an intermediate-metallicity globular cluster.  In his catalog Harris (1996) adopted for it ${\\rm [Fe/H]}=-1.54 $.  The color magnitude diagrams of this cluster were obtained by Harris and Racine (1973), Brocato et al. (1998) and Piotto et al. (1999).  The cluster was extensively searched for variable stars by Sawyer-Hogg \\& Wehlau (1980), who listed 51 variable stars, 50 of which were RR Lyr variables. They have found mean periods for 45 RRab and 5 RRc stars equal to $0.552\\;$days and $0.294\\;$days, respectively. These values place NGC~6934 among Oosterhoff type I clusters. ",
        "conclusions": "We have presented the photometry of 85 variables in the globular cluster NGC 6934. The photometry was obtained using the newly developed image subtraction method (Alard \\& Lupton 1998; Alard 2000). As first demonstrated by Olech et al.~(1999), image subtraction method provides a very powerful tool for extracting photometry of variables located in central regions of globular clusters. As many as 35 variables from our sample are new discoveries. Among these newly identified stars we have detected 24 RRab stars, 5 RRc stars, 2 eclipsing W UMa systems, one SX Phe star and three other variables. Our total sample contains photometry for 68 RRab and 10 RRc stars.  Suntzeff et al.~(1991) estimated that only 6\\% of the RR~Lyr variables hosted by galactic GCs remain to be discovered.  The case of NGC~6934 and recent results obtained by our group for some other clusters (M5, M55 and NGC~6362) shows that incompleteness of available samples is much higher. In case of NGC~6934 we have identified 29 new RR~Lyr stars what amounts up to about 37\\% of the total sample for this cluster.  The periods of RRab variables from NGC 6934 are between 0.4548 and $0.956\\;$days with the mean value of $0.574\\;$days. Periods of RRc stars are between 0.2470 and $0.4008\\;$days with the mean at $0.310\\;$days.  Only four RRc stars from our sample have the light curves with quality good enough for computing their physical parameters from the Fourier decomposition coefficients. The mean values of the mass, luminosity, effective temperature and helium abundance for these RRc stars are $0.63\\pm 0.06~M_\\odot$, $\\log L/L_\\odot=1.72\\pm 0.01$, $T_{\\rm eff}=7290\\pm 27$, and $Y=0.27\\pm 0.01$, respectively.  Out of 69 RRab variables only 24 showed stable light curves during our observations.  For this sub-sample we used the method developed by Kov\\'acs \\& Jurcsik (1996, 1997) to derive values of the absolute magnitude, metallicity and effective temperature which are equal to $M_V=0.81\\pm 0.01$, ${\\rm [Fe/H]}=-1.53\\pm 0.04$ (Zinn-West scale) and $T_{\\rm eff}=6455\\pm 18$, respectively.  From the $B-V$ color at minimum light of the RRab variables we obtained the color excess to NGC 6934 equal to $E(B-V)=0.09 \\pm 0.01$. It is marginally consistent with the recent determination of Piotto et al. (1999), who derived $E(B-V)=0.05\\pm 0.02$, but agrees well with a value $E(B-V)=0.11$ obtained using the reddening map of Schlegel et al. (1999).  We obtained five estimates of an apparent distance modulus of the cluster using various calibrations of $M_{V}$ for RR~Lyr stars and HB stars. A calibration based on statistical-parallax method (Gould \\& Popowski 1998) which is preferred by us leads to $(m-M)_{\\rm V}=16.10\\pm 0.13$.  We noted a likely error in the zero point of photometry published for the cluster by Piotto et al (1999).   Among cluster RR~Lyr stars we have detected a short period and low amplitude variable whose characteristic make it a good candidate for the second overtone pulsator.  Both eclipsing variables detected in this search are W UMa-type systems and both most likely are the foreground stars.  The magnitude and the color of the single detected SX Phe variable suggest that it belongs to the cluster blue stragglers group.  Its distance modulus computed based on relation of McNamara (1997) is compatible with the cluster membership of that star.  Our sample of newly identified variables includes also one AGB star with possible period $P\\approx 49\\;$days, one periodic variable with $P=1.62\\;$days or $P=2.61\\;$days and one star which is a likely binary but for which we have no clue neither for its type of variability nor for possible periodicity of light variations."
    },
    "0010/astro-ph0010135_arXiv.txt": {
        "abstract": " ",
        "introduction": "\\label{intro}  Deep submm surveys of small patches of sky have only just become feasible, largely due to the development of the SCUBA instrument on the James Clerk Maxwell Telescope (Holland \\et\\ 1999). Several groups have pursued surveys on scales of a few arcminutes to mJy levels (Smail, Ivison, and Blain 1997; Hughes \\et\\ 1998; Barger \\et\\ 1998, 1999; Eales \\et\\ 1998; Lilly \\et\\ 1999; Blain \\et\\ 1999a), finding a very large sky density of faint submm sources. These surveys seem to have justified the long held belief that submm surveys would be a short-cut to the high-redshift universe, as the so-called ``negative k-correction\" effect largely counterbalances the dimming due to distance and redshift, so that objects can be seen just as easily at very high redshift ($z=3-10$) as they can at moderate redshift ($z\\sim 1$). High redshift optical identifications have indeed been proposed for many of the submm sources in the above papers. The IR-mm luminosities of the sources are then very large, similar to local ultraluminous IRAS galaxies, with implied star formation rates of hundreds of solar masses per year. However, given the sky density of these sources, the high redshift luminosity density must be large. Hughes \\et\\ (1998), in their study of submm sources in the Hubble Deep Field (HDF), show that the implied star formation rate at $z=2-5$ is {\\em larger} than that deduced from the UV fluxes of all the optically selected high-z galaxies in the HDF. The optically derived star-formation history curve, or Lilly-Madau plot, has since undergone revision but the qualitative point remains unchanged. This has striking consequences - most of the star formation in the young universe was obscured, and most of the star formation in the young universe went on in a tiny minority of starburst galaxies, not in the great majority of quiescent or moderately active objects. This is quite the reverse of the situation today. There may be problems with such a picture - Blain \\et\\ (1999a) show that a star-formation history of the kind deduced by Hughes \\et\\ may, over the history of the universe, overproduce the metals, and overproduce the number of observed stars.  It may very well be that the above story is broadly correct, but it is so important that we must make sure that we have excluded reasonable alternatives. One such alternative is that some fraction of the submm sources are actually AGN rather than starbursts (Almaini Lawrence and Boyle 1999;  McMahon \\et\\ 1999; Fabian \\et\\ 2000;  Gunn \\et\\ in preparation). Another more radical alternative is that the submm surveys, like all other first surveys of the sky at a new wavelength, are discovering a {\\em new class of object}. We are driven to examine such an idea because the optical identifications of submm sources with high redshift galaxies, while highly plausible, are often only circumstantial. The problem of course is the large beam size of SCUBA (14\\arcsec ) so that identifications are nearly always ambiguous. In some cases, a relatively bright galaxy is within the error circle, and so statistically secure. But these are the less interesting cases - reasonably secure identifications at $z\\sim0.5-1$. In some other cases, identifications in the range $z\\sim 1-3 $ are secure because of confirming CO detections, or because of the presence of extremely red objects (EROs) which are relatively rare (Ivison \\et\\ 1998; Frayer \\et\\ 1998, 1999; Smail \\et\\ 1999). In a large number of cases however, several faint blobs are positionally consistent, and the identification rests on the constraints given by limits on the spectral energy distribution (SED) - for example non-detection at 15\\um\\ in the ISO data together with a detection at 850\\um\\ ruling out objects with an ARP~220-like SED at low redshift (see Hughes \\et\\ 1998), or the submm/radio ratio (Carilli and Yun 1999; Blain 1999). In all such cases, the submm source is equally consistent with having no optical counterpart at all.  For the brightest SCUBA sources, mm-wave interferometry is just feasible and can of course substantially increase locational accuracy. Downes \\et\\ (1999) observed HDF~850.1, the brightest SCUBA source in the HDF, with IRAM, and achieved a position accurate to 0.3\\arcsec . Within the $2\\sigma$ contour, there is {\\em no object at all} to the limit of the HDF. Two very faint objects are consistent at $3\\sigma$. However, the tentative photometric redshifts of those objects are inconsistent with the SED of HDF~850.1, if the intrinsic SED is like that of ARP~220 or M82 (Hughes \\et\\ 1998). There are several possible explanations. (i) It could be that the intrinsic SED is different from those classic exemplars, and indeed Downes \\et\\ find a local object whose SED does appear to be consistent. (ii) It could simply be that the tentative redshifts are wrong. (iii) The most interesting possibility is that the correct identification is at very high redshift, $z>5$. The combination of distance dimming, k-correction, and some reddening could move such an object out of the detection limit for the HDF. Deep K-band imaging by Smail \\et\\ (1999) has shown that perhaps 10\\% of the SCUBA sources are ``Extremely Red Objects (EROs)\", appearing only in the IR. Unpublished deep K-band imaging of the CFRS survey sources show that only one third appear to identified, even to $K=21.5$, suggesting that {\\em most} counterparts are at $z>2$ (Eales, private communication).  In some ways the simplest explanation for the non-identification of HDF~850.1 is that it is an object that is bright in the submm but emits negligible light at optical wavelengths. Is it possible that such a previously unsuspected population of objects exists, and at such a high sky density  ? We must at least discount such a hypothesis before believing the consequences of submm surveys for cosmology. In this paper I consider three such possible populations and test them against the constraints we have. The first two - brown dwarfs and solar system objects - are quickly dispensed with. The third hypothetical population examined - very cold dense dark dust clouds - turns out to be much harder to dismiss, as long as the clouds are in the sub-stellar mass range. Furthermore such objects may plausibly contribute substantially to the dark matter problem, as has been suggested by other authors for quite different reasons (Pfenniger Combes and Martinet 1994; Gerhard and Silk 1996; Walker and Wardle 1998). Once we have arrived at the concept of such objects, they seem more than plausible, as a likely end-product of cloud collapse, and as an alternative to brown dwarfs. Even if they do not turn out to constitute a large fraction of the SCUBA sources, it is of some interest to discover whether they exist at all. ",
        "conclusions": "\\label{conc}  The lack of clear optical identifications for a large fraction of faint submm sources is pushing us towards believing that they are very high redshift objects, but we must first scrupulously consider alternative possibilities. In local hypotheses, these sources are extremely cold, $T\\sim 7$ K. They cannot be brown dwarfs or solar system bodies without grossly violating simple constraints. Cold dusty clouds are harder to rule out completely, but if such objects are not to violate limits on local mass density, extinction features, and the FIR background, they are constrained to have masses similar to Jupiter to within an order of magnitude, and to be a Galactic Plane population. Such clouds do not explain the dark matter problem, unless there is a much larger population of similar but dust-free objects. (In this respect my proposal differs from most of the rest of the literature in this area which involves a pervasive but dust-free halo population). If such cold dusty clouds do exist, they are an important new component of the interstellar medium, and may be an important part of the star formation puzzle. The main theoretical objection is that such objects should be unstable to collapse on a thermal timescale, but they may be short-lived objects which are part of a dynamic ISM, as argued by Pfenniger and Combes (1994).  The majority of faint SCUBA sources clearly really are distant galaxies. However if even a quarter or a third of the SCUBA sources are actually local objects, the cosmological implications are significant, as this would selectively remove the objects believed to be at $z>3$."
    },
    "0010/astro-ph0010213_arXiv.txt": {
        "abstract": "We discuss the cooling history of the  low--mass, helium core white dwarfs in  short orbital period   millisecond pulsars PSR J0751+1807 and PSR J1012+5307. The revised cooling age estimated by Alberts et al. agrees with the age estimation for PSR J1012+5307, removing the discrepancy between the spin--down age and the cooling age. However, if we accept this model then the helium white dwarf in the binary pulsar system PSR J0751+1807 must be much hotter than is observed. We propose that this discrepancy may be resolved if, after detachment of the secondary star from its Roche lobe in PSR J0751+1807, the star loses its hydrogen envelope due to pulsar irradiation. When hydrogen burning stops, the white dwarf will cool down much more quickly than in the case of a thick hydrogen envelope with a hydrogen burning shell. We discuss several possibilities to explain different cooling histories of white dwarfs in both  systems. ",
        "introduction": "It is accepted that formation of a binary millisecond (or recycled) pulsar with a low--mass companion may be explained as the end--point of close binary evolution in which an old pulsar is spun--up by accretion from the red giant (secondary) (see Bhattacharaya \\& van den Heuvel 1991, for review). Once the secondary overflows its Roche lobe, it starts transferring mass and angular momentum to the neutron star. During the long term accretion phase the neutron star is generally assumed to be spun--up to the equilibrium period ($P_{eq}$). The equilibrium period can be written as (Bhattacharaya \\& van den Heuvel 1991):  \\begin{equation} P_{eq}=1.9~ms~ B_9^{6/7}\\left(\\frac{M_{ns}}{1.4M_\\odot}\\right)^{-5/7}\\left (\\frac{\\dot{M}}{\\dot{M}_{Edd}}\\right)^{-3/7} R_6^{16/7}~,~ \\end{equation}  where $B_9$ = B/$10^9$ G, $R_6$= $R_{ns}$/$10^6$ cm and $M_{ns} $ are the magnetic field strength, radius and mass of the neutron star respectively and ${\\dot M}_{Edd} $ is the Eddington mass accretion rate. After the whole envelope of the red giant has been transferred, the binary consists of a low--mass star with a helium core (progenitor of a low--mass helium white dwarf) and a fast rotating neutron star (millisecond pulsar). At this point the pulsar spin period starts to change due to magneto--dipole radiation, with a characteristic time {\\it t} elapsed from this epoch \\begin{equation} {\\it t} = \\tau \\times [\\frac{2}{n-1}][1- (\\frac{P_0}{P})^{n-1}]~,~ \\end{equation} where $\\tau$=P/(2$\\dot{P}$), $P_0$ and P  are the initial and current spin periods, and $\\dot{P}$ is the spin period derivative. n is the braking index (=3) and if $P_0\\ll$P then {\\it t}=$\\tau$. It is possible to estimate the age {\\it t} from the cooling age of the white dwarf. The latter can be calculated from the observed parameters of the ($T_{eff}$, $\\log ~g$). This gives a unique opportunity for  determination of neutron star age which is independently of the spin--down theory of pulsars.  In this short note we shall discuss  cooling properties of  helium white dwarfs in  short period  binary systems with millisecond pulsars, PSR J0751+1807 and PSR J1012+5307. ",
        "conclusions": "The first question is what is the real value of the initial pulsar period $P_0$ for both systems? Is it equal to $P_{eq}$? To explain the discrepancy between the cooling and spin--down ages of PSR J1012+5307 Burderi et al. (1996) have accepted the so called accretion--induced magnetic field decay model (see Bhattacharya \\& Srinivasan 1995 for review). In this case, at the end of the accretion phase, field decay takes place on a shorter time--scale than spin--up, and spin equilibrium is no longer achieved.  The second point is connected with the  discussion of Webbink's (1975) results, who found that there is a hydrogen burning in the thick envelope (of a few $\\times10^{-3} $$M_\\odot$) of low--mass helium white dwarfs ($< 0.2 ~M_\\odot$). Recently three different groups, Albert et al. (1996, using Eggleton evolutionary program), Driebe et al. 1998 (Kippenhahn program) and we (Sarna et al. 1999,2000, Paczy\\'nski program) have obtained the same results: small mass helium white dwarfs ($<$ 0.2$M_\\odot$), after detachment from the Roche lobe, have a massive hydrogen layer (a few $\\times 10^{-2}~M_\\odot$) in which stationary hydrogen burning continues. This keeps these low--mass helium white dwarfs hot and prolongs their cooling time.  However, there is a difference between Alberts et al. and our+Driebe's results, since we and Driebe found hydrogen flashes for M$>$ 0.2 $M_\\odot$ and Alberts et al. did not. We show that the transition from a cooling white dwarf with a temporarily stable hydrogen--burning shell to a cooling white dwarf in which almost all residual hydrogen is lost in a few thermal flashes (via Roche--lobe overflow) occurs between 0.183--0.213 $M_\\odot$, which well agrees with the cooling scenario proposed by Burderi et al. (1998).  Therefore we may accept that PSR J1012+5307 is in a long term hydrogen shell burning stage and this may be the reason why the white dwarf is too hot (accepting that spin--down age of 7 Gyr, which value could be even larger if the pulsar has a significant transverse velocity (Hansen \\& Phinney 1998), is almost equal to the cooling age). For PSR J0751+1807 the irradiation driven mass loss reduces the hydrogen envelope mass, the hydrogen burning deceases and the white dwarf is quickly cooled.  Taking into account the effect of illumination, $P_{eq}$ is $\\sim 1.6\\times 10^{-3}$s (Ergma \\& Sarna 1996). Therefore, if the pulsar in PSR J1012+5307 has been spun up to $P_{eq}$, it also has suffered irradiation driven mass loss. If we do not include illumination then $P_{eq}$ is longer, since at the beginning of RLOF mass transfer is proceeding on the thermal time--scale at a rather high mass accretion rate, and the neutron star spins--up to millisecond periods. After about 0.2 $M_\\odot$ have been lost, the mass transfer rate decreases and the neutron star enters a ``propeller phase'' during which the fast rotating neutron star will slow down. To avoid irradiation--driven mass loss from this system the external irradiation flux must be less than critical flux $\\sim 1.6\\times10^{10}$ erg$s^{-1}cm^{-2}$ (Podsiadlowski 1991). Let us estimate the value of the initial pulsar period $P_0$ to allow such a flux. Simple estimates show that $P_0$ must be longer than 3.85$\\times 10^{-3}$s, and the time elapsed from detachment of the Roche lobe is $\\sim 3.2\\times 10^9$ yrs. If $P_0$ $>$ 3.85$\\times 10^{-3}$s then there is no irradiation driven mass loss and the hydrogen burning is stationary, prolonging the cooling time. Since both  systems are in many parameters very similar to each other, only the white dwarf cooling time--scale is very different and there must be something which leads to the different cooling histories for the helium white dwarfs in these systems.  Different irradiation fluxes ($F_x$(J0751+1807) $\\approx 10\\times F_x$(J1012+5307)) may explain the discrepancy.  We would like to point to an important issue  neglected in many papers. Usually the comparison with observed system  is made for one or several parameters (for example the cooling age of the white dwarf, $T_{eff}$ and so on) but not with all available data. For binary systems we must build a selfconsistent picture  with the right orbital parameters (which may be a very hard task, as in the case of the two systems discussed here), masses (if known) and cooling parameters of the white dwarf.   The effect of irradiation on the secular evolution of binaries containing a millisecond pulsar has been discussed in a recent review paper by D'Antona (1996). She has pointed out that one needs to distinguish the two main effects of irradiation with following terms; i) illumination: structural changes in the whole stellar structure due to irradiation, which may cause the secondary to fill its Roche lobe again, ii) evaporation: direct driving of mass loss from the secondary. Both cases give the same final result -- the decrease of the mass of the hydrogen envelope.  We propose that for PSR J0751+1807 the irradiation by the pulsar causes irradiation--driven mass loss from secondary, resulting in exhaustion of the hydrogen envelope, and as a consequence of this, a hydrogen shell burning stops. The irradiation--driven wind influences also the cooling phase, decreasing the cooling time--scale by a factor 1.7--4 (depending on how efficient this process is).  For the PSR J1012+5307 system the irradiative flux may be too small to cause significant mass loss, and hydrogen shell burning will be maintained for a long enough time to keep the white dwarf hot."
    },
    "0010/astro-ph0010025_arXiv.txt": {
        "abstract": "We present the luminosity function of $90\\mu$m selected galaxies from the European Large Area ISO Survey (ELAIS), extending to $z=0.3$. Their luminosities are in the range $10^{9}<h_{65}^{-2}L/L_\\odot<10^{12}$, i.e. non-ultraluminous. From our sample of $37$ reliably detected galaxies in the ELAIS S1 region from the Efstathiou et al. (2000) $S_{90}\\ge 100$mJy database, we found optical, $15\\mu$m or $1.4$GHz identifications for $24$ ($65\\%$). We have obtained 2dF and UK Schmidt FLAIR spectroscopy of $89\\%$ of IDs to rigid multivariate flux limits. We construct a luminosity function assuming (a) our spectroscopic subset is an unbiased sparse sample, and (b) there are no galaxies which would not be represented in our spectroscopic sample at {\\it any} redshift. We argue that we can be confident of both assumptions. We find the luminosity function is well-described by the local $100\\mu$m luminosity function of Rowan-Robinson, Helou \\& Walker (1987). {\\it Assuming} this local normalisation, we derive luminosity evolution of $(1+z)^{2.45\\pm0.85}$ ($95\\%$ confidence). We argue that star formation dominates the bolometric luminosities of these galaxies and we derive comoving star formation rates in broad agreement with the Flores et al. (1999) and Rowan-Robinson et al. (1997) mid-IR-based estimates. ",
        "introduction": "\\label{sec:introduction} The study of the star formation history of the Universe is an extremely active field, in which much of the current debate is centred on the uncertainties in dust obscuration. Selecting star forming galaxies in the UV is relatively cheap in observing time, but such selection makes the samples extremely sensitive to dust obscuration. Detecting the reprocessed starlight requires sub-mm surveys (e.g. Smail, Ivison \\& Blain 1997; Hughes et al. 1998; Barger et al. 1998, 1999; Eales et al. 1999, Peacock et al. 2000; Ivison et al. 2000a,b, Dunne et al. 2000) and/or space-based mid-far-IR surveys (e.g. Rowan-Robinson et al. 1997, Taniguchi et al. 1997, Kawara et al. 1998, Flores et al. 1999, Puget et al. 1999, Oliver et al. 2000; see e.g. Oliver 2000 for a review). As such, the European Large Area ISO Survey (ELAIS) is well-placed for the study of the evolution and obscuration of the star formation in the Universe, and strong constraints at $z\\stackrel{<}{_\\sim}1$ are possible from ELAIS (Oliver et al. 2000).  \\begin{table*} \\begin{tabular}{llllllllll} Name  & RA          & Dec          & $S_{90}$/Jy & $S_{15}$/mJy & $S_{1.4}$/mJy & R & z & z$_{\\rm max}$ & $L/L_\\odot$ \\\\ & (J2000)     & (J2000)      &             &          &           &   &   &             \\\\ ELAISP90\\_J003857-424358        & 00 38 57.7 & -42 43 58.8  &   0.43 &     2.9 &    0.964    &             &          & &         \\\\ ELAISP90\\_J003056-441633        & 00 30 57.0 & -44 16 33.6  &   0.35 &     4.9 &             &     13.7    &   0.019  &0.0351$^P$ & $2.8\\times 10^{9}$\\\\ % ELAISP90\\_J003510-435906        & 00 35 10.5 & -43 59  6.0  &   0.41 &     9.6 &             &     12.1    &   0.136  &0.272$^P$ & $1.7\\times 10^{11}$\\\\ % ELAISP90\\_J003431-432613        & 00 34 31.6 & -43 26 13.2  &   0.59 &         &             &     14.0    &   0.053  &0.127$^P$ & $3.7\\times 10^{10}$\\\\ % ELAISP90\\_J003914-430437        & 00 39 14.8 & -43 04 37.2  &   1.79 &    16.2 &             &     11.6    &   0.012  &0.049$^P$ & $5.7\\times 10^{9}$\\\\ % ELAISP90\\_J003021-423657        & 00 30 21.1 & -42 36 57.6  &   1.17 &    22.3 &             &     16.0    &   0.149  &0.493$^P$ & $6.0\\times 10^{11}$\\\\ % ELAISP90\\_J003459-425718        & 00 34 59.0 & -42 57 18.0  &   0.55 &     5.5 &     1.371   &     14.9    &   0.055  &0.101$^R$ & $3.7\\times 10^{10}$\\\\ % ELAISP90\\_J003134-424420        & 00 31 34.5 & -42 44 20.4  &   0.31 &         &             &     13.1    &   0.027  &0.048$^P$ & $5.1\\times 10^{9}$\\\\ % ELAISP90\\_J003242-423314        & 00 32 42.9 & -42 33 14.4  &   0.24 &     2.6 &    0.463    &     15.2    &   0.053  &0.081$^P$ & $1.5\\times 10^{10}$\\\\ % ELAISP90\\_J003615-424344        & 00 36 15.4 & -42 43 44.4  &   0.31 &         &             &     17.2    &   0.115  & &        \\\\ % ELAISP90\\_J003516-440448        & 00 35 16.1 & -44 04 48.0  &   0.29 &         &             &             &          & &        \\\\ ELAISP90\\_J003741-440227        & 00 37 41.6 & -44 02 27.6  &   0.21 &         &             &     19.9    &   0.348  & &        \\\\ % ELAISP90\\_J003431-433806        & 00 34 31.2 & -43 38  6.0  &   0.75 &         &    0.595    &     18.4    &   0.199  &0.357$^O$ & $6.8\\times 10^{11}$\\\\ % ELAISP90\\_J003405-423816        & 00 34  5.3 & -42 38 16.8  &   0.23 &         &             &             &          & &        \\\\ ELAISP90\\_J003358-441102        & 00 33 58.2 & -44 11  2.4  &   0.23 &     5.0 &             &     18.9    &   0.305  &0.446$^O$ & $5.2\\times 10^{11}$\\\\ % ELAISP90\\_J003501-423852        & 00 35  1.9 & -42 38 52.8  &   0.28 &         &    0.606    &     15.9    &   0.054  &0.090$^P$ & $1.8\\times 10^{10}$\\\\ % ELAISP90\\_J003254-424608        & 00 32 54.2 & -42 46  8.4  &   0.46 &     2.5 &    0.469    &     17.7    &   0.190  &0.207$^C$ & $3.8\\times 10^{11}$\\\\ % ELAISP90\\_J003531-423314        & 00 35 31.4 & -42 33 14.4  &   0.16 &         &             &             &          & &        \\\\ ELAISP90\\_J003635-430148        & 00 36 35.7 & -43 01 48.0  &   0.12 &     4.5 &    0.358    &     21.5    &          & &        \\\\ ELAISP90\\_J003019-432006        & 00 30 19.8 & -43 20  6.0  &   0.20 &         &             &             &          & &        \\\\ ELAISP90\\_J003839-422324        & 00 38 40.0 & -42 23 24.0  &   0.14 &         &             &             &          & &        \\\\ ELAISP90\\_J003014-440509        & 00 30 14.3 & -44 05  9.6  &   0.10 &     3.6 &             &             &          & &        \\\\ ELAISP90\\_J003912-431203        & 00 39 12.2 & -43 12  3.6  &   0.15 &     5.0 &             &     14.8    &          & &        \\\\ ELAISP90\\_J003133-425100        & 00 31 33.9 & -42 51  0.0  &   0.11 &     6.3 &             &     18.0    &   0.211  &0.222$^P$ & $1.1\\times 10^{11}$\\\\ % ELAISP90\\_J003304-425212        & 00 33  4.3 & -42 52 12.0  &   0.19 &     1.4 &             &     15.6    &          & &        \\\\ ELAISP90\\_J003046-432204        & 00 30 46.4 & -43 22  4.8  &   0.11 &     6.2 &             &     15.6    &   0.073  &0.075$^P$ & $1.3\\times 10^{10}$\\\\ % ELAISP90\\_J003027-433050        & 00 30 27.7 & -43 30 50.4  &   0.11 &     4.9 &             &     15.7    &   0.072  &0.074$^P$ & $1.2\\times 10^{10}$\\\\ % ELAISP90\\_J002845-424420        & 00 28 45.5 & -42 44 20.4  &   0.11 &         &             &             &          & &        \\\\ ELAISP90\\_J003719-421955        & 00 37 19.4 & -42 19 55.2  &   0.16 &         &             &             &          & &        \\\\ ELAISP90\\_J003725-424554        & 00 37 25.2 & -42 45 54.0  &   0.12 &         &             &             &          & &        \\\\ ELAISP90\\_J003731-440758        & 00 37 31.3 & -44 07 58.8  &   0.32 &         &             &     20.0    &          & &        \\\\ ELAISP90\\_J003625-441127        & 00 36 25.0 & -44 11 27.6  &   0.14 &         &     4.265   &             &          & &        \\\\ ELAISP90\\_J003721-434228        & 00 37 21.5 & -43 42 28.8  &   0.12 &    17.3 &    0.451    &     17.4    &   0.225  &0.248$^P$ & $1.4\\times 10^{11}$\\\\ % ELAISP90\\_J003348-433032        & 00 33 48.3 & -43 30 32.4  &   0.15 &         &             &             &          & &        \\\\ ELAISP90\\_J003149-423628        & 00 31 49.2 & -42 36 28.8  &   0.11 &         &             &             &          & &        \\\\ ELAISP90\\_J003244-424803        & 00 32 44.4 & -42 48  3.6  &   0.12 &     1.2 &             &     19.0    &  0.192   & &        \\\\ % ELAISP90\\_J003323-432634        & 00 33 23.2 & -43 26 34.8  &   0.19 &         &             &     18.7    &  0.316   & &        \\\\ % \\end{tabular} \\caption{\\label{tab:sample} ELAIS S1 $90\\mu$m sample. The $90\\mu$m photometry has a $\\sim30\\%$ possible systematic uncertainty (see Paper III), and the $15\\mu$m photometry is subject to a possible systematic scaling to fainter fluxes by up to a factor of $\\sim1.5$ (see Paper II). We assume $H_0=65$ km s$^{-1}$ Mpc$^{-1}$, $\\Omega_0=1$, $\\Lambda=0$, and bolometric luminosities are calculated at $90\\mu$m assuming a constant $\\nu L_\\nu$ spectrum. The maximum accessible redshifts $z_{\\rm max}$ quoted assume no evolution, and are labelled respectively as $P$, $R$, $O$ and $C$ for maximum redshifts imposed by the $90\\mu$m PHOT flux limit, the radio flux limit, the optical magnitude limit and the CAM flux limit. Two galaxies (ELAISP90\\_J003721-434228 and ELAISP90\\_J003021-423657) are Seyfert II galaxies and two further galaxies (ELAISP90\\_J003431-433806 and ELAISP90\\_J003133-425100) have an early type spectrum; all the remainder show starburst emission line spectra. Also listed are $4$ redshifts for candidate optical IDs with the optical magnitudes listed, but which lack other multi-wavelength detections and are too faint optically to be certain that the correct ID has been found. All $4$ such identifications show starburst emission line spectra, but nevertheless are excluded from the luminosity function analysis as they fail the strict selection function defined in the text. }\\end{table*}   The scientific aims and strategy of ELAIS are presented in detail in Paper I (Oliver et al. 2000). In Papers II and III (Serjeant et al. 2000, Efstathiou et al. 2000) we presented respectively the ELAIS Preliminary Analysis source counts from the CAM ($6.7\\mu$m and $15\\mu$m) and PHOT ($90\\mu$m) instruments on ISO. The $90\\mu$m sample covered $11.6$ square degrees, and the source counts were found to agree well at the bright end with the IRAS $100\\mu$m counts. Excellent agreement was also achieved with a parallel independent pipeline (Surace et al. in prep.).  In this paper we present the first $90\\mu$m luminosity function from our initial spectroscopic campaigns in the ELAIS S1 field (Oliver et al. 2000 in prep., Gruppioni et al. 2000 in prep., La Franca et al. 2000 in prep., Linden-V\\o rnle et al. 2000 in prep.). Section \\ref{sec:method} defines our sample. In section \\ref{sec:results} we use this catalogue to derive a $90\\mu$m luminosity function. We discuss the implications of our results in section \\ref{sec:discussion}. We assume $H_0=65$ km s$^{-1}$ Mpc$^{-1}$, $\\Omega_0=1$, $\\Lambda=0$ throughout. ",
        "conclusions": "\\label{sec:discussion} We have found that both the shape and normalisation of the $90\\mu$m luminosity function is well-described by the local luminosity function of Rowan-Robinson, Helou and Walker (1987), with evolution consistent with $(1+z)^{2.45 \\pm 0.85}$ pure luminosity evolution, using the transformation $S_{100}\\simeq S_{90}$. Pure density evolution is already excluded in this population as it conflicts with sub-mJy radio source counts (Rowan-Robinson et al. 1993).  We can use these results to predict the redshift ranges of current and future $90\\mu$m surveys. We assume the evolution extends to between $z=1-2$ and is then unevolving. Figure \\ref{fig:frac} shows the $95\\%$ confidence limit for surveys of various depths, incorporating the uncertainty in the ELAIS $90\\mu$m absolute flux calibration. One corollary is  that at least $10\\%$ of the currently unidentified $90\\mu$m galaxies in ELAIS extend to $z>0.5$, possibly even $z>2$.  Is it reasonable to assume that star formation dominates the bolometric luminosities of our $90\\mu$m sample? Genzel et al. (1998), Lutz et al. (1998) and Rigopoulou et al. (1999) have presented mid-infrared spectroscopy of ultraluminous galaxies ($L\\ge 10^{12} L_\\odot$) in the same redshift interval as our sample, and using the PAH features as a star formation indicator found that star formation dominates the bolometric power outputs in $70-80\\%$ of ULIRGs. The AGN bolometric fraction derived this way decreases monotonically at lower luminosities (e.g. Lutz et al. 1998). In local IR galaxies between $10^{11}$ and $10^{12} L_\\odot$, $\\sim15\\%$ have Seyfert II spectra (Telesco 1988), though care should be taken in interpreting AGN dominance from optical spectra (e.g. Lutz et al. 1999, Taniguchi et al. 1999). At $L<10^{11}L_\\odot$ most local IR galaxies are single, gas rich spirals (e.g. Sanders \\& Mirabel 1996) in which many lines of argument point to starburst dominance in the IR (e.g. Telesco 1988): for example, the similarity of the spectral energy distributions to those of galactic star forming regions; the linear $L_{\\rm IR}-L_{\\rm CO}$ correlation over many orders of magnitude in luminosity; and optical/IR spectroscopic confirmation of star formation activity. It therefore seems likely that our sub-ultraluminous population should not be powered by active nuclei, as is also suggested by the low fraction of spectra in our sample ($10\\%$) with Seyfert II features. (This does not imply that AGN activity cannot be present at a weak or bolometrically-neglectable level, e.g. Ho, Filippenko \\& Sargent 1997.) At the lowest luminosities there may be a significant contribution from cirrus which is at least partly illuminated by the old stellar population in the galaxies (Telesco 1988, Morel et al. 2000), though such galaxies nevertheless also still obey the starburst radio-far-IR relation (e.g. Condon 1992). We conclude that the galaxies in our sample have their bolometric luminosities dominated by star formation.  The Saunders et al. (1990) local luminosity density implies a local star formation rate of $0.023\\pm0.002$ M$_\\odot$ yr$^{-1}$ Mpc$^{-3}$ (Oliver et al. 1998) assuming a Salpeter initial mass function from $0.1$ M$_\\odot$ to $125$ M$_\\odot$. Our results imply this rises to $0.036\\pm0.009$ M$_\\odot$ yr$^{-1}$ Mpc$^{-3}$ by $z=0.2$ ($95\\%$ confidence), significantly higher than the Tresse \\& Maddox (1998) H$\\alpha$-based estimate of $0.022\\pm0.007$ M$_\\odot$ yr$^{-1}$ Mpc$^{-3}$. (Note that the two Seyfert II spectra in our sample are at the brightest end of the luminosity function where the contribution to the luminosity density is the smallest. The luminosity density is dominated by the galaxies around or below the break in table \\ref{tab:local_lf}.) We believe this is due to two factors: firstly, the H$\\alpha$-based estimates are not immune to extinction effects, even if performing Balmer decrement reddening corrections (e.g. Serjeant, Gruppioni \\& Oliver 1998); secondly, the Tresse \\& Maddox (1998) survey area is sufficiently small to be affected by large-scale structure (e.g. Oliver, Gruppioni \\& Serjeant 1998). ELAIS is immune to these problems, as the cosmic variance over the PHOT survey area is $<20\\%$ at $z<0.3$ (Paper I). These star formation rates are however consistent with extrapolations from those derived from mid-IR data at higher redshift by Flores et al. (1999) and Rowan-Robinson (1997).  The prospects for improving on the results presented here are excellent. A large-scale spectroscopic follow-up of the ELAIS northern areas is currently underway, which will have a major impact on the science analysis of ELAIS. In particular, the ELAIS $15\\mu$m luminosity function will probe higher redshifts than accessible to PHOT at $90\\mu$m, from which the $15\\mu$m luminosity density can constrain the cosmic star formation history (e.g. Serjeant 2000)."
    },
    "0010/astro-ph0010449_arXiv.txt": {
        "abstract": "Observations of the November 1995 Sun crossing of the Saturn's ring-plane made with the 3.6m CFH telescope, using the UHAO adaptive optics system, are presented here. We report the detection of four arcs located in the vicinity of the F ring. They can be seen one day later in HST images. The combination of both data sets gives accurate determinations of their orbits. Semi-major axes range from 140\\,020 km to 140\\,080 km, with a mean of $140\\,060 \\pm 60$ km. This is about 150 km smaller than previous estimates of the F ring radius from Voyager 1 and 2 data, but close to the orbit of another arc observed at the same epoch in HST images. ",
        "introduction": "The Sun's crossing of Saturn's ring-plane is a rare opportunity to observe the unlit face of the rings. Results of observations made at the 3.6m CFH Telescope are reported here. Thanks to the use of the UHAO adaptive optics system (Roddier et al. \\cite{roddier_2000}), high resolution ground-based images were obtained ($\\sim$0.15\\arcsec\\,in August 1995 and $\\sim$0.25\\arcsec\\,in November 1995). The Sun's crossing lasted from November 17$^{th}$ to 21$^{st}$ 1995. It was accompanied by three Earth ring plane crossings (1995 May 22$^{nd}$, August 10$^{th}$, and 1996 February 11$^{th}$). Several teams observed these events (Nicholson et al. \\cite{nicholson_96}, hereafter N96; Bosh \\& Rivkin \\cite{bosh}; Sicardy et al. \\cite{sicardy}; Roddier et al. \\cite{roddier_96a}; Ferrari et al. \\cite{ferrari}).\\\\ The unusual viewing of the rings during these events provides a rare opportunity to detect faint structures  such as small satellites, clumps, or arcs, that are usually lost in the glare of the bright main rings. In this paper, we focus our attention on the azimuthal structure of the F ring. Located at the Roche limit of Saturn and bounded by two shepherding satellites Pandora and Prometheus, the F ring is very intriguing. The Voyager images revealed its complex radial and azimuthal structure including  multiple strands, clumps and braids. Interactions with Saturn's satellites might explain some features, but they are not completely understood.\\\\ During the 1995 ring crossing observing campaign, the F ring  appeared populated with numerous features, either point-like or longitudinally extended. The observation of two point-like objects (S/1995 S1 and S/1995 S3) was reported by Bosh \\& Rivkin (\\cite{bosh}) during the May crossing. During the August crossing, our team discovered at least 6 new features with semi-major axes compatible with the F ring (Roddier et al. \\cite{roddier_96a}; Ferrari et al. \\cite{ferrari}). At least one of them was identified as an arc-like object. During the same period three objects (S/1995 S5, S/19995 S6 and S/1995 S7) were detected by N96. Finally, during the November crossing, two wide arcs ($7\\degr$ and 10\\degr-long) were observed by N96. The brightest was also seen by Poulet et al. (\\cite{poulet}, hereafter P2000). The most striking aspect of those discoveries is that no evident correlation can be found between features observed in May, August and November (N96, Bosh \\& Rivkin 1996, P2000).  In the present paper, we report the detection of 4 arcs in the region of the F ring on November 20$^{th}$. The same features can be seen in some HST images taken 24 to 36 hours later. These images have been already presented in N96. By combining both sets of data, accurate orbital solutions are derived. The data processing of CFHT and HST images is presented in the first section. Results are presented in the second section and discussed in the final section. ",
        "conclusions": "We have reported observations of the crossing of Saturn's ring-plane by the Sun, on 1995 November 20$^{th}$, made with the 3.6m CFH Telescope, using the UHAO adaptive optics system. Four azimuthally extended structures (A,B,C,D) have been detected, with central longitude at epoch of 123\\degr, 136\\degr, 177\\degr\\, and 193\\degr\\ (see Table~\\ref{table2}), on 1995 November 21.5 TDT at Saturn. These structures have been also seen one day later in HST images (N96). The combination of both CFHT and HST data sets provides an accurate estimate of their orbit. A circular orbital fit locates them between 140\\,020 km and 140\\,080 km, with error bars of 120 km on average. The mean circular orbit is 140\\,060 $\\pm$ 60 km, that is consistent with the orbit of another bright arc observed at the same epoch (P2000). However, this represents a significant discrepancy ($3\\ \\sigma$) with previous estimates of the F ring's radius derived from Voyager images (located at $140\\,200$\\,km on average). If the five arcs observed in November 1995 belong to a F ring core, this would imply a radial restructuring or a spreading of the F ring between 1980 and 1995, whose explanation has still to be found. In order to understand the F ring's dynamics, better spatial resolution and longer time baselines observations are required. The Cassini encounter in 2004 will probably be a major step in the understanding of the F ring mysteries."
    },
    "0010/astro-ph0010480_arXiv.txt": {
        "abstract": "Analysis of the peculiar velocities of galaxies should take account of the uncertainties in both redshifts and distances. We show how this can be done by a numerical application of the action principle.  The method is applied to an improved catalog of the galaxies and tight systems of galaxies within 4$h_{75}^{-1}$~Mpc, supplemented with a coarser sample of the major concentrations at 4$h_{75}^{-1}$~Mpc to 20$h_{75}^{-1}$~Mpc distance. Inclusion of this outer zone improves the fit of the mass tracers in the inner zone to their measured redshifts and distances, yielding best fits with reduced $\\chi ^2$ in redshift and distance in the range 1.5 to 2.  These solutions are based on the assumption that the galaxies in and near the Local Group trace the mass, and a powerful test would be provided by observations of proper motions of the nearby galaxies.  Predicted transverse galactocentric velocities of some of the nearby galaxies are confined to rather narrow ranges of values, and are on the order of 100 km~s$^{-1}$, large enough to be detected and tested by the proposed SIM and GAIA satellite missions. ",
        "introduction": "This paper continues the analysis of the dynamics of relative motions of the galaxies in and near the Local Group. In previous numerical applications of the action principle either the present distances of the assumed mass tracers are given, and the measure of goodness of fit is the difference between model and catalog redshifts (\\cite{Peeb89} 1989), or the present redshifts are given, and the measure of goodness of fit is the difference between model and catalog distances.\\footnote{The use of variants of the action principle to find solutions to the equation of motion with given present redshifts rather than distances is discussed by \\cite{Maro} (1993), \\cite{Peeb94} (1994), and \\cite{SPT} (1995). We use a canonical transformation of the action, which was introduced, we believe independently, in \\cite{Schmoldt} (1998), \\cite{Whiting} (2000), and \\cite{SDP} (2000).} Since errors in redshift and distance both are important we have adapted the numerical action method to allow relaxation to a stationary point of the action and a minimum of the sum of the squares of the weighted differences of model and catalog redshifts and distances. The method is described in \\S~3.  In \\S~4 we present an application of the method to the catalog discussed in \\S~2. Recent advances in measurements of distances of nearby galaxies are incorporated in a catalog of the distances and radial velocities of the galaxies and tight concentrations of galaxies within 4$h_{75}^{-1}$~Mpc. (Hubble's constant is $H_o=75h_{75}$ km~s$^{-1}$~Mpc$^{-1}$). This inner zone catalog is supplemented by a coarser catalog of distances and radial velocities of the main galaxy concentrations in an outer zone at 4--20$h_{75}^{-1}$~Mpc distance. We work with the assumption that the luminosities of the galaxies trace the mass. Two control cases support this assumption: (i) inclusion of the outer zone in the dynamical solutions improves the fit to the measured redshifts and distances in the inner zone, and (ii) replacing the angular positions by random positions on the sky, while keeping the catalog redshifts and distances, worsens the fit.  The application presented here is meant to illustrate the new numerical method and catalogs; we hope to present a systematic study of parameters in a later paper. With the parameter choices for cosmology and mass-to-light ratios ($M/L$, using blue-band light) listed in \\S~4 we get values of the reduced $\\chi^2$ for distances and redshifts that in the best cases are between 1.5 and 2, close enough to unity to add to the evidence that our model is a useful approximation to reality. A considerably stronger test, from measurements of transverse motions of nearby galaxies, is discussed in \\S~5. ",
        "conclusions": "\\subsection{The Assumption Galaxies Trace Mass}  If the nearby galaxies trace the mass on scales $\\gsim 100$~kpc then our inner catalog in Table~1 gives a reasonably complete description of the mass within 4$h_{75}^{-1}$~Mpc, and Table~2 gives a first approximation to the effect of external masses on the Local Group and its immediate neighbors. We have found dynamical solutions for the orbits of the assumed mass tracers that end up in our inner zone (within 4$h_{75}^{-1}$~Mpc distance); in the best cases the rms difference between model and catalog redshifts and distances is in the range 1.25 to 1.5 times the nominal catalog errors. The better fits use a more liberal estimate for the redshift errors, an issue that requires further consideration. The difference of the reduced $\\chi ^2$ from unity is considerably larger than statistical, but small enough to argue for the assumption galaxies trace mass.  One of our controls keeps the catalog distances, redshifts, and masses but places the objects at random positions in the sky. Since this preserves the near Hubble flow of the more distant members of the inner zone and the infall of most of the closer members, it is not surprising that we get a rough fit. But this scrambling of the present angular positions worsens the fit, as expected if galaxies trace mass. This result is quite stable under changes in the random numbers for the scrambling and computation.  Our second control removes the outer masses from the dynamics. It shows that the outer masses improve the fit to the measured redshifts and distances in the inner zone, again as expected if galaxies trace mass. This also suggests our solutions for the inner zone would be improved by using a better model for the external mass distribution, another point that requires further consideration.  Figures 6 to 8 show that, if galaxies trace mass, there are solutions with similar values of $\\chi ^2$ and transverse velocities that differ by 300 km~s$^{-1}$. These solutions tend to occupy restricted regions of transverse velocity space. The predictions for the galactocentric transverse velocity of M31 are only weakly perturbed by the masses beyond 4$h_{75}^{-1}$~Mpc. Thus we suspect that under the assumption galaxies trace mass the allowed velocity of M31 is rather tightly constrained by the present catalogs. A measurement of the transverse velocity of this and other nearby galaxies could critically test the picture of the local mass distribution.  \\subsection{Comments on Future Work}  This new version of the numerical action method could and certainly should be applied to numerical N-body simulations. We hope to continue work on this topic. Our exploration of the effect of adjusting the mass-to-light ratios in the data has been quite limited, and here too we hope to be able to report on the values of $M/L$ for late and early type systems that yield the best fit to catalog redshifts and distances, to compare to global measures of the density parameter.  The allowance for errors in both redshift and distance is more realistic than previous applications of the action method, but it still ignores the errors in masses. A study of this issue certainly will be part of a satisfactory  assessment of the dynamics of the nearby galaxies.  A fundamental limitation of our numerical application should be noted. The matrix inverse of the second derivatives of the action with respect to the coordinates of an orbit gives the rate of change of the present distance with respect to the redshift (eq.~[\\ref{eq:dddz}]), and it reaches saddle points and local maxima as well as the local minima that are reached by following the gradient of the action. Because the computation time for the matrix inverse varies as the cube of the matrix size we relax the coordinates for each particle separately, toward a stationary point of the action and a minimum of the single-particle $\\chi ^2$. This is iterated through all particles many times. We are really interested in the minimum of the sum of the $\\chi ^2$ for all particles, which need not minimize any single-particle $\\chi ^2$. The inversion of the much larger matrix of all coordinates of all particles would be a heavy computation, but worth attempting.  The galaxy NGC~6822 illustrates the problem of dealing with complex orbits. We have found reasonably close fits to its positive radial velocity by relaxing from random orbits, but it would be feasible and useful to develop a more systematic sample of the many allowed orbits for such cases. The results would challenge us to decide which solutions could have come from realistic initial conditions; we need something better than the simple estimator in equation~(\\ref{eq:deltao})."
    },
    "0010/astro-ph0010119_arXiv.txt": {
        "abstract": "We present new mid-infrared (N-band) images of a sample of eight nearby Seyfert galaxies.  In all of our targets, we detect a central unresolved source, which in some cases has been identified for the first time. In particular, we have detected the mid-infrared emission from the active nucleus of NGC\\,4945, which previously remained undetected at any wavelength but hard X-rays. We also detect circumnuclear extended emission in the Circinus galaxy along its major axis, and find marginal evidence for extended circumnuclear emission in NGC\\,3281.  The high spatial resolution ($1.7\\as$) of our data allows us to separate the flux of the nuclear point sources from the extended circumnuclear starburst (if present).  We complement our sample with literature data for a number of non-active starburst galaxies, and relate the nuclear N-band flux to published hard ($2-10\\kev$) X-ray fluxes. We find tight and well-separated correlations between nuclear N-band flux and X-ray flux for both Seyfert and starburst nuclei which span over 3 orders of magnitude in luminosity. We demonstrate that these correlations can be used as a powerful classification tool for galactic nuclei.  For example, we find strong evidence against NGC\\,1808 currently harbouring an active Seyfert nucleus based on its position in the mid-infrared-X-ray diagram. On the other hand, we confirm that NGC\\,4945 is in fact a Seyfert~2 galaxy. ",
        "introduction": "The classification of ``active'' galaxies based on their optical spectra alone provides an incomplete or even deceiving description of the true nature ot this class of objects.  In some ``classical'' starburst galaxies, X-ray and VLBI observations have found indications for the presence of an active galactic nucleus (AGN) \\citep*{mas99, vig99, smi98}. Mid-infrared (MIR) spectroscopy with ISO has shown that some galaxies, which were optically classified as AGNs, are in fact dominated by starbursts. Conversely, the presence of a heavily obscured AGN was revealed in some systems previously classified as starburst \\citep*{gen98,lut99a,rig99}.  Mid-IR observations are particularly useful for detecting dust-enshrouded AGNs for a number of reasons. Most importantly, the MIR is much less affected by dust extinction than, e.g., the optical regime \\cite[$A_{8-13\\mum}\\approx0.06\\cdot \\av$,][]{lut96}. While the extinction is even lower at far-IR wavelengths, warm dust emission in the  $60-100\\mum$ regime can be powered both by AGNs and starbursts and is therefore not distinctive of the nature of activity. In contrast, significant emission from hot dust at wavelengths between 2 and $10\\mum$ is generally characteristic of heating by an active nucleus \\citep{oli95,mai98,moo96a,lut99a,cla00}.  Finally, the MIR allows diffraction-limited, arcsecond-scale imaging from large ground-based telescopes, in contrast to observations at far-IR wavelengths which have to rely on space observatories with limited aperture sizes and hence limited resolution. As a case in point, MIR spectroscopy with ISO is often unable to detect the nuclear hot dust emission, since the host galaxy and circumnuclear starburst emission dilute the nuclear AGN emission in the large beam of the ISO observations.  A classical example is NGC\\,4945. This galaxy hosts one of the brightest AGNs at $100\\kev$, although the X-ray source is heavily absorbed \\cite*[$\\rm N_{H}\\approx 5\\cdot 10^{24}\\>cm^{-2}$,][]{gua00}. However, both optical and near-IR data do not show any indication for the presence of the active nucleus and appear, instead, dominated by a powerful nuclear starburst along with its superwind cavity \\citep{moo96b,mai99,mar00}. Even the MIR ISO spectra appear dominated by the starburst \\citep{spo00}.  Because of the reasons outlined above, sensitive MIR observations with high spatial resolution hold the promise to identify heavily obscured AGNs within starburst galaxies, even if they are missed in optical and near-IR imaging and/or in ISO spectroscopy, both because of heavy obscuration or dilution by the nuclear starbursting activity. If such MIR observations were to find that the fraction of galaxies with AGNs is significantly higher than estimated from optical, near-IR, or ISO MIR surveys, it would provide support to the idea of a connection between starbursts and AGNs, which has been suggested by various studies \\citep{hec97,mai95}. ",
        "conclusions": "We have presented new N-band maps of a sample of 8 nearby Seyfert galaxies. In all of our targets, we have identified an unresolved nuclear emission source, some of which were previously undetected. In those cases where additional extended emission is present, we have separated the unresolved component by means of PSF fitting. This demonstrates the importance of high resolution MIR imaging for detecting AGN and isolating them from the extended emission component.  After complementing our sample by literature data for a number of starburst galaxies and some additional Seyferts, we have compared the MIR luminosity of the nuclear sources to their hard X-ray emission. We find that both ``pure'' starbursts and AGN follow a tight MIR--X-ray correlation, but the two populations are separated by about a factor of 15 without and 190 with aperture correction in their hard X-ray fluxes. We propose to use this correlation as a powerful diagnostic tool to distinguish between these two classes of objects in cases where evidence for nuclear activity is ambiguous."
    },
    "0010/astro-ph0010269_arXiv.txt": {
        "abstract": "Using Eggleton's stellar evolution code, we carry out 150 runs of Pop I binary evolution calculations, with the initial primary mass between 1 and $8M_\\odot$, the initial mass ratio $q=M_1/M_2$ between 1.1 and 4, and the onset of Roche lobe overflow (RLOF) at an early, middle, or late Hertzsprung-gap stage. We assume that RLOF is conservative in the calculations, and find that the remnant mass of the primary may change by more than 40 per cent over the range of initial mass ratio or orbital period, for a given primary mass. This is contrary to the often-held belief that the remnant mass depends only on the progenitor mass if mass transfer begins in the Hertzsprung gap. We fit a formula, with an error less than 3.6 per cent, for the remnant (white dwarf) mass as a function of the initial mass $M_{\\rm 1i}$ of the primary, the initial mass ratio $q_{\\rm i}$, and the radius of the primary at the onset of RLOF. We also find that a carbon-oxygen white dwarf with mass as low as $0.33M_\\odot$ may be formed if the primary's initial mass is around $2.5M_\\odot$. ",
        "introduction": "Binary evolution plays an important role in the formation of many interesting stellar objects and binary evolution theory has been successful in solving many puzzles in the past. In particular, double degenerates, as possible supernovae Type Ia progenitors \\cite{ibe84,web84}, are receiving more and more attention. The confrontation of the double degenerate theory of Iben et al.\\ \\shortcite{ibe97} and Han \\shortcite{han98} with the observations of Marsh's group \\cite{mar00} shows a conflict between the theoretical and observational mass ratio distribution of double degenerates. An Algol-like phase is an important or even the main channel for the formation of double degenerates. The confrontation indicates that we need a systematic investigation of binary evolution with the onset of Roche lobe overflow (RLOF) in the Hertzsprung gap (HG).  Case-B binary evolution, in which mass transfer begins after central hydrogen exhaustion, has been studied  in the past. Refsdal \\& Weigert \\shortcite{ref69} and Giannone \\& Giannuzzi \\shortcite{gia70} calculated several runs of case B evolution for low-mass binaries. Van der Linden \\shortcite{van87} calculated conservative evolution for initial primary masses  between 3 and $12M_\\odot$. De Loore \\& De Greve \\shortcite{del92}, De Greve \\& de Loore \\shortcite{deg92}, De Greve \\shortcite{deg93} and de Loore \\& Vanbeveren \\shortcite{del94a,del94b,del95} carried out non-conservative evolution (conservative evolution in some cases) for initial primary masses between 3 and $40M_\\odot$. Sarna et al.\\ \\shortcite{sar96} investigated evolutionary scenarios for double degenerates. The evolutionary sequences calculated previously mostly cover different initial primary masses, initial orbital periods that make RLOF begin in the early HG, and only one or two initial mass ratios. Furthermore there is a lack of low-mass binary sequences.  A binary population synthesis of double degenerates \\cite{han98} requires a detailed knowledge of binary evolution. In this paper, we investigate binary evolution with the onset of RLOF in the Hertzsprung gap in a systematic way. We calculate 150 runs of binary evolution with a rather complete parameter space, i.e. varying the primary's initial mass between 1 and $8M_\\odot$, the initial mass ratio $q=M_1/M_2$ between 1.1 and 4, and the onset of Roche lobe overflow (RLOF) at early, middle, or late Hertzsprung gap. ",
        "conclusions": "Van der Linden \\shortcite{van87} and de Loore \\& Vanbeveren \\shortcite{del95} have calculated conservative Case-B evolution of intermediate mass binaries, though only for one or two mass ratios. However, it is not possible to compare our results with theirs in a one-to-one way, as none of their binaries has exactly the same initial parameters (primary mass, mass ratio and orbital period) as any of ours. But we do make a comparison and find that our results agree with theirs reasonably for binaries with initial primary mass less than $4M_\\odot$. The difference between our results and theirs is small and is mainly due to the following differences between our work and theirs. We use the OPAL opacity tables \\cite{rog92} while van der Linden used the LAOS ones. The treatment of RLOF in our code is different from theirs. Ours is more efficient in computation, but theirs may be more realistic physically. We do not include convective core overshooting but de Loore \\& Vanbeveren \\shortcite{del95} do. And, perhaps more importantly, our stellar evolution code is different from theirs.  For binaries with initial primary mass larger than $5M_\\odot$, there exist very large differences between our results and theirs. For example, the hydrogen abundance at the primary's surface is almost zero after RLOF in our model for a binary with initial parameters of $M_{\\rm 1i}=5M_\\odot$, $q_{\\rm i}=1.1$ and $P_{\\rm i}=22.52$ days, while the hydrogen abundance of a similar binary in their model is about 0.2. The timescale of RLOF in our model is much larger than that in their model. This is explained as follows. For a binary with its primary initially more massive than $5M_\\odot$, the primary fills its Roche lobe in the Hertzsprung gap and RLOF begins. The primary becomes a red giant during the RLOF, and its envelope is stripped away gradually. When the envelope is very small, the primary contracts and becomes a helium star (but with a hydrogen-helium envelope). Helium is ignited in the centre and the star expands after central helium is exhausted. The star becomes a red giant again and fills its Roche lobe and the second episode of RLOF begins. At the end of the second episode of RLOF, the surface hydrogen abundance becomes zero obviously. Van der Linden \\shortcite{van87} and de Loore \\& Vanbeveren \\shortcite{del95} have only followed the first episode of RLOF and they stop the evolution when the central helium is exhausted. We have followed the two episodes and our calculation is terminated when the primary becomes a helium WD or carbon-oxygen WD in most cases. Therefore the timescale of RLOF in our model (the time between the onset of the first episode of RLOF and the end of the last episode by our definition) is much larger than that of van der Linden \\shortcite{van87} and de Loore \\& Vanbeveren \\shortcite{van87}.  Most of the double degenerates observed by Marsh's group \\cite{mar00} have primary masses less than $0.5M_\\odot$ and they are thought to be helium WDs. Our calculation shows that a low mass (as low as $0.33M_\\odot$) carbon-oxygen WD may be formed from stable RLOF in HG, so that some of the presumed helium WDs may actually be carbon-oxygen WDs.  In order to get better consistency with observations, Iben et al.\\  \\shortcite{ibe97} use a rather large common-envelope ejection efficiency in their binary population synthesis studies. They assume that RLOF in the HG is always dynamically unstable and that RLOF leads to the formation of a common envelope. Our calculation shows however that RLOF is very likely to be stable, even if the initial mass ratio is as large as 3. If the initial mass ratio is even larger, RLOF is stable at the beginning and part of the primary's envelope is transferred to the secondary before the RLOF becomes dynamically unstable. If the process of RLOF is treated as common-envelope evolution, as by Iben et al., a high efficiency is required for common-envelope ejection.  Most of the previous studies treated binaries with only one or two initial mass ratios and with the onset of RLOF early in the HG only. Our study is more comprehensive in that our calculation is for a rather complete space of initial parameters. It has been believed that the WD mass derived from the RLOF does not depend much on the initial mass ratio or the initial orbital period \\cite{deg93}. However, our investigation shows that the dependence is quite significant. Therefore the binary evolutionary models in this paper should be included in any realistic binary population synthesis model.  To obtain a copy of all the evolutionary tracks (similar to figure \\ref{2q15p3hrd}), all tables for characteristics of RLOF (similar to table \\ref{2q15p3tab}), and equations \\ref{mif} to \\ref{mifr2} in FORTRAN code, please send an email to ZH who will provide them via ftp."
    },
    "0010/astro-ph0010575_arXiv.txt": {
        "abstract": " ",
        "introduction": "The cosmologically most important cluster parameter is its gravitational mass. In order to constrain cosmological models of structure formation by comparing simulations or analytic models with, e.g., observational mass functions of statistical cluster samples, cluster masses are needed. Precise cluster mass determinations for large cluster surveys, e.g.~\\cite{bs00}, however, require extensive observations and are currently not feasible. An alternative approach is to use established relations between more easily observed quantities and cluster mass. For optical cluster samples the observed number of cluster galaxies has sometimes been used to relate to cluster mass. For X-ray cluster samples a three component approach is commonly used. First, measured X-ray luminosities are converted to X-ray temperatures using an observational relation. Then, a slope for the X-ray temperature -- gravitational mass relation is assumed on theoretical grounds. And finally, the X-ray temperatures are converted to masses by normalizing this relation with hydrodynamical simulations.  Here we show that cluster surveys selecting clusters by their X-ray luminosities effectively select by cluster mass. It is shown that a selection based solely on Abell galaxy richness is less efficient in terms of mass. Furthermore we quantify the empirical X-ray luminosity -- gravitational mass relation and its scatter, offering the possibility to directly use this observed relation in wide angle X-ray cluster surveys for the luminosity -- mass conversion. Throughout $H_0=50\\,\\rm km/s/Mpc$, $q_0=0.5$ and $\\Lambda=0$ is used. ",
        "conclusions": "In the course of constructing a highly complete X-ray flux-limited sample of the brightest galaxy clusters in the sky (\\gcs\\index{HIFLUGCS}, \\cite{rb99a}) cluster masses for more than 100 clusters have been determined individually from X-ray data. In Fig.~\\ref{lbm} the bolometric ($0.01-40\\,\\rm keV$) X-ray luminosity, $L_{\\rm Bol}$, is plottet versus the gravitational mass within an overdensity of 200 times the critical density, $M_{200}$. \\begin{figure} \\centering \\includegraphics[width=.95\\textwidth]{reiprichf1.eps} \\caption[]{Bolometric X-ray luminosity versus gravitational mass for 106 clusters.} \\label{lbm} \\end{figure} The solid line shows the best fit relation for 106 clusters, the triple-dot-dashed line is the best fit relation determined using the 63 clusters included in the strictly flux-limited sample (\\gcs ), the dot-dashed line is the self-similar relation (slope = 4/3) normalized by the simulations of Navarro, Frenk \\& White \\cite{nfw95}, the dashed line is the `pre-heated'\\index{pre-heating} relation given by Evrard \\& Henry \\cite{eh91} (slope = 11/6). The normalization given by Evrard \\& Henry, determined theoretically, agrees with the normalization found in the simulations of pre-heated clusters by Navarro, Frenk \\& White. The measured relations lie in between the relations with and without pre-heating. The physical implications will be discussed in Reiprich et al.\\ (in prep.), here we emphasize that the plot shows measured and predicted $L_{\\rm Bol}-M_{200}$ relation to be in rough agreement. Using a bisector linear regression fit routine in log--log space which takes into account errors in both variables and allows for intrinsic scatter \\cite{ab96}, the following best fit relations are found using 106 clusters: $L_{\\rm Bol} = 9.25\\times 10^{-23}\\,M_{200}^{1.81\\pm 0.08}$ and for the luminosity in the ROSAT energy band $L_{\\rm X}(0.1-2.4\\,{\\rm keV}) = 1.47\\times 10^{-19}\\,M_{200}^{1.57\\pm 0.08}$ ($L_{\\rm X}$ in units of $10^{40}\\,\\rm erg/s$ and $M_{200}$ in solar masses). Note that different best fit values are obtained when errors in the variables are neglected and only one variable is treated as dependent (e.g., for the latter relation then a slope of 1.33, in agreement with the slope of the preliminary $L_{\\rm X}-M_{500}$ relation quantified by Reiprich \\& B\\\"ohringer \\cite{rb99b}, is found instead of 1.57). The bisector method used here ensures that variables are treated symmetrically \\cite{ifa90}. The 1-$\\sigma$ scatter in log space for the mass around the $L_{\\rm Bol} - M_{200}$ relation equals 0.19, corresponding to a relative mass error of +55\\% and $-$35\\% when converting $L_{\\rm Bol}$ to $M_{200}$. This error includes intrinsic and measurement scatter.  \\begin{figure} \\hspace{-2.5cm} \\includegraphics[width=9.25cm]{reiprichf2.eps} \\hspace{-2.2cm} \\includegraphics[width=9.25cm]{reiprichf3.eps} \\caption[]{Number of cluster galaxies for 66 clusters (left) and galaxy velocity dispersion for 58 clusters (right) versus mass as compared to X-ray luminosity versus X-ray mass for the same clusters.} \\label{lr} \\vspace{-0.3cm} \\end{figure} In Fig.~\\ref{lr} the measured number of cluster galaxies, $N_{\\rm gx}$, as taken from Abell, Corwin \\& Olowin \\cite{aco89} and the measured radial galaxy velocity dispersion, Sigma, as taken from Struble \\& Rood \\cite{sr99} are compared to $L_{\\rm X}$ as gravitational mass tracers. It is clear from the plot on the left that, assuming $M_{200}$ to be a good estimate of the true cluster mass, solely taking the Abell richness to select clusters is not a very effective way to select clusters by their mass. From the plot on the right one can see that the velocity dispersion correlates better with mass than the galaxy richness but also it confirms that careful analyses are required, e.g.~\\cite{ggm98}, when determining the cluster mass from the velocity dispersion.  For further infos and more comprehensive articles please check\\\\ http://www.xray.mpe.mpg.de/$\\sim$reiprich/ \\vspace{-0.27cm}"
    },
    "0010/hep-th0010232_arXiv.txt": {
        "abstract": "This paper investigates Starobinsky's model of inflation driven by the trace anomaly of conformally coupled matter fields. This model does not suffer from the problem of contrived initial conditions that occurs in most models of inflation driven by a scalar field. The universe can be nucleated semi-classically by a cosmological instanton that is much larger than the Planck scale provided there are sufficiently many matter fields. There are two cosmological instantons: the four sphere and a new ``double bubble'' solution. This paper considers a universe nucleated by the four sphere. The AdS/CFT correspondence is used to calculate the correlation function for scalar and tensor metric perturbations during the ensuing de Sitter phase. The analytic structure of the scalar and tensor propagators is discussed in detail. Observational constraints on the model are discussed. Quantum loops of matter fields are shown to strongly suppress short scale metric perturbations, which implies that short distance modifications of gravity would probably not be observable in the cosmic microwave background. This is probably true for any model of inflation provided there are sufficiently many matter fields. This point is illustrated by a comparison of anomaly driven inflation in four dimensions and in a Randall-Sundrum brane-world model. ",
        "introduction": " ",
        "conclusions": ""
    },
    "0010/astro-ph0010561_arXiv.txt": {
        "abstract": "We present a new Unbiased Minimal Variance (UMV) estimator for the purpose of reconstructing the large--scale structure of the universe from noisy, sparse and incomplete data. Similar to the Wiener Filter (WF), the UMV estimator is derived by requiring the linear minimal variance solution given the data and an assumed \\prior\\ model specifying the underlying field covariance matrix. However, unlike the WF, the minimization is carried out with the added constraint of an unbiased reconstructed mean field. The new estimator does not necessitate a noise model to estimate the underlying field; however, such a model is required for evaluating the errors at each point in space. The general application of the UMV estimator is to predict the values of the reconstructed field in un-sampled regions of space (\\eg\\ interpolation in the unobserved Zone of Avoidance), and to dynamically transform from one measured field to another (\\eg\\ inversion of radial peculiar velocities to over-densities). Here, we provide two very simple applications of the method. The first, is to recover a 1D signal from noisy, convolved data with gaps, \\eg\\ CMB time-ordered data. The second application is a reconstruction of the density and 3D peculiar velocity fields from mock SEcat galaxy peculiar velocity catalogs. ",
        "introduction": "\\label{intro}  Mapping the distribution of galaxies and their peculiar velocity field constitutes a major research area in modern astronomy setting both the observational and theoretical foundations of cosmology and, in particular, of large scale structure (LSS).  Within the framework of gravitational instability, the large scale galaxy distribution offers a probe of the nature of the primordial perturbation field, and can be used to set strong constraints on the values of cosmological parameters.  However, since astronomical observations give only incomplete and noisy information on the real universe, the recovery of the underlying signal from these observations can be a non-trivial task, forcing one to resort to regularization methods, \\eg, Wiener filtering (Wiener 1949; Rybicki \\& Press 1992; Zaroubi \\etal\\ 1995) and the Maximum Entropy algorithm (Gull 1989).  In particular, the WF has been widely applied to galaxy surveys (Lahav \\etal\\ 1994; Fisher \\etal\\ 1995; Schmoldt \\etal\\ 2000), CMB studies (Bunn \\etal\\ 1994; Tegmark \\& Efstathiou 1997, Bouchet \\& Gispert, 1999), and galaxy peculiar velocity catalogs (Hoffman \\etal\\ 2000; Zaroubi 2000; Zaroubi \\etal\\ 1999 \\& 2001a). The WF provides an optimal estimator of the underlying field in the sense of a minimum-variance solution given the data and an assumed \\prior\\ model (Wiener 1949; Press \\etal\\ 1992). The \\prior\\ defines the data auto-correlation and the data-field cross-correlation matrices. In the case where the data is drawn from a random Gaussian field, the WF estimator coincides with the conditional mean field and with the most probable configuration given the data (see Zaroubi \\etal\\ 1995).  Although the application of the WF is very simple and has proven to be useful for many purposes, it is easy to show that the estimator is intrinsically biased, often in a scale dependent manner.  The main cause of this bias stems from the modulation introduced by the Signal/(Signal + Noise) weighting it invokes. This drawback has prevented the use of the Wiener reconstructed maps in many areas, \\eg\\ power spectrum estimations, bias parameter extraction from galaxy peculiar velocity data comparison with Galaxy surveys data, etc.  To account for this bias in the power spectrum estimation a correction factor is often applied to the Wiener reconstructed signal (Rybicki \\& Press 1992; Tegmark \\& Efstathiou 1997).  In this paper, we propose a new linear unbiased minimal variance (UMV) estimator that is designed to avoid the intrinsic bias that exists in the WF. This is achieved by solving the minimization equation subject to the constraint of unbiased mean underlying field. To test the UMV estimator we apply it: 1- To recover a 1D time series from convolved and noisy data with gaps. 2- To mock SEcat peculiar velocity data set, a combination of the SFI (Giovanelli \\etal\\ 1998) and the ENEAR (da Costa \\etal\\ 2000) galaxy peculiar velocity catalogs, where we reconstruct the distribution of the true density and velocity fields from which the mock catalog is constructed.   The outline of this paper is as follows.  The method is presented in \\S~\\ref{sec:theory}.  In \\S~\\ref{sec:regularization} two simple extra regularization methods are discussed. \\S~\\ref{sec:CR} shows how constrained realizations could be produced within the UMV framework. The method is tested using artificial data based on simulations in \\S~\\ref{sec:applications}. The results are discussed and the conclusions are summarized in \\S~\\ref{sec:summary}. ",
        "conclusions": "A general framework of linear estimation and prediction by minimal variance subject to a linear constraint in the data has been introduced. The solution of the minimization problem yields the UMV estimator, which is shown to be a very useful tool for reconstructing the large scale structure of the universe from incomplete, noisy and sparse data.  The UMV estimator has been designed to overcome one of the main drawbacks of the WF which is that it predicts the null field in the absence of good data, \\ie\\ in the limit of very poor signal-to-noise data the cosmological mean field is estimated, \\ie\\ zero perturbation. In contrast, the UMV estimator does not alter the values of the reconstructed field at the locations of the data point, because it lacks the filtering aspect of the WF, instead it keeps the values of the measured data at the locations of the data points and interpolates between them in accordance with the correlation function assumed in the model. Like the WF the new estimator can be used for dynamical reconstruction, \\ie\\ to reconstruct one dynamical field, \\eg\\ over-density, from another measured field, \\eg\\ radial peculiar velocity. These two properties make the UMV estimator a very appealing tool for various applications in LSS and CMB problems.  The regularization strength of the new estimator stems from the cross correlation between the signal and the data points at different locations. Which allows a reconstruction that is consistent with the data points and the correlations among them. Lacking such a correlation, the UMV estimator is equivalent to direct inversion. In this case an additional regularization is required.  This issue could be viewed as follows. The UMV produces an unbiased estimate of the underlying signal subject to the minimal variance requirement. This minimal variance could sometimes be too large to be useful. In comparison the WF produces an {\\it truly} minimal variance but with a biased mean. Therefore, when applying the UMV estimator, sometimes an extra regularzation is needed in order to, on the one hand, reduce the variance in the UMV reconstruction to manageable value, while keeping the reconstructed underlying signal as unbiased as possible, on the other. In this paper two extra regularization methods, that are both simple and easily integrated in the UMV method, have been discussed.  Constrained realizations of the underlying field (Hoffman \\& Ribak 1991) will not be possible with the UMV estimator. However, constrained realizations of a noisy underlying field are possible. Such noisy constrained realizations could be very useful if one wishes to fill gaps in the data, \\eg\\ from balloon-born CMB measurements, with a realization that has the same assumed properties of the Signal+Noise.  An apparent difficulty arises from the fact that the current UMV reconstruction assumes linear gravitational instability, yet it is applied to a universe that is {\\it non-linear} on scales smaller than a few Mpc's.  To obtain non-linear reconstruction of the underlying field, $\\bf s_{nl}$, one can include the non-linear correlation by substituting $\\langle {\\bf s_{nl} o^+} \\rangle$ in eq.~\\ref{eq:estimator}. This of course would only take into account the contribution of the variance and ignore the contribution of higher moments but in many cases this will do.  In one of the exampled presents here, we have applied the method to galaxy radial peculiar velocity data and shown that the reconstruction is unbiased and trustworthy within a very large region of the volume covered by the data. In this context the UMV reconstruction could be viewed as a compromise between the POTENT algorithm (Bertschinger \\& Dekel 1989; Dekel, Bertschinger \\& Faber 1990), which assumes no regularization, and the WF, which relies too heavily on it.  The UMV reconstruction of the over-density and the 3D velocity field from galaxy peculiar velocity catalogs is suitable for bias parameter extraction from a comparison with the respective fields obtained from redshift galaxy catalogs like the PSCz (Saunders \\etal\\ 2000; Branchini \\etal\\ 2000). The current estimator allows carrying out density-density and velocity-velocity comparisons using the same reconstruction technique (Zaroubi \\etal\\ 2001b)."
    },
    "0010/astro-ph0010082_arXiv.txt": {
        "abstract": "We study the behaviour of inflationary density perturbations in the vicinity of horizon crossing, using numerical evolution of the relevant mode equations. We explore two specific scenarios. In one, inflation is temporarily ended because a portion of the potential is too steep to support inflation. We find that perturbations on super-horizon scales can be modified, usually leading to a large amplification, because of entropy perturbations in the scalar field. This leads to a broad feature in the power spectrum, and the slow-roll and Stewart--Lyth approximations, which assume the perturbations reach an asymptotic regime well outside the horizon, can fail by many orders of magnitude in this regime. In the second scenario we consider perturbations generated right at the end of inflation, which re-enter shortly after inflation ends --- such perturbations can be relevant for primordial black hole formation. ",
        "introduction": "The inflationary cosmology remains the leading candidate theory for the origin of structure (see Refs.~\\cite{LLrep,LLbook} for reviews).  Considerable attention has been focussed on producing highly accurate calculations of the perturbations produced during inflation \\cite{perts}, which ultimately may be measured at the percent level via the microwave anisotropies they induce \\cite{measure}. Special interest has been given to the simplest case, where there is only a single scalar field degree of freedom during inflation, and we will focus exclusively on that case in this paper.  In considering how a perturbation might evolve, a crucial quantity is the comparison of the inverse wavenumber with the Hubble length, given by the ratio $aH/k$. By definition inflation corresponds to a decreasing comoving Hubble length, $d(aH)/dt >0$; consequently when modes of a given wavenumber are considered, they begin their evolution well inside the horizon\\footnote{We use `horizon' as shorthand for the more precise `Hubble length'; it does not refer to the particle horizon.} and cross outside during inflation. At early times, flat space-time quantum field theory can be used to fix the initial normalization of the perturbations. Around the epoch of horizon crossing, the perturbation becomes frozen in, corresponding to a constant curvature perturbation once the mode is well outside the horizon. This leads to standard expressions for the spectrum of density perturbations such as \\begin{equation} \\label{eqn:scalpert} {\\cal P}_{{\\cal R}}^{1/2}(k) = \\left. \\frac{1}{2\\pi} \\, \\frac{H^2}{|\\dot{\\phi}|} \\right|_{k=aH} \\,, \\end{equation} where ${\\cal P}_{{\\cal R}}$ is the power spectrum of the curvature perturbation ${\\cal R}$, using the notation of Refs.~\\cite{LLrep,LLbook}.  This expression gives the amplitude of the perturbations in terms of the values of the Hubble parameter $H$ and scalar field velocity $\\dot{\\phi}$ at the time the mode crossed the horizon, i.e.~when $k=aH$. A similar expression holds for the gravitational wave amplitude, and more sophisticated higher-order versions have also been derived.  Eq.~(\\ref{eqn:scalpert}) gives the impression that the amplitude of perturbations is being quoted at the instant of horizon crossing. However, it is important to realize that that is not in fact the case. The value quoted is the perturbation amplitude attained in the asymptotic limit $k/aH \\rightarrow 0$ when the perturbation is well outside the horizon; it just happens to be written in terms of the values the background parameters had at the instant of horizon crossing. In fact the value of the curvature perturbation at the instant of horizon crossing typically differs by a significant factor from the asymptotic value.  Bearing that in mind, in this paper we consider two circumstances where the standard formula for the perturbation amplitude may not be valid, due to a failure to reach the asymptotic limit. In each case, this is because inflation ends before the true asymptotic regime has been reached. In our first scenario, we consider a temporary interruption to inflation, where the field driving inflation has a region where the potential is too steep to sustain inflation. We will see that this can lead to significant modifications to the standard results, and indeed that even modes significantly outside the horizon can receive a large change in amplitude due to entropy perturbations in the scalar field. In the second scenario, we consider perturbations produced at the end of inflation, where their amplitude on re-entry has relevance to the production of primordial black holes. ",
        "conclusions": "The accuracy of the usual analytic expressions for the inflationary power spectrum depends on scales evolving smoothly through horizon crossing and into the asymptotic regime $k^2 \\ll z''/z$. In this paper, we have investigated two situations where this is not achieved, one being a temporary end to inflation and the other the true end. In the former case we have seen that the modes can have a very complicated evolution, including the possibility of amplification on super-horizon scales via an exponential driving term corresponding to an entropy perturbation in the scalar field. Such behaviour can be traced to the complicated evolution of the scalar pump field $z$, which no longer grows monotonically. The net effect is the insertion of a broad feature into the power spectrum that can only be computed by mode-by-mode integration, and which can differ wildly from the slow-roll and Stewart--Lyth approximations. Typically the features are sufficiently non-scale-invariant to be excluded already on astrophysical scales, but we have also seen that a very flat spectrum can be obtained while far from slow-roll, confirming an analytic analysis by Wands~\\cite{W}.  The second scenario we studied was the true end of inflation, where the failure to reach an asymptotic regime means that perturbations re-enter the horizon with a higher amplitude. The most interesting physical consequence of such modes is in primordial black hole formation, and this result indicates that earlier treatments assuming the slow-roll formulae are somewhat conservative, as the mode equation solutions indicate that these formulae underestimate the perturbation amplitude at horizon re-entry."
    },
    "0010/astro-ph0010427_arXiv.txt": {
        "abstract": "We present the results of a series of imaging and spectroscopic observations aimed at identifying and studying the galaxy responsible for the $z_{\\rm abs} = 3.38639$ damped \\lya\\ system in the $z_{\\rm em} = 3.61$ QSO  Q0201+1120. We find that the DLA is part of a concentration of matter which includes at least four galaxies (probably many more) over linear comoving dimensions greater than $5h^{-1}$~Mpc. The absorber may be a $0.7 L^{\\ast}$ galaxy at an impact parameter of $15 h^{-1}$~kpc, but follow-up spectroscopy is still required for positive identification. The gas is turbulent, with many absorption components distributed over $\\sim 270$~km~s$^{-1}$ and a large spin temperature, $T_s \\simgt 4000$~K. The metallicity is relatively high for this redshift, $Z_{\\rm DLA} \\simeq 1/20 Z_{\\odot}$. From consideration of the relative ratios of elements which have different nucleosynthetic timescales, it would appear that the last major episode of star formation in this DLA occurred at $z \\simgt 4.3$, more than $\\sim 500$~Myr prior to the time when we observe it. ",
        "introduction": "The study of quasar absorption lines is a unique tool with which to investigate the high redshift universe on a variety of scales, ranging from the intergalactic medium (IGM) to proto-galaxies. Since the detection of material intercepting our line of sight to a given QSO is dependent only on the column density of gas and the luminosity of the quasar, this is a particularly powerful and sensitive technique for probing the chemical composition and physical conditions in the interstellar media (ISM) of high $z$ galaxies.  Damped Lyman Alpha systems (DLAs) are the highest column density ($N$(H~I) $\\geq 2 \\times 10^{20}$ atoms \\cm) absorbers identified in QSO spectra and the major contributors to the census of neutral gas available for star formation. DLAs are therefore widely believed to be the absorption signatures of galaxies in early stages of evolution; the study of their associated metal lines has provided us with the first accurate measurements of the abundances of a wide variety of chemical elements at high $z$ (Lu et al. 1996; Prochaska \\& Wolfe 1999). Indeed, measuring the abundances in the ISM of DLAs not only gives a direct view of the evolving chemistry of galaxies over time, but is also a valuable testbed for supernova models and their theoretical yields (e.g. Pettini et al. 2000).  However, the exact nature of DLAs is still a contentious issue. Results from early DLA surveys (e.g. Wolfe et al. 1986; Lanzetta et al. 1991) led to the proposal that these systems trace the progenitors of present day spirals.  More recently, Prochaska \\& Wolfe (1997; 1998) have argued that this hypothesis is further supported by the kinematics of the metal absorption lines which they can best explain with large rotating disks with velocities $v_{\\rm rot} \\simlt 200$~\\kms\\ (although see Ledoux et al. 1998). On the other hand, the presence of such massive disks at $z \\sim 3$ is in stark contrast with the predictions of hierarchical CDM models (e.g. Kauffmann 1996) where galaxies at high redshifts are relatively compact and underluminous.  In such models, the kinematics of merging proto-galactic clumps can also reproduce the shapes of the absorption lines (Haehnelt, Steinmetz, \\& Rauch 1998; McDonald \\& Miralda-Escud\\'{e} 1999; Gardner et al. 2000). From the point of view of their chemical enrichment, it was shown by Pettini et al. (1997) that the distribution of DLA metallicities is displaced to lower values relative to those those of the thin and thick disk stellar populations of the Milky Way (see also Wolfe \\& Prochaska 1998).  At intermediate redshifts the picture is a little clearer. Direct imaging with {\\it HST} of galaxies associated with DLAs at $z \\simlt 1$ has revealed a range of morphologies with only a minority of spirals (Le Brun et al. 1997; Rao \\& Turnshek 1998; Pettini et al. 2000). However, there are still concerns regarding sample bias. An absorption system at $z \\simlt 1.5$ can only be recognised as a DLA with {\\it HST} ultraviolet spectroscopy (or $21\\,$cm absorption, but such systems are rare). Thus QSOs behind metal- and presumably dust-rich DLAs may well be under-represented in existing surveys, given the bright magnitude limit for spectroscopy imposed by the relatively small aperture of {\\it HST\\/}. Since $B$-band luminosity and metallicity are related (e.g. Kobulnicky \\& Zaritsky 1999), {\\it HST}-selected DLAs are likely to be biased in favour of absorbers of low luminosity and metallicity.  In order to make further progress on these issues, it is necessary to bring together all the information provided by deep imaging, kinematics, and chemical abundance studies into one coherent picture. In this paper we explore this approach for a high redshift DLA in the direction of the QSO Q0201+1120 which we have targeted with a number of different observations.  The compact, flat spectrum, radio source PKS~0201+113 with a faint ($R = 19.5$) optical counterpart (Condon, Hicks, \\& Jauncey 1977) was identified as a $z_{\\rm em} = 3.61$ QSO by White, Kinney, \\& Becker (1993) who also reported the presence of a strong DLA at $z_{\\rm abs} = 3.3875$ with an estimated $N{\\rm (H~I)} = 2.5 \\times 10^{21}$~cm$^{-2}$. This value of the neutral hydrogen column density is sufficiently high to produce absorption in the redshifted 21~cm line against the background radio source (de Bruyn, O'Dea, \\& Baum 1996; Briggs, Brinks, \\& Wolfe 1997), although there is still some uncertainty about the reality of these weak detections (Kanekar \\& Chengalur 1997). Here we present deep imaging, both from the ground and with {\\it HST}, of the field of Q0201+1120 aimed at detecting the galaxy responsible for the DLA. In addition, we have obtained intermediate and low resolution spectroscopy of the QSO and of a number of faint sources in its proximity revealed by our images. High resolution spectroscopy of the DLA is used to study its chemical composition and velocity structure. All these data are scrutinised for clues to the nature of the object producing the damped \\lya\\ system.  Unless otherwise stated, we use a $\\Omega_M = 0.3$, $\\Omega_{\\Lambda} = 0.7$ cosmology; $h$ is the Hubble constant in units of 100~km~s$^{-1}$~Mpc$^{-1}$; and all redshifts reported in this paper are vacuum heliocentric. \\\\ ",
        "conclusions": "In this paper we have presented imaging and spectroscopic observations aimed primarily at investigating the nature of the absorber giving rise to the strong damped \\lya\\ system at $z_{\\rm abs} = 3.38639$ in front of Q0201+1120. Our main findings can be summarised as follows.  1. The DLA is part of a concentration of matter which includes at least four galaxies at $z \\simeq 3.37$ over transverse dimensions of at least $5 h^{-1}$~Mpc (comoving). Two of the galaxies are seen directly via their UV stellar continua, while two others (including the DLA) are detected in absorption against higher redshift sources. Since we have so far obtained spectra for only about one quarter of the $U$-drop candidates in this field, it is likely that the structure includes other objects at $z \\simeq 3.37$.  2. We have found a promising candidate for the absorber in a photometric Lyman break object only 2.9~arcsec away from the QSO sight-line. If confirmed by future spectroscopy to be at the absorption redshift, then the galaxy associated with this DLA has a UV luminosity of about $0.7 L^{\\ast}$ and a linear extent of at least $15 h^{-1}\\,$kpc. Otherwise the absorber is likely to be fainter than about $0.4 L^{\\ast}$.  3. The H~I gas producing the damped system is highly turbulent, with a spin temperature $T_s \\simgt 4000$~K and complex absorption line profiles, consisting of many discrete components spanning $\\sim 270$~km~s$^{-1}$. Neither is what one may have expected from a cold, rotationally supported, disk.  4. The DLA has a metallicity $Z \\simeq 1/20 Z_{\\odot}$, at the upper end of the distribution of values of $Z_{\\rm DLA}$ at these redshifts, but lower than those deduced for other luminous Lyman break galaxies, although the data here are still very sparse. Further, it exhibits no marked overabundance of the $\\alpha$-elements (in this case S) relative to Fe, and a relatively high (N/$\\alpha$) ratio for its low metallicity. Both element ratios can be understood if there has been a relatively quiescent interval, lasting more than $\\sim 500$~Myr, since the last major burst of star formation in this galaxy, allowing the delayed release of Fe from Type~Ia SN and of N from AGB stars of intermediate mass. If this reasoning is correct, the star formation episode responsible for producing the heavy elements we see occurred at $z \\simgt 4.3$\\,. The chemical properties of this absorber are similar to those of another high-$z$ DLA, at $z_{\\rm abs}  = 3.3901$ in Q0000$-$2620, which has been extensively studied with both the Keck and VLT echelle spectrographs.  Firm conclusions on the nature of DLAs at high redshifts are hampered by the lack of follow-up spectroscopy of galaxy G2. If this is indeed the absorber, it would be only the second case where a high-$z$ DLA has been identified with a Lyman break galaxy, after the $z_{\\rm abs} = 3.151$ system in Q2233+1310 (Steidel, Hamilton, \\& Pettini 1995; Djorgowski et al. 1996), although others have been detected in \\lya\\ emission (e.g. Fynbo, Burud, \\& M{\\o}ller 2000 and references therein). In general our deep Lyman break imaging has shown that the galaxies producing DLAs at $z \\simgt 3$ must be significantly sub-$L^{\\ast}$ in their stellar continua (e.g. Steidel et al. 1995; Steidel et al. 1998) and that detections such as this one are the exception rather than the rule.  We conclude that, although the statistics are still very limited, DLAs seem to sample a wide range of the luminosity function of galaxies at $z \\simeq 3$. QSO absorption line spectroscopy still gives us the most precise measurements of many important physical properties of high redshift gas. However, only by combining it with deep imaging will it be possible to realise its full potential for unravelling the nature and evolution of galaxies at early times.\\\\  We are grateful to the staff of the Palomar and Keck Observatories for their competent technical assistance. Our collaborators in the LBG survey project, Kurt Adelberger, Mark Dickinson, Mauro Giavalisco, and Mindy Kellogg, generously helped with various aspects of the data acquisition and reduction. C.C.S. acknowledges support from the National Science Foundation through grant AST 95-96229 and from the David and Lucile Packard Foundation.  \\newpage"
    },
    "0010/astro-ph0010611_arXiv.txt": {
        "abstract": " ",
        "introduction": "The improvement by four orders of magnitude in wide-angle astrometric measurements in the last 30 years, provided mainly by Very Long Baseline Interferometry (VLBI), have resulted in a better definition of the fundamental astronomical reference frame (Johnston \\& de Vegt 1999), the International Celestial Reference Frame (ICRF), based on the radio positions of 212 compact extragalactic radio sources. Up to the present, the group delay VLBI observable has been regularly used for such a task.  The use of the more precise phase delay  observable should provide an immediate improvement in accuracy. A phase-connection process is needed (Shapiro \\et\\ 1979) to overcome the inherent $2\\pi$ ambiguity of the phases. When used with radio source pairs, the phase delay becomes the most accurate observable in astrometry. The phase-delay astrometric technique has been applied to different pairs of radio sources with separations from minutes to some degrees in the sky (among others, 3C\\,345/NRAO512, Shapiro \\et\\ (1979), Bartel \\et\\ (1986); 1038+528\\,A/B, Marcaide \\et\\ (1983, 1995) and Rioja \\et\\ (1997); 4C\\,39.25/0920+390, Guirado \\et\\ (1995b); QSO\\,1928+738/BL\\,2007+777, Guirado \\et\\ (1995a, 1998); 3C\\,395/3C\\,382, Lara \\et\\ (1996)).  A significant technical problem in the astrometric data reduction is the modelling and removal of the ionospheric effect.   Since the latter is a frequency-dependent effect, the removal was done in the past by observing at two frequencies. The Global Positioning System (GPS) allows now to obtain ionospheric data of unprecedented accuracy. Ros \\et\\ (2000a) and P\\'erez-Torres \\et\\ (2000) used GPS data to successfully remove the ionospheric effect.  Such an approach can be applied to any VLBI astrometric observations, without the need for dual-frequency observations.  Many compact radio sources show variable structures on scales larger than the accuracy of their position estimates. If the source structure is ignored, the group-delay astrometric results may incorrectly be interpreted as a proper motion of the radio sources. Because the radio sources, with typical redshifts of 1.0, are very distant, proper motions should not be detectable. It is true that most of the emission is generally produced in the core (brightest feature) of the source, but for low frequencies, due to opacity effects, this core may not be close to the central engine of the source (mass center of the system, which sould be stationary).  Until the efforts of Fey \\et\\ (1996) and Fey \\& Charlot (1997, 2000) the effect of source structure had been ignored in the group-delay based reference system maintenance.  These authors have initiated a programme to image the radio reference frame sources and define a source ``structure index\" to provide a quantitative estimate of the possible astrometric quality of them. In contrast, the phase-delay analysis from the early studies of Shapiro \\et\\ (1979) takes into account the effect of the structure and to date, the phase-delay astrometry is the only rigorous way to register radio maps with precisions of 100\\,$\\mu$as or better.  As mentioned above, this is the technique that we use in our astrometric data analysis. ",
        "conclusions": "Following recent progress in the phase-delay astrometric technique (extension to larger distances, larger sets of radio sources, and higher frequencies) we have initiated a program to study the 13 radio sources of the S5 polar cap sample.   We observed the sources twice at 8.4\\,GHz and twice at 15\\,GHz, and new observations are planned at these frequencies and at 43\\,GHz.  The astrometric data reduction includes the modelling of the atmospheric, ionospheric and source-structure effects. Our first results show the feasibility of phase-connecting the 13 radio sources at 8.4\\,GHz, and obtain precisions in the relative separation determinations better than 0.1\\,mas. The absolute kinematics of the radio sources will thus be determined unambiguously.  Furthermore, using the astrometric information available at 8.4, 15 and 43\\,GHz, it will be possible for the first time to make rigorous spectral-index maps, together with a multiwavelength study of the absolute kinematics of the S5 polar cap sources. This observing scheme could be extended in the future to further regions of the sky, on-route towards global phase-delay astrometry."
    },
    "0010/astro-ph0010177_arXiv.txt": {
        "abstract": "The nature of the binary systems giving rise to Type Ia supernovae (SNeIa) remains an unsolved problem. In this {\\it Letter} we calculate, from the statistics of initial conditions (masses and binary separations), the mass, luminosity, and velocity distributions of the possible binary companions (main--sequence star, subgiant, red giant) following the explosion of the white dwarf which gives rise to the SNeIa. Those companions could be detected from either their proper or their radial motions, by means of high--precision astrometric and radial--velocity measurements in young, nearby supernova remnants. Peculiar velocities typically ranging from 100 to 450 km s$^{-1}$ should be expected, which places proper--motion measurements within reach of HST instruments and makes radial--velocity ones feasible with 2.5--4m class telescopes from the ground. Detections would solve the long--standing problem of which kind of binaries do produce SNeIa and clear up the way to accurate physical modeling of the explosions. ",
        "introduction": "Type Ia supernovae (SNeIa) are unanimously attributed, nowadays, to the thermonuclear explosion of a white dwarf (WD) due to accretion of matter from a companion in a close binary system. The mass--accreting WD is also generally thought to be made of carbon and oxygen (CO WD). There the agreement stops, and different kinds of systems are proposed as candidates to SNeIa progenitors (see Iben 1997, for a review).  The nature of the WD companion, the mass--donor, is unclear: it could either be another WD (double--degenerate or DD systems) or a still thermonuclearly evolving star (single--degenerate or SD systems). The evolutionary stage of the companion in the SD systems could be anything from a main--sequence star to a supergiant,  and mass accretion proceed via Roche--lobe overflow (driven by thermonuclear evolution of the companion, magnetic braking, or gravitational wave radiation) or by capture of matter from stellar wind. All that also bears on the explosion mechanism: on how the explosion is triggered and how it develops. Those uncertainties still raise doubts as to the use of SNeIa as {\\it calibrated candles} to probe the dynamics of the Universe, the calibration being purely empirical and based on the nearby SNeIa sample (see Drell, Loredo, \\& Wasserman 2000, for instance).  Recently, evidence has been gathered against the merging of WDs giving rise to SNeIa, from the high efficiencies found in producing SNeIa out of star formation. These efficiencies have now been measured at several redshifts and appear to be similar, every $\\sim 700\\ M_{\\odot}$ going into star formation giving rise to 1 SNeIa (Ruiz--Lapuente \\& Canal 2000).  No previously observed system has ever been identified as the progenitor of any SNeIa. The historical SNeIa took place at a time when astronomical instrumentation was rudimentary. They can still give us, nonetheless, important and maybe decisive clues on the nature of the long--sought SNeIa progenitor systems (Ruiz--Lapuente 1997). If they were DD systems, there should be no companion left since it is destroyed in the accretion process already. On the contrary, in SD systems the companion can survive the explosion in most cases and show characteristics allowing its identification.  Of the historical SN, from the records on lights curves only two have been identified as SNeIa: SN 1006 and Tycho's SN (SN 1572). From their X--ray morphology (spherically--symmetric shell), X--ray spectra (showing high abundances of Fe, Ni, Si, Ca, Ar), Fe enrichment in their optical filaments, and lack of any central X--ray source, a number of nearby, young Galactic supernova remnants (SNR) can be attributed to SNeIa (Ruiz--Lapuente et al. 2000). As we will see, precise enough astrometric and radial--velocity measurements of the central regions of such SNR should allow detection of the SNeIa companions from either their large proper motions or high radial velocities, due to disruption of the binary orbit plus kick from the impact of the SN ejecta on them. For SNR ages $\\lapprox 10^{4}\\ yr$, the companions should still exhibit in their outer layers the consequences of the strong perturbation of thermal equilibrium from partial stripping of the envelopes and deposition of energy from the ejecta into the layers that remain bound, and they may also show chemical contamination from the SNeIa ejecta. ",
        "conclusions": "We have added to the results from existing hydrodynamical studies of the impact of SNeIa into their companion stars the statistical evaluation of the distribution of velocities to be expected for different kinds of companions. We have shown that they should be moving at velocities much higher than those of the surrounding stars: for RG companions there is a probability $P > 90\\%$ that the velocities are higher than $100\\ km\\ s^{-1}$; for SG $P > 99\\%$ that $v > 200\\ km\\ s^{-1}$, and for MS $P > 99\\%$ that $v > 450\\ km\\ s^{-1}$. Such velocities make their identification possible in the central regions of young, nearby SNR of SNeIa origin. For distances $d\\lapprox 5 - 8\\ kpc$ and SNR ages $\\lapprox 10^{3} - 10^{4}\\ yr$, detection of proper motion is within the reach of instruments on board the HST, and that of radial motions can be made with 2.5--4m class telescopes from the ground. The angular sizes of the regions to be surveyed are small enough that the numbers of stars within them, bright enough to be candidates to SNeIa companions, remains low. There are 8 Galactic SNR (including those of SN 1006 and SN 1572) that most likely formed from SNeIa explosions and whose distances and ages allow thorough searches (Ruiz--Lapuente et al. 2000).  Candidates selected from their peculiar velocities and positions close to the centers of the SNR could be confirmed by spectroscopic characteristics arising from their being out of thermal equilibrium due to heating by the impact of the SN debris, according to recent hydrodynamic simulations (Marietta, Burrows, \\& Fryxell 2000). Contamination by material accreted from the low--velocity tail of the ejecta is also possible.  Our conclusions are largely independent from the unknowns still affecting close binary evolution, mass transfer, and growth of WDs up to the Ch--mass.  Therefore, it now appears feasible to test the SD system hypothesis for the origin of SNeIa by means of observations of Galactic SNR. Detection of confirmed companions would not only validate the SD system hypothesis, but it would also allow physical modeling of SNeIa on a firmer basis and, through that, make much more reliable than at present the use of SNeIa in Cosmology."
    },
    "0010/astro-ph0010494_arXiv.txt": {
        "abstract": "The giant flare of 1998 August 27 from SGR 1900+14 was extraordinary in many ways: it was the most intense flux of gamma rays ever detected from a source outside our solar system; it was longer than any previously detected burst from a soft gamma repeater (SGR) in our Galaxy by more than an order of magnitude; and it showed a remarkable four-peaked, periodic pattern in hard X-rays with the same rotation period that was found modulating soft X-rays from the star in quiescence. The event was detected by several gamma-ray experiments in space, including the \\it Ulysses \\rm  gamma-ray burst detector and the BeppoSAX Gamma Ray Burst Monitor. These instruments operate in different energy ranges, and comparisons of their measurements reveal complex patterns of spectral evolution as the intensity varies.  In this paper, we present a joint analysis of the BeppoSAX and \\it Ulysses \\rm  data and discuss some implications of these results for the SGRs.  We also present newly-analyzed Venera/SIGNE and ISEE-3 data on the 1979 March 5 giant flare from an SGR in the Large Magellanic Cloud (SGR 0526-66), and compare them with the August 27 event. Our results are consistent with the hypothesis that giant flares are due to catastrophic magnetic instabilities in highly magnetized neutron stars, or ``magnetars\".  In particular, observations indicate that the initial hard spike involved a relativistic outflow of pairs and hard gamma rays, plausibly triggered by a large propagating fracture in the crust of a neutron star with a field exceeding 10$^{14}$ Gauss. Later stages in the light curve are accurately fit by a model for emission from the envelope of a magnetically-confined pair-photon fireball, anchored to the surface of the rotating star, which contracts as it emits X-rays and then evaporates completely in a finite time.  The complex four-peaked shape of the light curve likely provides the most direct evidence known for a multipolar geometry in the magnetic field of a neutron star. ",
        "introduction": "During recent years, soft X-ray observations of the quiescent counterparts to SGR 1806-20 and SGR 1900+14 have revealed periodicities in the 5--8~s range and spindown rates of $\\sim 10^{-11}$--$10^{-10}$ s s$^{-1}$ (Kouveliotou et al. 1998a, Hurley et al. 1999c). These  rotation periods are similar to that of SGR 0526-66, which displayed an 8~s periodicity during the giant flare of 1979 March 5 (Mazets et al.~1979b, Barat et al.~1979). Precise localizations of SGRs indicate that they may be associated with supernova remnants (SNRs) of ages $\\loe 10^4$ years. SGR 1806--20 was localized by detecting its bursts in X--rays (Murakami et al.~1994) to a  position consistent with a radio synchrotron nebula that may be a plerionic SNR (Kulkarni \\& Frail 1993); however a recent more accurate localization of the bursting source provides evidence that the SGR is displaced from the radio nebula core (Hurley et al. 1999d). SGR 1627-41, discovered in 1998 (\\cite{kouveliotou98c,feroci98}), is also positionally coincident with a young SNR (Hurley et al.~1999e, Woods et al.~1999a, Smith, Bradt \\& Levine 1999). However, deep X--ray observations of SGR 1627-41 (Hurley et al. 2000) have so far failed to verify any X--ray periodicity in the quiescent source.  The association of SGRs with young SNRs suggests that they are young neutron stars. The long rotation periods and rapid spindown rates, in the absence of any evidence for binary companions, can be accounted for if the spindown is driven by an ultra-strong magnetic field, e.g., $B_{dipole}\\goe10^{14}$~Gauss, as invoked in the magnetar model (Duncan \\& Thompson 1992, ``DT92\"; Paczy\\'nski 1992; Thompson \\& Duncan 1995, ``TD95\").   Magnetars, or ``magnetically-powered neutron stars,\"  could form via an $\\alpha$-$\\Omega$ dynamo action in hot, nascent neutron stars if they are born spinning rapidly enough (DT92; Thompson \\& Duncan 1993, ``TD93\").  Magnetism may be strong enough within these stars to evolve diffusively over SGR lifetimes of $\\sim 10^4$ years, driving internal heat dissipation that would keep the neutron stars hot and X-ray bright (Thompson \\& Duncan 1996; Heyl \\& Kulkarni 1998). Above a flux density of $\\sim 10^{14}$ G, the evolving field inevitably induces stresses in the solid crust that cause it to yield or fracture (TD95). Such magnetically-driven starquakes may account for the ``ordinary\" SGR bursts, which have many statistical properties in common with earthquakes (Cheng et al.~1996; Palmer 1999; Gogus et al.~1999;  Gogus et al.~2000). Less common and more catastrophic magnetic instabilities, perhaps involving large propagating fractures, are thought to produce giant flares (TD95).  The non-thermal persistent emission of the SGRs has also been proposed to be a consequence of magnetic activity:  either through persistent fracturing of the crust driven by the Hall electric field (Thompson \\& Duncan 1996) or through persistent magnetospheric currents that are excited by twisting motions of the crust during outbursts (Thompson et al.~1999).  We note that the observed spindown histories of SGRs are {\\it not} well-fit by the idealization of vacuum magnetic dipole radiation [MDR] (Kouveliotou et al.~1998a, 1999; Marsden et al.~1999; Woods et al.~1999b, 2000). Indeed, spindown torques in active magnetars can be enhanced by persistent outflows of relativistic particles and Alfv\\'en waves, channeled by a strong magnetic field (Thompson \\& Blaes 1998; Harding, Contopoulos \\& Kazanas 1999; Thompson et al.~1999) and also strongly modulated by material ejected during bursting activity (Woods et al.~2000). It remains possible that the smooth spindown in some {\\it inactive} candidate magnetars (such as 1E 1841-045) is well fit by MDR (Vasisht \\& Gotthelf 1997; Gotthelf, Vasisht \\& Dotani 1999). The observed spindown in SGRs is consistent with dipole fields weaker than the QED strength ($4.4\\times 10^{13}$ G), but only if a persistent outflow of particles carries away substantially more energy than is observed in X-rays  (Harding, Contopoulos \\& Kazanas 1999; Marsden et al.~1999).  Accretion-powered alternatives to the magnetar model have also been considered (e.g., Van Paradijs, Taam \\& Van den Heuvel 1995;  Li 1999; Chatterjee, Hernquist \\& Narayan 2000; Alpar 2000).  Such models make reasonable fits to the continuous X-ray emissions and spindown histories of SGRs and Anomalous X-ray Pulsars (AXPs), but they offer no good explanation for the hyper-Eddington burst and flare emissions that are the defining property of SGRs (cf.~\\S 7.3 in TD95; \\S 5.2 in Thompson et al.~1999).  This paper will focus on SGR 1900+14.  Of the four known SGRs, this object is the only one without an identified  SNR surrounding it. However, the recently-verified source position (Vasisht et al.~1994; Hurley et al.~1999a; Murakami et al.~1999; Frail, Kulkarni \\& Bloom 1999) lies just outside the edge of G42.8+0.6, a $\\sim 10^{4}$-year-old galactic SNR. \\ A parallel can be drawn with the other giant flare source, SGR 0526-66, which lies near the edge of, but just {\\it inside}, the young SNR N49 in the LMC (Cline et al.~1982). Note that several mechanisms could plausibly impart a recoil velocity $\\sim 1000$ km s$^{-1}$ to a magnetar at birth, sufficient to propel it outside its remnant in $\\sim 10^4$ years (DT92).  SGR 1900+14 was first detected in 1979, when it emitted three bursts of soft gamma-rays of moderate intensity (Mazets et al.~1979a). The source again became burst active for a short while in 1992 when a handful of events were recorded (Kouveliotou 1993).  In 1998 May, SGR 1900+14 entered an unprecedented level of burst activity in both frequency and intensity (Kouveliotou et al.~1998b, Hurley et al.~1999a). Purely by chance, an ASCA (2--10 keV) observation less than one month prior to the source reactivation in 1998 led to the discovery of 5.16 pulsations from the quiescent counterpart to the SGR (Hurley et al. 1999c). Subsequent RXTE PCA (2--20 keV) observations following the reactivation of the SGR allowed for a measurement of the period derivative $\\sim 1 \\times 10^{-10} \\rm s \\, s^{-1}$ (Kouveliotou et al.  1999; Woods et al. 1999b). This spindown rate is consistent with a magnetar-strength field (Kouvelioutou et al.~1999; Thompson et al.~1999 and references therein).  On 1998 August 27 a giant flare from SGR1900+14, lasting more than five minutes, was detected by Konus-Wind, \\it Ulysses \\rm  and BeppoSAX (Cline et al. 1998; Hurley et al. 1999b; Feroci et al. 1999; Mazets et al.~1999). Gamma rays during the first second were extraordinarily intense, overwhelming detectors on several other spacecrafts as well, such as the Near Earth Astreroid Rendevous mission (which went into a protective shut-down mode) and the Proportional Counter Array on the Rossi X-ray Timing Explorer. The Compton Gamma-Ray Observatory was Earth-occulted for this flare.  The 5.16~s neutron star rotation period was clearly detected during the giant flare.  Indeed, the periodic signal was intense enough to produce a marked 5.16-second modulation in the height of the Earth's ionosphere, which affected long-wavelength radio transmissions (Inan et al.~1999), a remarkable effect for a star $\\sim 20,000$ light years away. About 40 seconds after the onset of the flare,a 1.03 s repetitive pattern (4th harmonic of 5.16 s fundamental) was observed to set in gradually (Feroci et al.~1999; Mazets et al.~1999), unlike any emission previously detected from any source. A radio afterglow was soon found with the Very Large Array (Frail, Kulkarni \\& Bloom 1999).  This source, the only radio point source unambiguously associated with an SGR, was apparent in the error box of SGR1900+14 on 1998 September 3, but it faded away in less than one week, giving evidence for an abrupt outflow of relativistic particles during the flare.  In this paper we carry out a comparative analysis of BeppoSAX and {\\it Ulysses} observations of the August 27th event.  Our main goal is a thorough description of the data that may be cogent to understanding SGR giant flares, including some novel comparisons with the 1979 March 5th event. The observed properties are then qualitatively interpreted in the context of the magnetar hypothesis, deriving insights and constraints on the magnetic field structure and mechanisms for large flares in SGRs. A companion paper (Thompson et al.~2000, hereafter Paper II) derives a more quantitative and rigorous treatment of the physics and magnetic field structure in this and similar objects. ",
        "conclusions": "In this paper we present an analysis of the data on the giant flare from SGR 1900+14 available from the BeppoSAX Gamma Ray Burst Monitor and from the \\it Ulysses \\rm Gamma Ray Burst detector. We observe three clearly distinct stages in the August 27 event. During the first 0.5 seconds the star emitted {\\it an intense spike of hard gamma-rays} which saturated all detectors. The  fluence in this hard spike was at least half the total photon fluence of the event at energies above 15~keV (Mazets et al. 1999). Its duration and fast ($\\lte 4$ ms) rise time are consistent with a magnetically-driven instability in an ultramagnetized ($B > 10^{14}$ Gauss)  neutron star, involving a large propagating fracture in the crust. After the hard spike, the flare's light curve shows two distinct phases: a {\\it smooth, soft tail} that lasted $\\sim 40$ s, followed by the emergence of a striking pattern of {\\it four large-amplitude peaks} that repeat coherently on the 5.16-s rotation period of the neutron star.  During both these phases, the rotation-averaged emissions declined nearly monotonically, and released a total energy not much larger than the initial spike (cf. Mazets et al.~1999).  Taken together, these observations suggest that most of the burst energy was released during this initial half-second. Half or more of the energy escaped promptly in a {\\it relativistic} outflow of electron-positron pairs and hard gamma rays, and the remainder was trapped in the star or its vicinity (TD95). Independent evidence for a particle outflow comes from the detection of a radio afterglow which faded in the days following the event (Frail, Kulkarni \\& Bloom 1999).  The rapid drop in X-ray flux at the end of the burst gives strong evidence for an emitting resevoir of energy that evaporates completely in a finite time.   The envelope of the light curve, from 40 s to the termination of the burst, is well fit by the contracting surface of a thermal photon-pair plasma, confined by the strong magnetic field and anchored to the rotating surface of the star (eqn.~(1); cf.~TD95). About 80\\%  of the post-spike fluence is associated with this phase of large amplitude pulsations.  The smooth, soft tail has a harder spectrum than the subsequent emissions, and the measured X-ray flux lies well above the trapped fireball light curve that fits so well after $\\sim$ 40 s.  The flattening of the light curve, and the decay of this excess emission, occurs simultaneously with the emergence of the four large-amplitude sub-peaks. Given these facts, we associate the excess emission (representing $\\sim 20$\\% of the post-spike fluence) with Compton heating by an extended corona of hot pair plasma (Paper II). This heating must occur outside the trapped fireball, and is plausibly driven by a persistent output of Alfv\\'en waves from the neutron star. In this model, the appearance of the large-amplitude pulsations coincides with the evaporation of the pair plasma and a contraction of the scattering photosphere to the surface of the trapped fireball, which sits congruent with the confining magnetic field.   The persistent Alfv\\'enic heating could be driven by crustal shear waves, excited at the initial instability (Blaes et al 1989; Duncan 1998); or by persistent but declining seismic activity in an active fault zone.  The large-amplitude pulsations require narrow collimation of the X-ray flux, which is provided by the rapid increase in the E-mode scattering cross section with distance from the neutron star. In addition, the O-mode  has a much stronger coupling to matter than the E-mode and will flow hydrodynamically, at mildly relativistic speed, along open channels of magnetic flux.  Both modes are emitted through X-ray fan jets, which are fed from below by rapid radiative diffusion of E-mode photons out of the trapped reservoir of hot plasma (followed by conversion to the O-mode through scattering or splitting).  Collimated X-ray emission is a natural consequence of super-Eddington radiative transfer in magnetic fields stronger than $B_{QED}= 4.4 \\times 10^{13}$ Gauss (\\S 6.4 in TD95; \\S \\ref{fourpeaksec}; Paper II).  The light curve of the August 27 flare provides valuable information about the magnetic geometry in SGR 1900+14.  The four sub-pulses require the presence of higher magnetic multipoles -- perhaps the most tangible evidence yet for a complex magnetic geometry in an isolated neutron star.  In this picture, the observation of two (rather than four) sub-pulses in the March 5 event simply reflects the chance location of the outburst, and the differing structure of the magnetic field in that second source.  The joint analysis of the \\it Ulysses \\rm and GRBM spectra shows that a hard spectral component persisted long after the initial hard spike.  This conclusion is corroborated by the time-resolved hardness ratio, which shows that the burst spectrum is hardest during rotation-cycle minima. We have noted two possible mechanisms for accelerating non-thermal particles: persistent Alfv\\'enic heating, for which there is a clear evidence at times $\\loe$40~s; or magnetohydrodynamic turbulence at the boundaries of the X-ray jets responsible for the four-peaked repetitive pattern.  Finally, we re-analyzed the SIGNE data of the first 32~s of the 1979 March 5th event from SGR 0526-66, and ISEE-3 data for the first $\\sim$150~s of the same event. We found more properties that are shared with the August 27 event, lending support to the hypothesis that similar physical instabilities drove both outbursts. In particular, we find that the spectral evolution across the pulse is strikingly similar to that of the giant flare from SGR 1900+14.  A trapped fireball model appears to fit the March 5 light curve better than a simple exponential, just as in the case of the August 27 event, although the short duration of the available light curve and the small dynamic range ($\\sim 30$) in flux do not allow a definitive conclusion. We also considered the suggestion of jetlike substructure on 1-s timescales in the SIGNE March 5th light curve (Barat et al.~1979). By a standard epoch-folding approach, we found no statistically-compelling evidence that such substructure emerged during the first 32 seconds of the event.  We conclude that the magnetar model offers a promising framework for understanding giant flares from SGRs.  The basic picture of an initial relativistic outflow followed by diffusive cooling of a trapped, thermal fireball seems robust.  But a detailed understanding of the radiative transport will involve mastering several complications, including: the effects of bulk hydrodynamical streaming, the large variations of the radiative transport coefficients arising from inhomogeneities in the background magnetic field, a non-dipolar magnetic geometry, and the behavior of a pair atmosphere that is simultaneously heated by Alfv\\'en waves and irradiated with X-rays from a trapped fireball.  Much work will be needed to test the physical mechanisms suggested here and in Paper II, and to make a quantitative comparison with alternative models."
    },
    "0010/astro-ph0010388_arXiv.txt": {
        "abstract": "We discuss the main ingredients necessary to build models of chemical evolution of spiral galaxies and in particular the Milky Way galaxy. These ingredients include: the star formation rate, the initial mass function, the stellar yields and the gas flows.  Then we discuss models for the chemical evolution of galaxy disks and compare their predictions with the main observational constraints available for the Milky Way and other spirals.  We conclude that it is very likely that the disk of our Galaxy and other spirals formed through an ``inside-out'' mechanism, where the central parts collapsed much faster than the external ones. This mechanism has important consequences for the appearance of galaxy disks as a function of redshift. ",
        "introduction": "In order to predict the chemical evolution of galaxy disks one needs to know several fundamental quantities. In particular, to build a model which predicts the evolution of the abundances of the most common chemical elements in the gas as functions of space and time, some assumptions on the following physical processes are required: i) the star formation, ii) the stellar nucleosynthesis, iii) the possible gas flows entering and/or leaving the system. Unfortunately, point i) is still very uncertain since we do not know the physics of star formation and we have to adopt simplifying assumptions about the star formation law.  Generally we divide the stellar birthrate into two independent functions, one is the so-called star formation rate (SFR) and is only a function of time, the other is the initial mass function (IMF) and is only a function of the stellar mass. Concerning point ii), the knowledge of stellar evolution and nucleosynthesis is more advanced than that of the star formation process although large uncertainties still exist also in the calculation of the stellar yields; among these there are the treatment of convection, the rates of nuclear reactions and the explosive nucleosythesis. In particular, the yield of iron from massive stars ($M > 10 M_{\\odot}$) seems to be particularly uncertain owing to the poorly known ``mass-cut'', namely how much mass stays in the collapsed remnant and how much is ejected in the supernova event. Finally, point iii) is also highly uncertain since the gas dynamics is a very complex phenomenon and pure chemical evolution models take into account dynamics only in a very simplified manner.  In past years a great deal of theoretical work has appeared concerning the chemical evolution of the disk of the Milky Way and other spirals (Matteucci and Fran\\c cois 1989; Yoshii and Sommer-Larsen, 1990; Burkert et al. 1992; Carigi 1994; Timmes et al. 1995; Tsujimoto et al. 1995; Chiappini et al. 1997, 1999,2000a,b; Chang et al. 1999; Portinari \\& Chiosi 1999, 2000, Boissier \\& Prantzos 1999; Goswami \\& Prantzos 2000;Prantzos and Boissier, 2000; Romano et al. 2000).  Most of the more recent models agree that is important to assume an ``inside-out'' formation for disks. This implies that the inner parts of galaxy disks should form on relatively fast timescales whereas the outermost regions might be still forming now.  In this paper we focuse on one particular model for the chemical evolution of the Galactic disk, the model of Chiappini et al. (1997, hereafter CMG97) and its more recent development (Chiappini et al. 2000b).  This model assumes two main episodes of gas infall of primordial chemical composition. The first episode occurs on a short timescale ($\\sim 0.8 $ Gyr) and contributes to the formation of the halo, the second takes much longer (several Gyrs) and gives rise to the disk.  In this picture the formation of most of the disk is disentangled from that of the halo which is assumed to have lost part of its gas to form the bulge.  Therefore, the bulge formed on a timescale similar to that of the halo.  The formation of the innermost disk regions also occurs on a short timescale whereas the outermost regions form on much longer timescales.  >From the observational point of view, the distributions of abundances, stars and gas in the disk of our Galaxy are the best studied.  In particular, the presence of abundance gradients along the Galactic disk is now well established. Abundance gradients represent an important constraint on the history of the formation of the disk especially if considered together with the distribution of gas, stars and star formation as functions of the Galactocentric distance.  Abundance gradients are also measured in other spirals and in some cases the gas distribution is also available.  Finally, in the local disk (solar neighbourhood) the abundances in stars of different ages are very well studied.  Very good constraints are represented by the G-dwarf metallicity distribution which imposes constraints on the birthrate history of the local disk and the [el/Fe] versus [Fe/H] relations which help us in understanding the stellar nucleosynthesis and supernova (SN) progenitors. ",
        "conclusions": "The comparison between observations and models suggests that:  \\begin{itemize}  \\item The disk of the Galaxy formed mostly by infall of primordial or very metal poor gas accumulating faster in the inner than in the outer regions (inside-out scenario).  \\item In the framework of the inside-out scenario the SFR should be a strongly varying function of the galactocentric distance.  This can be achieved by assuming a dependence of the SFR either on the total surface mass density (feed-back mechanism) or on the angular circular velocity of gas (both are supported by observations), besides the dependence upon the surface gas density.  \\item Radial flows probably are not the main cause of gradients but can help in reproducing the gas profile.  \\item An IMF constant in space and time should be preferred, as shown by several numerical experiments (e.g. Chiappini et al. 2000a).  \\item The disks of other spirals also indicate an inside-out formation, as shown by Prantzos and Boissier (2000) who described the galaxy disks by means of scaling laws, taken from semy-analytical models of galaxy formation (Mo et al. 1998), calibrated on the Galaxy.  \\item Dynamical processes such as the formation of a central bar can influence the evolution of disks and deserve more attention in the future.  \\item Abundance ratios (e.g. [$\\alpha$/Fe]) in stars at large galactocentric distances can provide clues for understanding the formation of the disk and the halo (inside-out or outside-in) (see Chiappini et al., 2000b).  \\item The predictions of chemical evolution models can be tested in a cosmological context by studying galaxy surface brightness and size evolution as a function of redshift. Roche et al. (1998) concluded that a size-luminosity evolution, as predicted by the inside-out model of CMG97, represents a good fit to the observations. \\end{itemize}"
    },
    "0010/astro-ph0010341_arXiv.txt": {
        "abstract": "The Solar Tower Atmospheric Cherenkov Effect Experiment (STACEE) is located at the National Solar Thermal Test Facility of Sandia National Laboratories in Albuquerque, New Mexico, USA. The field of solar tracking mirrors (heliostats) around a central receiver tower is used to direct Cherenkov light from atmospheric showers onto secondary mirrors on the tower, which in turn image the light onto cameras of photomultiplier tubes. The STACEE Collaboration has previously reported a detection of the Crab Nebula with approximately 7 standard deviation significance, using 32 heliostats (STACEE-32).  This result demonstrates both the viability of the technique and the suitability of the site.  We are in the process of completing an upgrade to 48 heliostats (STACEE-48) en route to an eventual configuration using 64 heliostats (STACEE-64) in early 2001.  In this paper, we summarize the results obtained on the sensitivity of STACEE-32 and our expectations for STACEE-48 and STACEE-64. ",
        "introduction": " ",
        "conclusions": ""
    },
    "0010/astro-ph0010207_arXiv.txt": {
        "abstract": "We will review our mid-infrared and HST imaging surveys of the circumstellar dust shells of proto-planetary nebulae. While optical imaging indirectly probes the dust distribution via dust-scattered starlight, mid-IR imaging directly maps the distribution of warm dust grains. Both imaging surveys revealed preferencially axisymmetric nature of PPN dust shells, suggesting that axisymmetry in planetary nebulae sets in by the end of the asymptotic giant branch phase, most likely by axisymmetric superwind mass loss. Moreover, both surveys yielded two morphological classes which have one-to-one correspondence between the two surveys, indicating that the optical depth of circumstellar dust shells plays an equally important role as the inclination angle in determining the morphology of the PPN shells. ",
        "introduction": "While circumstellar dust shells (CDSs) of asymptotic giant branch (AGB) stars display high spherical symmetry (e.g. Habing \\& Blommaert 1993), planetary nebulae (PNs) are renowned for their spectacular axisymmetry (e.g. Zuckerman \\& Aller 1986). To understand {\\sl when} and {\\sl how} such a drastic morphological conversion takes place, the proto-planetary nebula (PPN) phase, a brief transitionary phase between the AGB and PN phases (e.g. Kwok 1993) has been extensively investigated. In fact, PPNs are the most ideal space laboratories for such studies: AGB mass loss histories are imprinted on the PPN shells in the most pristine form because the PPN central stars are not hot enough to generate disturbing fast winds and/or ionizing UV photons as PN central stars.  Therefore, by conducting imaging surveys on a large number of PPN CDSs, we can pinpoint the epoch of morphological transformation and obtain insights about yet unknown mechanisms for the AGB mass loss. To achieve our goals, we employed mid-infrared (IR) and optical wavelength ranges, both of which have strengths and weaknesses that would complement one another. Diffraction-limited mid-IR images give marginal resolution ($1^{\\prime\\prime}$) but allow one to directly probe dust mass distribution through thermal dust emission. On the other hand, optical images are of high resolution ($0^{\\prime\\prime}.1$), while they only allow an indirect mass probing through dust-scattered starlight escaping from the dust shells. ",
        "conclusions": ""
    },
    "0010/astro-ph0010295_arXiv.txt": {
        "abstract": "We motivate the study of chiral cosmic strings through a scenario of structure formation which mixes D-term inflation and strings. We then discuss some properties of chiral cosmic strings, and results regarding their evolution and possible cosmological consequences are presented. ",
        "introduction": " ",
        "conclusions": ""
    },
    "0010/astro-ph0010589_arXiv.txt": {
        "abstract": "Spectra (1951-78) of the central object in $\\eta$ Car, taken by A.D. Thackeray, reveal three previously unrecorded epochs of low excitation. Since 1948, at least, these states have occurred regularly in the 2020 day cycle proposed by Damineli et al. They last about 10 percent of each cycle. Early slit spectra (1899-1919) suggest that at that time the object was always in a low state. $JHKL$ photometry is reported for the period 1994-2000. This shows that the secular increase in brightness found in 1972-94 has continued and its rate has increased at the shorter wavelengths. Modulation of the infrared brightness in a period near 2020 days continues. There is a dip in the $JHKL$ light curves near 1998.0, coincident with a dip in the X-ray light curve. Evidence is given that this dip in the infrared repeats in the 2020 day cycle. As suggested by Whitelock \\& Laney, the dip is best interpreted as an eclipse phenomenon in an interacting binary system; the object eclipsed being a bright region (`hot spot'), possibly on a circumstellar disc or produced by interacting stellar winds. The eclipse coincides in phase and duration with the state of low excitation.  It is presumably caused by a plasma column and/or by one of the stars in the system. ",
        "introduction": "Despite a very extensive literature there is as yet no general consensus on the nature of $\\eta$ Carinae and the cause of the Great Eruption (c. 1838 - 1858) (see Morse et al. 1999 and Davidson \\& Humphreys 1997 for recent summaries). The present paper is the third in a series reporting observations of this object. The earlier papers are Whitelock et al. 1983 (henceforth Paper~I) and Whitelock et al. 1994 (henceforth Paper~II). Paper~I reported and discussed infrared, $JHKL$, photometry of the object in the years 1972 to 1982 as well as optical and infrared spectroscopy. Paper~II continued the photometric series up to 1994 and showed that, besides a gradual increase in brightness over the entire time interval 1972 to 1994, there was evidence for variations on a time scale of about 5 years. This was probably the first convincing evidence for periodicity or quasi-periodicity in the system. Damineli (1996) combined these data with observations of the variations in the He\\,{\\sc i} 10830A emission line and other spectroscopic results in the literature to suggest that there was a rather stable periodicity of 5.52 years. Later Damineli et al. (1997) found evidence for a variation in the radial velocity of Pa$\\gamma$ emission in this same period. They interpreted this as showing that $\\eta$ Car was a spectroscopic binary in a highly eccentric orbit and they suggested that various phenomena were associated with periastron passage (see also Damineli et al. 2000). However, Davidson et al. (2000) report HST observations of the central object, which is not resolved from the ground. These do not confirm the velocity behaviour predicted by the Damineli et al. orbit. In the present paper we report an examination of spectroscopic material, particularly that collected by A.D. Thackeray over the years 1951 to 1978, and also give the results of the continued monitoring of $\\eta$ Car in $JHKL$. We discuss these data, as well as published material, with particular reference to the light they throw on the periodicity and nature of the object. ",
        "conclusions": "The photometric and spectroscopic results discussed in this paper seem most naturally explicable by an interacting-binary model of some kind. A binary model involving a `hot spot' is basically that proposed originally by Bath (1979); though the long period presumably indicates that in the case of $\\eta$ Car, the interaction involves a stellar wind from at least one of the stars rather than Roche-lobe overflow. It is evident that a full quantitative model of the $\\eta$ Car system is still a matter for the future. However the results of the present paper provide considerable constraints for any such model.  \\subsection*{Acknowledgements} We are grateful to Dave Laney, Tom Lloyd Evans, Enrico Olivier and Ronald Heijmans for making some of the $JHKL$ observations.  The re-examination of the archival spectra arose out of discussions with Roberta Humphreys. We thank the referee (Kris Davidson) for some pertinent comments. This paper is based partly on observations made from the South African Astronomical Observatory."
    },
    "0010/astro-ph0010540_arXiv.txt": {
        "abstract": "We have investigated the infrared properties of  a sample of objects with \\iras\\ colors typical of planetary nebulae. The selected objects have not yet evolved to the planetary nebula stage since they have no detectable radio continuum emission and they are therefore likely post-AGB stars. ",
        "introduction": "A most intriguing challenge is to understand how Asymptotic Giant Branch (AGB) stars transform their surrounding mass-loss shells in a couple of thousand years into the variety of shapes and sizes observed in Planetary Nebulae (PNe).  There are a number of theories currently being investigated.  No consensus about the dominant physical process responsible for the shaping of PNe has emerged so far, but there is agreement that the shaping occurs during the early AGB-to-PN transition stage.  The details of the rate of evolution and the strength of the stellar wind during the AGB-to-PN transition phase are essential ingredients in understanding the shaping of the PN. However, both are very poorly known, either theoretically or observationally.  To remedy this we have started to study a sample of \\iras-selected post-AGB candidates. ",
        "conclusions": ""
    },
    "0010/astro-ph0010406_arXiv.txt": {
        "abstract": "The KPNO International Spectroscopic Survey (KISS) is a new objective-prism survey for extragalactic emission-line objects.  It combines many of the features of previous slitless spectroscopic surveys with the advantages of modern CCD detectors, and is the first purely digital objective-prism survey for emission-line galaxies.  Here we present the first list of emission-line galaxy candidates selected from our red spectral data, which cover the spectral range 6400 to 7200 \\AA.  In most cases, the detected emission line is H$\\alpha$.  The current survey list covers a one-degree-wide strip located at $\\delta$(1950) = 29$\\arcdeg$~30$\\arcmin$ and spanning the RA range 12$^h$~15$^m$ to 17$^h$~0$^m$.  An area of 62.2 deg$^2$ is covered.  A total of 1128 candidate emission-line objects have been selected for inclusion in the survey list (18.1 per deg$^2$).  We tabulate accurate coordinates and photometry for each source, as well as estimates of the redshift and emission-line flux and equivalent width based on measurements of the digital objective-prism spectra.  The properties of the KISS emission-line galaxies are examined using the available observational data. ",
        "introduction": "The KPNO International Spectroscopic Survey (KISS) is an ongoing objective-prism survey which targets the detection of large numbers of extragalactic emission-line sources.  The survey method is patterned after previous surveys for these types of objects carried out with Schmidt telescopes and photographic plates (e.g., Markarian 1967, Smith \\etal 1976, MacAlpine \\etal 1977, Pesch \\& Sanduleak 1983, Wasilewski 1983, Markarian \\etal 1983, Zamorano \\etal 1994, Popescu \\etal 1996, Surace \\& Comte 1998). The primary characteristic that distinguishes KISS from these previous surveys is that we utilize a CCD as our detector.  With the advent of large format CCDs, the areal coverage possible with the combination of Schmidt telescopes and CCDs makes digital surveys like KISS both possible and highly desirable.  The obvious advantages of CCDs over photographic plates include much higher quantum efficiency, lower noise, good spectral response over the entire optical portion of the spectrum, and large dynamic range.  In addition, CCDs give us the ability to use automated selection methods to detect emission-line galaxies (ELGs), and allow us to quantify the selection function and completeness limit directly from the survey data.  The combination of increased depth and large areal coverage is a powerful one, and allows for substantial improvements compared to the previous photographic surveys listed above.  The primary goal of KISS is to produce a high-quality survey whose selection function and completeness limits can be accurately quantified so that the resulting catalog of ELGs will be useful for a broad range of studies requiring statistically complete galaxy samples.  We also want to reach substantially deeper than previous objective-prism surveys.  A complete description of the survey method employed for KISS is given in the first paper in this series (Salzer \\etal 2000, hereafter Paper I). As described there, the first survey strip was observed in two distinct spectral regions.  One covered the blue portion of the spectrum (4800 -- 5500 \\AA), while the second covered the wavelength range 6400 -- 7200 \\AA\\ in the red part of the spectrum.  The first blue survey list is given in Salzer \\etal (2001, hereafter KB1).  The present paper presents the initial KISS list of H$\\alpha$-selected ELG candidates.  In addition to listing the ELGs, we provide substantial observational data for each object.  This includes accurate photometry and astrometry for each source, as well as estimates of each galaxy's redshift, line flux, and equivalent width.  These data are used to examine the properties of the KISS ELGs in Section 4. ",
        "conclusions": "We present the first list of emission-line galaxy candidates from the KPNO International Spectroscopic Survey.  KISS is an objective-prism survey, similar in nature to a number of important surveys carried out in the past using Schmidt telescopes and photographic plates.  The crucial improvement incorporated into KISS is the use of a CCD as the detector.  This allows us to detect fainter ELGs than previous surveys.  Our specially chosen filters also allow us to detect emission lines out to higher redshifts than did most previous line-selected surveys.  The combination of higher sensitivity, lower noise, and larger volumes surveyed yield huge improvements in the depth of the resulting survey. KISS finds 181 times more AGN and starburst galaxy candidates per unit area than did the Markarian (1967) survey, and 32 times more than the UCM survey (Zamorano \\etal 1994).  This first installment of KISS includes 1128 ELG candidates selected from 54 red survey fields covering a total of 62.2 deg$^2$.  The KISS catalog has a surface density of over 18 galaxies per deg$^2$.  The primary emission line we are sensitive to is H$\\alpha$.  The survey follows a narrow strip across the sky at a declination of $\\delta$(1950) = 29$\\arcdeg$~30$\\arcmin$ and spanning the RA range 12$^h$~15$^m$ to 17$^h$~0$^m$.  This region was chosen to overlap the Century Redshift survey (Geller \\etal 1997).  For each object in the catalog we tabulate accurate equatorial coordinates, B \\& V photometry, and estimates of the redshift and line strength measured from the objective-prism spectra. Also provided are finder charts and extracted spectral plots for all galaxies.  In addition to the main survey list, we include a supplementary list of 189 ELG candidates with weaker (lower significance) emission lines.  Since the survey data themselves provide such a large amount of observational data for each KISS ELG, we are able to develop a fairly complete picture of the survey constituents even before follow-up spectral information is available.  The median apparent magnitude of the sample is B~=~18.08, which is substantially fainter than previous ELG surveys.  Objects fainter than B~=~20 are routinely cataloged.  Measurement of the line strengths in the objective-prism spectra show that KISS is sensitive to objects with H$\\alpha$ + [\\ion{N}{2}] equivalent widths of less than 20 \\AA, and that most objects with EW $>$ 30 \\AA\\ are detected.  The median line flux of the KISS sample is more than three times lower than that of the UCM survey (Gallego \\etal 1996).  The luminosity distribution of the KISS ELGs is dominated by intermediate- and low-luminosity galaxies, although luminous AGN and starbursting galaxies are also represented.  The median absolute magnitude of M$_B$ = $-$18.96 is characteristic of a small spiral galaxy or large Magellanic irregular, and underscores the fact that strong-lined galaxies of the type cataloged by KISS tend to be less luminous than the types of objects found in more traditional magnitude-limited samples.  Despite the fact that one can learn a great deal about each KISS object from the survey data alone, detailed follow-up spectra are still required to get a more complete picture.  For example, one cannot distinguish between AGN and star-formation activity in the KISS galaxies based on the objective-prism spectra.  Further, the redshifts derived from the KISS spectral data are too coarse to be used in detailed spatial distribution studies (e.g., Lee, Salzer \\& Gronwall 2001).  We are in the process of obtaining spectra for a large number of KISS ELGs in order to better assess the nature of the individual galaxies, as well as to allow for a wide range of science applications, many of which are outlined in Paper I."
    },
    "0010/astro-ph0010360_arXiv.txt": {
        "abstract": "}[2]{{\\footnotesize\\begin{center}ABSTRACT\\end{center} \\vspace{1mm}\\par#1\\par \\noindent {~}{\\it #2}}}  \\newcommand{\\TabCap}[2]{\\begin{center}\\parbox[t]{#1}{\\begin{center} \\vglue-9mm \\small {\\spaceskip 2pt plus 1pt minus 1pt T a b l e} \\refstepcounter{table}\\thetable \\\\[1mm] \\footnotesize #2 \\end{center}}\\end{center}}  \\newcommand{\\TableSep}[2]{\\begin{table}[p]\\vspace{#1} \\TabCap{#2}\\end{table}}  \\newcommand{\\FigCap}[1]{\\footnotesize\\par\\noindent Fig.\\  % \\refstepcounter{figure}\\thefigure. #1\\par}  \\newcommand{\\TableFont}{\\footnotesize} \\newcommand{\\TableFontIt}{\\ttit} \\newcommand{\\SetTableFont}[1]{\\renewcommand{\\TableFont}{#1}}  \\newcommand{\\MakeTable}[4]{\\begin{table}[htb]\\TabCap{#2}{#3} \\begin{center} \\TableFont \\begin{tabular}{#1} #4 \\end{tabular}\\end{center}\\end{table}}  \\newcommand{\\MakeTableSep}[4]{\\begin{table}[p]\\TabCap{#2}{#3} \\begin{center} \\TableFont \\begin{tabular}{#1} #4 \\end{tabular}\\end{center}\\end{table}}  \\newenvironment{references}% { \\footnotesize \\frenchspacing \\renewcommand{\\thesection}{} \\renewcommand{\\in}{{\\rm in }} \\renewcommand{\\AA}{Astron.\\ Astrophys.} \\newcommand{\\AAS}{Astron.~Astrophys.~Suppl.~Ser.} \\newcommand{\\ApJ}{Astrophys.\\ J.} \\newcommand{\\ApJS}{Astrophys.\\ J.~Suppl.~Ser.} \\newcommand{\\ApJL}{Astrophys.\\ J.~Letters} \\newcommand{\\AJ}{Astron.\\ J.} \\newcommand{\\IBVS}{IBVS} \\newcommand{\\PASP}{P.A.S.P.} \\newcommand{\\Acta}{Acta Astron.} \\newcommand{\\MNRAS}{MNRAS} \\renewcommand{\\and}{{\\rm and }} {We present results of age determination based on the standard procedure of isochrone fitting for about 600 star clusters younger than about 1.2~Gyr from the central parts of the LMC. Comparison of age distributions of star clusters from the LMC and the SMC shows that cluster formation histories are different in these galaxies. The age distribution of the LMC clusters reveals bursty nature of cluster formation in this galaxy, with contrast to the relatively uniform distribution of cluster ages in the SMC. We detected three extended peaks in the distribution of ages of LMC clusters, located at about 7~Myr, 125~Myr and 0.8~Gyr. All detected peaks have complex structure. While the structure of the youngest and the oldest peaks may be spurious due to accuracy of age determination, in the middle peak two evident sub-peaks at 100~Myr and 160~Myr are clearly seen. Similar peaks are seen in the distribution of ages of clusters from the SMC, which indicates that increased cluster formation ratio during these periods might be caused by the last encounter between the Magellanic Clouds. The sample of clusters from the LMC is overabundant in very young (${<10}$~Myr) and old (${>400}$~Myr) clusters, and underabundant in clusters having ages in the range from 80 to 15~Myr with respect to the studied sample of clusters from the SMC.}{Magellanic Clouds -- Galaxies: star clusters} ",
        "introduction": "The analysis of rich systems of clusters in the Magellanic Clouds and in the Galaxy provides many information about parent galaxies and processes of star and cluster formation. Previous investigations revealed that star clusters from the Magellanic Clouds and from the Galaxy differ in a number of important respects (van den Bergh 1991). In particular, the cluster formation histories in these galaxies are very different. In the Milky Way there are two distinct classes of old massive globular clusters and less massive, young open ones. In the Magellanic Clouds there are also objects of intermediate mass. The age distribution of clusters from the LMC shows bursts of star formation and quiescent periods in contrast to the continuous age distribution of clusters from the SMC (van den Bergh 1991). Unfortunately, the number of cluster from the Magellanic Cloud having precise data for age determination was still to small to make final statement about star cluster formation history in these galaxies. In particular information about the ages of significant fraction of the most numerous clusters, younger than one Gyr, was lacking.  The microlensing experiments have been providing us with incredible amount of photometric data. The observations collected in the course of the OGLE-II project (Udalski, Kubiak and Szyma{\\'n}ski 1997), the only microlensing experiment that uses standard {\\it BVI} passbands, are especially well suited for detecting and analysis of properties of star clusters from the Magellanic Clouds. In a series of paper we present results of our investigations. Using the automated and algorithmic method the catalogs of almost 1000 clusters from the Magellanic Clouds (Pietrzy{\\'n}ski \\etal 1998, 1999) were constructed. Based on these catalogs 99 multiple cluster candidates were selected and described by Pietrzy{\\'n}ski and Udalski (1999c, 2000). Many observations (typically 400) for each star collected during several years provide unique possibility to explore populations of variable stars from the Magellanic Clouds clusters (Pietrzy{\\'n}ski and Udalski 1999 ad). Precise photometry down to ${V\\approx21.5}$~mag (Udalski \\etal 1998) allowed us to derive ages for 93 clusters younger than 1~Gyr from the SMC (Pietrzy{\\'n}ski and Udalski 1999b). In this contribution we present the catalog of ages of about 600 clusters from the Large Magellanic Cloud and compare distributions of ages of clusters from the SMC and LMC. ",
        "conclusions": "We present results of age determination based on the standard procedure of isochrone fitting for about 600 star clusters from the central parts of the LMC. Based on the dataset, which is order of magnitude larger than available so far, we confirmed that cluster formation history of the LMC and the SMC are different. The distribution of ages of clusters from the LMC reveals periods of bursty cluster formation and quiescent periods. We detected three extended periods of enhanced cluster formation at about 7~Myr, 125~Myr and 0.8~Gyr ago. No such features are seen in the relatively uniform age distribution of the SMC clusters. Both age distributions reveal peaks located at about 100~Myr and 160~Myr, which might be connected with the last encounter of the Magellanic Clouds.  \\Acknow{The paper was partly supported by the Polish KBN grants 2P03D00814 to A.~Udalski. Partial support for the OGLE project was provided with the NSF grant AST-9530478 to B.~Paczy\\'nski.}"
    },
    "0010/astro-ph0010156_arXiv.txt": {
        "abstract": "The effect of the nebula optical depth on the determination of the temperature (T$_*$) of the central stars in planetary nebulae is discussed. Based on photoionization models for planetary nebulae with different optical depths, we show, quantitatively, that the details of the distribution of the H and He II Zanstra temperatures are mainly explained by an optical depth effect; in particular, that the discrepancy is larger for low stellar temperatures. The results also show that for high stellar temperatures the He II Zanstra temperature underestimates the stellar temperature, even for high optical depths. The stellar temperature, as well as the optical depth, can be obtained from a Zanstra temperature ratio (ZR) plot ZR = T$_Z$(He II)/T$_Z$(H) versus T$_Z$(He II). The effects of departures from a blackbody spectrum, as well as of the He abundance in the nebulae, are also discussed. For nebulae of very low optical depth and/or high stellar temperature the distribution ZR versus T$_Z$(He II) only provides lower limits for T$_*$. In order to obtain better values for the optical depth and T$_*$, we propose the use of the line intensity ratio He II/He I versus T$_Z$(He II) diagram. ",
        "introduction": "The temperature of the central stars of planetary nebulae (PNs) is an essential parameter for evolutionary studies as well as for an analysis of the nebulae themselves. The different methods for the determination of the stellar temperature and the problems involved were extensively discussed by Pottasch (1984) and Kaler (1985a, 1989). The most common method was suggested by Zanstra (1931) and further developed by Harman \\& Seaton (1966). The temperature of the ionizing star of a planetary nebula is calculated from the ratio between the flux of a recombination line and the stellar continuum flux at a given frequency. The Zanstra method assumes that all photons above the H (or the He$^+$) Lyman limit are absorbed within the nebula and that each recombination eventually results in a Balmer photon. The total ionizing flux can then be related to the total flux of a recombination line. The Zanstra method yields then two values for the stellar temperature: the Zanstra temperature obtained from the intensity of a hydrogen recombination line, T$_Z$(H), and the Zanstra temperature obtained from a He II recombination line, T$_Z$(He II). When applied to observed nebulae,  T$_Z$(H) is generally lower than T$_Z$(He II); the difference can reach values of the order of 60,000 K (Kaler 1983b), and the temperature ratio can reach a factor higher than 3 (Kaler 1983b, 1985a). This is the well known Zanstra discrepancy, which is stronger for PNs with low stellar temperatures. In fact, many PNs with  T$_Z$(H) $<$ 100,000 K have T$_Z$(He II) $>$ T$_Z$(H) while, for higher stellar temperatures, both Zanstra temperatures are similar (Pottasch, 1984; Gathier \\& Pottasch 1988, 1989). Another important point is that the calculated Zanstra temperatures do not reproduce the high temperatures predicted by evolutionary models (Kaler 1985a, 1989; Stasi\\'nska \\& Tylenda 1986).  Another interesting feature that could be related to the Zanstra temperature issues is the distribution of objects in a log L - log T plane. As noted by Shaw \\& Kaler (1989), when this distribution is based on Zanstra temperatures there is a dense crowd of planetary nebula nuclei with temperatures of $\\sim$ 100,000 K and a strong decrease towards higher temperatures, what these authors call the ``Zanstra wall''.  The causes of the Zanstra  discrepancy have been discussed by several authors (Pottasch 1984; Kaler 1985a, 1989; Henry \\& Shipman 1986;Stasi\\'nska \\& Tylenda 1986; Kudritzki \\& M\\'endez 1989; Gabler, Kudritzki \\& M\\'endez 1991; M\\'endez, Kudritzki \\& Herrero 1992) and can be related to  the following effects: (1) optical effects, i.e., nebulae exhibiting a Zanstra discrepancy would be optically thin to photons ionizing H yet optically thick to those ionizing He$^+$; (2) differential dust absorption in the nebula; (3) the stellar continuum differing from the usually  assumed blackbody spectrum; for example, an excess of photons with energies beyond the He$^+$ ionization potential would result in a high He II Zanstra temperature.  Objects with a large discrepancy show fainter low-ionization lines, suggesting that the effect 1 is the correct interpretation (Kaler 1983b). Simple evolutionary models of planetary nebulae predict T$_Z$(H) $\\neq$  T$_Z$(He II) during the nebula lifetime because of the variation of their optical depth (Tylenda et al. 1994; see also Sch\\\" onberner \\& Tylenda 1990). The effect of dust on the calculated Zanstra temperature is discussed by Helfer et al. (1981). For an assumed absorption law (with a high opacity around 50 eV), Zanstra temperatures underestimate the actual stellar temperature. The effect is more intense for T$_Z$(H) and leads to the Zanstra discrepancy. However, dust appears to be only important in some specific nebulae (Kaler 1985a). Departures from the blackbody, in particular an excess of photons beyond the He$^+$ ionization potential, are suggested by observational studies of  many central stars of planetary nebulae (Kaler 1985a; see also references in Henry \\& Shipman 1986). An ionizing spectrum with an excess of high energy photons (relative to a blackbody), produced by a star with a less than solar atmospheric He abundance, leads to a He II Zanstra temperature higher than T$_Z$(H) (Henry \\& Shipman 1986).  Detailed theoretical analyses, using photoionization models and applied to optically {\\it thick} nebulae, are presented by Stasi\\'nska \\& Tylenda (1986) and Henry \\& Shipman (1986). Stasi\\'nska \\& Tylenda (1986) show that for low stellar temperatures (T$_*$ $\\leq$ 100,000 K), both Zanstra temperatures are similar. For higher stellar temperatures, T$_Z$(HI) is larger and T$_Z$(He II) is lower than T$_*$. However, this result is  opposite to what is obtained from observations. Analyzing only models with T$_*$ $\\leq$ 150,000 K, Henry and Shipman (1986) conclude that T$_Z$(H) is a good measure of the stellar temperature. In fact, for these values of T$_*$, T$_Z$(H) and T$_Z$(He II) are similar (Stasi\\'nska \\& Tylenda 1986) and provide a good stellar temperature estimation {\\it if the nebula is completely optically thick}.  Other methods for determining the stellar temperature include modelling of stellar absorption line profiles, ionic ratios, fitting of model atmospheres, the energy balance or Stoy method, stellar UV energy distribution, etc. The results from these different methods were compared with the values given by the Zanstra method in order to explain the Zanstra discrepancy; however, these methods give discordant results and have many uncertainties (see, for instance, Kaler 1985a, Stasi\\'nska \\& Tylenda 1986, and Kaler 1989).  Many authors have adopted T$_Z$(He II) as representative of the stellar temperature assuming that the Zanstra discrepancy is due to an optical depth effect (Kaler 1983b; Gleizes, Acker \\& Stenholm 1989; Kaler, Shaw \\& Kwitter 1990; Kaler \\& Jacoby 1991; Stanghellini, Corradi \\& Schwarz 1993). However, T$_Z$(He II) may not be a good indicator of the stellar temperature since it never reaches values as high as predicted by theoretical stellar evolutionary studies.  Usually the discussion of the Zanstra discrepancy in terms of an optical depth effect is based on optically thin {\\it or} thick nebulae at 13.6 eV or 54.4 eV. It is also implicitly assumed that the nebula is optically thick at 54.4 eV at a distance to the central star smaller than that corresponding to 13.6 eV. Harman \\& Seaton (1966) suggest the following criteria for the complete absorption of H$^0$, He$^0$, and He$^+$ ionizing-photons: presence of [OI] lines, He I images smaller than H I images, and He$^{++}$ fractional abundance $\\leq$ 0.75, respectively. For Pottasch (1984) the Zanstra method must work for $\\tau_{13.6}$ $>$ 1. Some authors define a criterion to distinguish between optically thin and thick objects using the ratio of the Zanstra temperatures, ZR = T$_Z$(He II)/T$_Z$(H). For example, for Shaw \\& Kaler (1985) the nebula is optically thick to the H Lyman continuum when ZR $\\lsim$ 1.2, while for ZR $\\gsim$ 2.5 and He II $\\lambda$4686/H$\\beta$ $\\gsim$ 0.9 it is thin for He$^+$ Lyman continuum photons, T$_Z$(He II) being a lower limit for the stellar temperature. The criterion used by Kaler \\& Jacoby (1989), based on line intensities of low-ionization lines, states that a nebula is thick when [O II] $\\lambda$3727/H$\\beta$ $\\geq$ 1 and [N II] $\\lambda$6584/H$\\alpha$ $\\geq$ 1. However, planetary nebulae can present a large range of optical depths, depending on the quantity of matter. Furthermore, even considering the central stellar radiation as a blackbody, different stellar temperatures correspond to different ratios between the number of ionizing photons with energy higher than 54.4 eV and those higher than 13.6 eV. Thus, the radial ionic distribution for H and He varies with T$_*$, changing the relative sizes of the H$^+$ and He$^{++}$ zones with the nebula optical depth at 13.6 eV and 54.4 eV. As remarked by Stasi\\'nska \\& Tylenda (1986), the radiation transfer is much more complicated than assumed by the Zanstra method.  In brief, the Zanstra temperature is commonly used in the literature for planetary nebula modelizations as well as evolutionary analysis, and low optical depth must be at least a partial explanation for the Zanstra discrepancy. However, a more detailed analysis is required in order to explain the issues listed above. In this paper, a careful analysis of the effect of the nebula optical depth on the determination of the Zanstra temperatures is intended. The effects on the Zanstra temperatures due to deviations of a blackbody spectrum and due to an overabundance of He in the nebula are also discussed. The theoretical models used in our analysis are described in \\S 2. The results for the Zanstra temperature ratio (ZR) and its behavior with the stellar temperature and with the nebula optical depth appear in \\S 3, which also includes a comparison with values derived from PNs observations. An alternative method to estimate the temperature of central stars of planetary nebulae is suggested and discussed in \\S 4. The conclusions are outlined in \\S 5. ",
        "conclusions": ""
    },
    "0010/astro-ph0010226_arXiv.txt": {
        "abstract": "In recent papers we have proposed that the dark matter of the Universe could be from scalar field origin. In this letter, we find that if the scale of renormalization of the model is of order of the Planck Mass, then a scalar field $\\Phi $ endowed with the scalar potential $V=V_{o}\\left[ \\cosh {\\left( \\lambda \\sqrt{\\kappa _{0}}\\Phi \\right) }-1\\right] $ can be a reliable model for dark matter in galaxies. The predicted scattering cross section fits the value required for self-interacting dark matter and the additional degree of freedom of the theory is of order of hundreds of TeV. ",
        "introduction": "Recent cosmological observations suggest a flat Universe in a current accelerated expansion but full with 95\\% of unknown matter. 25\\% of the matter in the Universe, widely known as dark matter, is responsible of the formation of the large scale structure we observe in the sky: galaxies, clusters of galaxies, voids, walls, etc. This model of dark matter has been very successful\\cite{triangle}.  However, at galactic scale, numerical simulations show some discrepancies between dark-matter predictions and observations\\cite{stein}. Dark matter simulations show cuspy halos of galaxies with an excess of small scale structure, while observations suggest a constant halo core density\\cite{avila} and a small number of subgalactic objects\\cite{avila2}. In order to solve these discrepancies, there are some proposals of scalar fields as dark matter in the literature nowadays\\cite{peebles,jeremy,hu,bento,tkachev}. All of them have assumed a polynomial scalar potential up to fourth order. At the cosmological scale, the main results are cold dark matter-like behavior, growing linear perturbations, the existence of a Jeans scale and suppression of the mass power at small scales, but they depend upon initial conditions. At the galactic scale, where the self-interaction becomes important, the scattering cross section is proportional to the quartic coupling square $g^2$, which is a free parameter of the models. Even if there is a model-independent result in which a particle with a value for the scattering cross section given by (\\ref{sigma}) has, straightforwardly, a mean-free path of order of 1 Mpc (see eq. (4) in\\cite{bento}), the free parameters of the potential are very related, as we can see in eq. (6) of ref.\\cite{bento}. Depending of the coupling $g$ the mass of the scalar field could be of order of MeV, inclusive. But nobody has determined the value of this coupling.  In this paper, we continue investigating the hypothesis of a scalar field as dark matter in the Universe with a $\\cosh$ scalar potential\\cite{luis,luis2,sahni}  \\begin{equation} V(\\Phi ) =V_0\\left[ \\cosh {(\\lambda \\,\\sqrt{\\kappa _{o}} \\Phi )}-1\\right] \\label{cosh} \\end{equation}  \\noindent with $\\kappa _{0}^{-1/2}=(8\\pi G)^{-1/2}=M_{Pl}/\\sqrt{8\\pi }=2.4\\times 10^{18}\\,GeV$ being the inverse square of the reduced Planck mass. Using results given in renormalization theory, we calculate the $2\\rightarrow 2$ cross section at the lowest order for the potential (\\ref{cosh}). It is found that the result corresponds to an effective $\\phi^4$-theory in which the effective coupling $g$ is exponentially enhanced by the scale of renormalization $\\Lambda$, which remains as a free parameter of the model. This result is compared with the value given for the cross section in self-interacting dark matter models. We found that if the scale of renormalization of potential (\\ref{cosh}) is of order of the Planck Mass, $\\Lambda ={\\mathcal {O}}(M_{Pl})$, the predicted scattering cross section by mass of the scalar particles $\\sigma_{2\\rightarrow 2}/m_{\\Phi }$ can fit the value predicted in numerical simulations of self-interacting dark matter in order to avoid high-density dark matter halos. ",
        "conclusions": ""
    },
    "0010/astro-ph0010010_arXiv.txt": {
        "abstract": "From 1995 to 1999, daily monitoring of the radio emission from the Crab pulsar at the Green Bank and Jodrell Bank observatories revealed a series of six sudden rotational spinups or glitches, doubling the number of glitches observed for this pulsar since 1969.  With these observations, the range of time intervals between significant Crab glitches has widened considerably, indicating that the occurrence of Crab glitches may be more random than previously thought. The new glitch amplitudes $\\Delta\\nu/\\nu$ span an order of magnitude from $2 \\times 10^{-9}$ to $3 \\times 10^{-8}$. Except in one case, which we suggest may represent an ``aftershock'' event, the frequency jumps display an exponential recovery with a timescale of $\\sim$3 days for the smaller glitches and $\\sim$10 days for the largest (1996) glitch. In the largest event, a portion of the spinup was resolved in time, as was previously reported for the 1989 glitch. A pronounced change in $\\dot{\\nu}$ also occurs after each glitch and is correlated with the size of the initial frequency jump, although for some of the smaller glitches this appears to be a temporary effect. We discuss the properties of the ensemble of observed Crab glitches and compare them with the properties of Vela glitches, highlighting those differences which must be explained by evolutionary models. ",
        "introduction": "}  Pulsar timing observations reveal that sudden increases in rotation rate, known as {\\it glitches}, are a common feature in some pulsars, especially younger ones \\citep{She96}. These spinup events are superposed on the long-term rotational slowdown of the pulsar due to magnetic dipole radiation and particle outflow, and are usually followed by a relaxation towards the extrapolated pre-glitch frequency over a period of days to weeks. Continuous monitoring of the Crab pulsar (PSR B0531+21) uncovered six significant glitches from 1969 to 1992, at intervals of 3--6 years \\citep*{LPS93}. High glitch activity has been observed in other pulsars as well. For example, at least twelve glitches of the Vela pulsar (PSR B0833$-$45) were observed between 1969 and 1996 \\citep[e.g.,][]{Chau93,Fla95}, usually at intervals of 2--3 years, and twelve glitches have been detected in the rotation of PSR J1341$-$6220 during just 8.2 years of observations \\citep{Wang00}, making it the most frequently glitching pulsar known.  At the same time, precise timing measurements have revealed that many pulsars exhibit an unpredictable fluctuation in pulse arrival time known as {\\it timing noise}. A number of previous studies \\citep{Boy72,Gro75,Cor80} have shown that timing noise in the Crab pulsar can be modeled as a random walk in frequency. The individual steps in this random walk have not been resolved, but can be of either sign and have amplitudes $|\\Delta\\nu/\\nu| \\ll 10^{-10}$, whereas major glitches of the Crab can be distinguished as sudden positive jumps in frequency with $\\Delta\\nu/\\nu \\gtrsim 10^{-9}$, followed by an exponential recovery over a period of days to months. Still, the possibility remains that glitches lie at the upper end of a continuous spectrum of frequency jumps \\citep{CH80}, and careful consideration of observational selection effects is essential before attempting to draw general conclusions about glitches.  Although our understanding of the glitch phenomenon is limited by our incomplete knowledge of neutron star structure, the most plausible explanation for the frequency spinups is the dumping of angular momentum from an unseen reservoir into the part of the star whose rotation we measure \\citep[e.g.,][]{Alp93}. Since the outer crust and core superfluid are believed to be coupled on short ($<$1 minute) timescales to the magnetic field and the observed pulsar beam (e.g., \\citealt*{Alp84}; \\citealt{Mend98}), the probable site of this reservoir is a superfluid component in the inner crust. As discussed by \\citet{And75} and others, the quantized vortex lines which carry angular momentum in the superfluid can be ``pinned'' to lattice sites in the neutron-rich inner crust, inhibiting the outward motion of vortex lines that occurs as the star spins down and creating an excess rotation of the superfluid relative to the outer crust. If large numbers of pinned vortices were to simultaneously unpin, the excess angular momentum would be shared with the rest of the star and a glitch would be observed. While this scenario provides an attractive theoretical framework for understanding glitches, the mechanisms that trigger glitches and lead to the observed postglitch recovery remain controversial \\citep[e.g.,][]{Alp96,Lin96,Jon98}.  Observations of pulsar glitches, in addition to providing insights into the glitch phenomenon itself, offer one of the few direct probes of neutron star structure and the physics of ultradense matter. For instance, the frequent large glitches ($\\Delta\\nu/\\nu \\sim 10^{-6}$) seen in the Vela pulsar indicate that the superfluid component in the inner crust comprises $\\sim$2\\% of the star's moment of inertia (e.g., \\citealt{Chau93}; \\citealt*{Lin99}). Crab glitches, on the other hand, exhibit much smaller relative frequency jumps ($\\Delta\\nu/\\nu \\sim 10^{-8}$), and tend to be dominated by persistent changes in the slowdown rate with $\\Delta\\dot{\\nu}/\\dot{\\nu} \\sim 10^{-4}$ \\citep[e.g.,][]{Gull77,Dem83}.  The persistent shift in $\\dot{\\nu}$ has been attributed to either a change in the external torque \\citep*{Lin92} or a change in the moment of inertia acted on by the torque \\citep{Alp96}. The various exponential relaxation timescales that follow glitches have been interpreted, in the framework of the ``vortex creep'' theory, as resulting from distinct regimes of superfluid pinning to the crustal lattice \\citep{Alp93}.  Since the occurrence of glitches is unpredictable and their relaxation timescales can be quite short (a few days or less), daily monitoring of frequent glitchers such as the Crab and Vela pulsars are necessary in order to study the glitch process in detail.  Here we present radio timing observations taken during a period of increased glitching in the Crab that began in 1995 and has extended into 2000.  The data were collected as part of regular pulsar monitoring programs with dedicated small telescopes at the NRAO\\footnote{The National Radio Astronomy Observatory (NRAO) is operated by Associated Universities, Inc, with funding from the National Science Foundation.} Green Bank and the University of Manchester Jodrell Bank sites.  The data acquisition and reduction procedures are described in \\S 2, and in \\S 3 we present our analysis of six distinct frequency jumps during this period.  A large glitch was observed in 1996 June and smaller events were observed in 1995 October, 1997 January, February, and December, and 1999 October. (The recent large glitch on 2000 July 15 is not covered by these observations).  In \\S 4 we examine the frequency (rate of occurrence) of Crab glitches, and \\S 5 we discuss the observed properties of the glitches.  Our conclusions are summarized in \\S 6.  Throughout this paper we define the rotation frequency $\\nu$ in Hz, which can be converted to angular frequency using $\\Omega$=2$\\pi\\nu$.  \\bigskip ",
        "conclusions": "}  We have presented new timing observations of the Crab pulsar at 610 MHz during a period of increased glitching.  The occurrence of 6 rotation glitches over a period of four years, of which 4--5 were large enough to have been detected by previous monitoring efforts, marks a departure from the relatively long intervals (3--6 years) between glitches reported prior to this period.  As a result, the range of interglitch intervals has widened considerably, and shows a distribution that is consistent with a random (Poisson) process.  Continued monitoring of this young pulsar will be valuable in determining whether the occurence of glitches is primarily a regular or stochastic process.  We have also fitted simple glitch models to the data to better characterize the basic properties of Crab glitches.  We summarize these properties as follows:  \\begin{enumerate}  \\item No correlation is found between the glitch amplitude and the time since the previous glitch.  If vortex pinning occurs at a constant rate, this implies that glitches do not lead to the complete unpinning of vortices in a certain region.  \\item In agreement with previous studies \\citep[e.g.,][]{LPS93}, we find that Crab glitches tend to be accompanied by long-lasting (in most cases permanent) changes in spindown rate $\\dot\\nu$.  This change in $\\dot\\nu$ is correlates well with the transient jump $\\Delta\\nu_1$ or the glitch amplitude $\\Delta\\nu_0$, and was observed to vanish or fade $\\sim$50--100 days after two of the smallest glitches.  \\item The glitch amplitudes span a range of over an order of magnitude, from $\\Delta\\nu/\\nu \\sim 2 \\times 10^{-9}$ to $3 \\times 10^{-8}$. Smaller events are unlikely to be detectable in many cases due to the substantial timing noise in this pulsar.  \\item The average rate of spindown due to glitches (given by the activity parameter $A_g$) has not changed significantly in recent years, despite the higher rate of glitching.  This is a result of the relatively small persistent frequency jumps $\\Delta\\nu_p$ accompanying the recent glitches.  \\item Although a strong correlation between glitch amplitude and recovery timescale is not found, there appears to be an absence of large glitches with short recovery times, consistent with the hypothesis that the recovery time increases with the vortex pinning energy.  \\item The large 1996 glitch exhibits a gradual spinup similar to that observed in the giant 1989 glitch, with a slightly shorter timescale (0.5~d rather than 0.8~d).  The next glitch was a factor of $\\sim$3 smaller and completed its spinup in under 1 hour, contradicting the hypothesis that smaller glitches have more gradual spinups \\citep{Lin96}.  \\item Detailed observations of the postglitch recovery suggest secondary spinups or ``aftershocks'' may follow 20--40 d after each glitch, although these events cannot clearly be distinguished from timing noise except in the case of Glitch 10 (1997 February).  \\item Two of the recent glitches (1996 June and 1997 February) are accompanied by substantial changes in the frequency second derivative $\\ddot\\nu$, which would imply a change in the braking index $n$ by 10--20\\% from its fiducial value of 2.51.  A more likely explanation is that these are due to long-term transients associated with the glitches, as was seen following the 1989 event, that were interrupted by further glitches.  \\end{enumerate}  Our observations confirm that important differences exist between glitches in the Crab and Vela pulsars, as has been noted by other recent studies of pulsar glitches \\citep{Lyn00a,Wang00}.  Vela glitches occur at more or less regular intervals of 2--3 years, and nearly all possess similar amplitudes and recovery timescales.  A spinup in Vela has never been resolved in time and the recovery fraction is fairly small ($\\sim$20\\%, \\citealt*{LPS95}).  All of these properties are at odds with what has been observed in the Crab.  Furthermore, most Crab glitches display persistent increases in $|\\dot\\nu|$, and no Crab glitches have been observed with sizes $\\Delta\\nu_g/\\nu \\sim 10^{-6}$, whereas such ``giant'' glitches have been seen in at least five young pulsars with characteristic ages $\\tau_c \\equiv \\nu/(2|\\dot\\nu|)$ = 7--16 kyr \\citep{Wang00}, including Vela.  If these differences in glitch behavior are primarily a result of evolution, they imply substantial structural changes occurring within neutron stars over their first 10 kyr."
    },
    "0010/astro-ph0010004_arXiv.txt": {
        "abstract": "s{There are two possible causes of variability in gravitationally lensed quasars: intrinsic fluctuations of the quasar and ``microlensing\" by compact objects along the line of sight. If disentangled from each other, microlens-induced variability can be used to study two cosmological issues of great interest, the size and brightness profile of quasars on one hand, and the distribution of compact (dark)  matter along the line of sight. Here we present a summary of recent observational evidence for quasar microlensing as well as of theoretical progress in the field. Particular emphasis is given to the questions which microlensing can address  regarding the search for  dark matter, both in the halos of lensing galaxies and in a cosmologically distributed form. A discussion of desired observations and required theoretical studies is presented as a conclusion/outlook. } ",
        "introduction": " ",
        "conclusions": ""
    },
    "0010/astro-ph0010232_arXiv.txt": {
        "abstract": "If the halo of the lensing galaxy 0957+561 is made of massive compact objects (MACHOs), they must affect the lightcurves of the quasar images Q0957+561 A and B differently. We search for this microlensing effect in the double quasar by comparing  monitoring data for the two images A and B -- obtained with the 3.5m Apache Point Observatory from 1995 to 1998  -- with intensive numerical simulations. This way we test whether the halo of the lensing galaxy can be made of MACHOs of various masses. We can exclude a halo entirely made out of MACHOs with  masses between $10^{-6}\\,M_{\\odot}$ and $10^{-2}\\,M_{\\odot}$  for quasar sizes of less than $3\\times 10^{14}\\,h_{60}^{-1/2}$\\,cm, hereby extending previous limits upwards by one order of magnitude. ",
        "introduction": "The nature of the dark matter is still one of the most pressing questions in current astrophysics. Many experiments are running or being built/planned in order to test the existence of various particle physics candidates, like wimps, axions, or neutralinos (for a review on the particle physics aspects of dark matter, see Raffelt 1997). There are also intense searches underway for astrophysical dark matter candidates, like black holes (from stellar to galactic mass scales), brown dwarfs, or other compact objects.  Gott (1981) and Paczy\\'nski (1986) put forward the idea to use gravitational lensing for the search of compact dark matter objects in galactic halos. Gott (1981) suggested to look for fluctuations in the lightcurves of multiply imaged quasars in order to detect compact objects along the line of sight in the halo of the lensing galaxy, which had been shown earlier by Chang and Refsdal (1979) to have observable consequences. Paczy\\'nski (1986) proposed to  probe the content of compact dark objects in the Milky Way halo by monitoring the brightnesses of about 10$^7$  individual stars in the Large Magellanic Cloud (LMC). The  latter paper led to a whole industry of observing projects, like MACHO (Alcock et al. 2000) and EROS (Lasserre et al. 2000). Gott (1981)'s suggestion to study the halos of lensing galaxies by way of monitoring background quasars -- though in principle as powerful a method -- was pursued, however, with much less fervor in terms of people involved, total observing nights or CPU-time used.  We followed the latter method and present here results on the possible matter contents of the halo of the lensing galaxy in the lens system 0957+561. They are based on comparing lightcurves of the gravitationally lensed double quasar Q0957+561A,B, obtained by four years of monitoring (1995 - 1998) at the 3.5m-Apache Point Observatory (Colley, Kundi\\'c \\& Turner 2000, hereafter CKT) with extensive numerical simulations. ",
        "conclusions": "The lightcurves of Q0957+561A, B as monitored between 1995 and 1998 do not show any significant differences beyond $\\Delta m = 0.05$\\,mag, when corrected for time delay and magnitude difference. In Fig.~\\ref{Fig_lego_1a} we present the resulting ``exclusion probabilities'' for the two parameters ``Macho mass'' $m_{\\rm macho}$ and quasar size $\\sigma_{\\rm Q}$ (assuming 100\\% of the halo mass is in MACHOs), % derived from four years of monitoring and comparison with numerical simulations. The numbers indicate the percentage of the 100,000 simulated lightcurves that showed fluctuations larger than the observed ones. In the diagram, we encircled and shaded the parts of parameter space that produced exclusion probabilities of 67\\% (1-$\\sigma$, thin shading), 95\\% (2-$\\sigma$, medium shading), and 99.7\\% (3-$\\sigma$, cross shading). It is obvious that  a mass range from $10^{-6} \\le m_{\\rm macho}/M_\\odot \\le 10^{-2}$ can be excluded at the 3-$\\sigma$ level for all quasar sizes except for the largest one, for which $10^{-2}\\,M_\\odot$ is barely allowed. For objects of mass 0.1 $M_\\odot$, the exclusion probability is at the 70\\% to 85\\% level. We need a longer time coverage, in order to improve significantly on these limits. In Figs.~\\ref{Fig_lego_1b} and \\ref{Fig_lego_1c} we present the resulting numbers and respective exclusion regions  for the assumption of only 50\\% or 25\\% of the halo mass in MACHOs. The exclusion probabilities get slightly smaller in these cases, and the exclusion regions shrink a bit as well, but the general picture does not change much.  \\begin{figure}[htb] \\resizebox{9.0cm}{!}{\\includegraphics{wambsganss_Ci291_fig3.ps}} \\caption{Exclusion Probability (in percent) as a function of Macho mass $m_{\\rm macho}$ and quasar size $\\sigma_{\\rm Q}$, for 100\\% of the halo mass in MACHOs. The shaded parts encircle regions of parameter space that can be excluded at the 67\\%, 95\\%, and 99.7\\% probability level (increasing line density; the latter two coincide). } \\label{Fig_lego_1a} \\end{figure} \\begin{figure}[htb] \\resizebox{8.7cm}{!}{\\includegraphics{wambsganss_Ci291_fig4.ps}} \\caption{Same as Fig.~\\ref{Fig_lego_1a} for 50\\% of the halo mass in MACHOs.} \\label{Fig_lego_1b} \\end{figure} \\begin{figure}[htb] \\resizebox{8.7cm}{!}{\\includegraphics{wambsganss_Ci291_fig5.ps}} \\caption{Same as Fig.~\\ref{Fig_lego_1a} for 25\\% of the halo mass in MACHOs.} \\label{Fig_lego_1c} \\end{figure}     Pelt et al. (1998)  and Refsdal et al. (2000) investigated the double quasar lightcurve as well. Their main focus was microlensing on medium and long time scales (baseline 15 years), including a continuous change in the difference lightcurve of about 0.25 mag in the first five years. Whereas their data set covers a longer baseline than ours, we have many more and more accurate data on short time scales ($\\le$ 100 days). Furthermore, we also take into account the       gaps in the difference lightcurve, which  means we treat the short term behaviour more realistically. Pelt el al. (1998)'s  finding  that the quasar size is about $3\\times 10^{15}$cm is consistent with our results (though we cannot put an upper limit on the size).  We extend the ``exclusion'' area by roughly one order of magnitude in mass, compared to the first results in SW98 and Wambsganss \\& Schmidt (1998). This is due to the fact that the coverage of the difference light curve increased by more than a factor of 4 (without showing any more variability), and the mass limits increase with the square of the length scale. But it also means that in order to increase the limits from short/medium term microlensing by another order of magnitude in mass -- reaching the very interesting regime of solar mass objects -- the frequent monitoring has to continue for another six or eight years.     The  method and results described here, in particular the exclusion diagram (Fig.~3; see also Fig.~2 of Refsdal et al. 2000), are very similar to those of the groups investigating microlensing of the Milky Way halo (e.g., Alcock et al. 2000, Lasserre et al. 2000). Hence it is obvious that monitoring multiple quasars (Gott 1981) is as powerful a tool in constraining the abundance of MACHOs in galactic halos as is monitoring LMC stars (Paczy\\'nski 1986)."
    },
    "0010/astro-ph0010142_arXiv.txt": {
        "abstract": "The intrinsic shape and orientation of the elliptical galaxy NGC 3379 are estimated by dynamical modeling. The maximal ignorance shape estimate, an average over the parameter space, is axisymmetric and oblate in the inner parts, with an outward triaxiality gradient. The $1\\sigma$ limits on total-mass triaxiality $T$ are $T<0.13$ at $0.33\\kpc$ and $T=0.08\\pm 0.07$ at $3.5\\kpc$ from the center. The luminous short-to-long axis ratio $c_L=0.79 {+0.05 \\atop -0.1}$ inside $0.82\\kpc$, flattening to $c_L=0.66 {+0.07 \\atop -0.08}$ at $1.9\\kpc$. The results are similar if the galaxy is assumed to rotate about its short axis. Estimates for $c_L$ are robust, but those for $T$ are dependent on whether the internal rotation field is disklike or spheroid-like. Short-axis inclinations $i$ between $30\\arcdeg$ and $50\\arcdeg$ are preferred for nearly axisymmetric models; but triaxial models in high inclination are also allowed, which can affect central black hole mass estimates. The available constraints on orientation rule out the possibility that the nuclear dust ring at $R\\approx 1\\farcs 5$ is in a stable equilibrium in one of the galaxy's principal planes. The ring is thus a {\\em decoupled nuclear component\\/} not linked to the main body of the galaxy. It may be connected with ionized gas that extends to larger radii, since the projected gas rotation axis is near the minor axis of the ring. The gas and dust may both be part of a strongly warped disk; however, if caused by differential precession, the warp will wind up on itself in a few $10^{7}$ years. The decoupling with the stellar component suggests that the gas has an external origin, but no obvious source is present. ",
        "introduction": "Like many astronomers, the ``standard elliptical galaxy'' NGC 3379 appears outwardly normal but, deep inside, has serious issues. To the casual observer this object is as ordinary as can be. Its optical surface brightness distribution follows an $r^{1/4}$ law over a span of 10 magnitudes (de Vaucouleurs \\& Capaccioli \\markcite{dVC79}1979). Its isophotes are almost perfectly elliptical and aligned (Peletier et al.\\ \\markcite{Pel90}1990). Its optical colors are typical of an old population, and color gradients are small (Goudfrooij et al.\\ \\markcite{Gou94}1994). It rotates slowly about its apparent minor axis (Davies \\& Illingworth \\markcite{DaI83}1983, Davies \\& Birkinshaw \\markcite{DB88}1988). It shows no sign of current or past interactions (Schweizer \\& Seitzer \\markcite{ScS92}1992), and contains only a tiny amount of detectable interstellar matter, seen only in H$\\alpha +$[\\ion{N}{2}] emission (Macchetto et al.\\ \\markcite{Mac96}1996) and dust absorption (van Dokkum \\& Franx \\markcite{vDF95}1995, Michard \\markcite{Mic98}1998).  On closer examination, the photometric and kinematic structure of the galaxy are puzzling. Deviations of the surface brightness profile from the $r^{1/4}$ law, though small, are significant (Capaccioli et al.\\ \\markcite{CaH87}1987). Based on photometric similarities to NGC 3115, Capaccioli et al.\\ \\markcite{CVHL}(1991) propose that NGC 3379 is a highly triaxial S0 galaxy seen at low ($31\\arcdeg$) inclination. The orientation and shape of the galaxy are difficult to constrain, however. Van der Marel et al.\\ \\markcite{vdM90}(1990), using stellar kinematic data and explicitly axisymmetric models, infer an inclination of $60\\arcdeg$, at which the galaxy would be intrinsically almost as round as it appears. Statler \\markcite{Sta94}(1994c, hereafter S94) considers triaxial models, introducing a methodology that relies on multi-position-angle observations of the mean velocity field. These models rule out the axis ratios advocated by Capaccioli et al.\\ \\markcite{CVHL}(1991) at $98\\%$ confidence,\\footnote{The model of Capaccioli et al.\\ is not internally consistent due to unfortunate trigonometric errors (S94, \\S\\ 5.2).}\\ but allow a wide variety of flattened nearly axisymmetric, or rounder triaxial shapes, and a correspondingly wide variety of inclinations.  Detailed kinematic studies of both stars (Bender, Saglia, \\& Gerhard \\markcite{BSG94}1994; Statler \\& Smecker-Hane \\markcite{SS99}1999, hereafter SS99) and planetary nebulae (Ciardullo, Jacoby, \\& Dejonghe \\markcite{CJD93}1993) show a richer kinematic structure than one might have expected from older data. Inflections in the rotation and dispersion profiles are suggestive of a two-component structure, and are similar to those of S0 galaxies (Fisher \\markcite{Fis97}1997). SS99 suggest a low ($30\\arcdeg$) inclination and a {\\em weakly\\/} triaxial S0-like shape on the basis of these similarities. At large radii the stellar dispersion profile joins with that of the planetary nebulae, which is consistent with a {\\em constant\\/} mass-to-light ratio out to $r \\approx 10\\kpc$ (Ciardullo, Jacoby, \\& Dejonghe \\markcite{CJD93}1993).  More recently, the nuclear regions of NGC 3379 have caught the attention of a number of workers, particularly as a likely home for a supermassive black hole. Kinematic data from the ground and from {\\em HST\\/} imply a central mass in the vicinity of $10^8 \\msun$ (Magorrian et al.\\ \\markcite{Mag98}1998; Gebhardt et al.\\ \\markcite{Geb00a}2000a, hereafter G00), placing it roughly in the middle of the $M_{\\rm BH}$-$L_{\\rm bulge}$ (Magorrian et al.\\ \\markcite{Mag98}1998) and $M_{\\rm BH}$-$\\sigma_{\\rm bulge}$ (Ferrarese \\& Merritt \\markcite{FeM00}2000, Gebhardt et al.\\ \\markcite{Geb00b}2000b) correlations, but the black hole mass is dependent, to within factors of order unity, on the orientation of the galaxy. The nuclear dust ring (\\markcite{Geb00a}G00, Pastoriza et al.\\ \\markcite{Pas00}2000, hereafter P00) seen on arcsecond scales and the slightly more extended H$\\alpha +$[\\ion{N}{2}] emission (Macchetto et al.\\ \\markcite{Mac96}1996, \\markcite{Pas00}P00) might be useful in constraining the orientation, if it could be assumed that the interstellar material were in an equilibrium configuration in one of the galaxy's principal planes. However, there seems not to be compelling justification for such an assumption, since the apparent principal axes of the dust and ionized gas are aligned neither with the stellar principal axes nor with each other. Figure \\ref{f.sketch} shows a sketch of these components and their relation to the starlight.  \\begin{figure}[t]{\\epsfxsize=3in\\hfil\\epsfbox{statler.fig1.eps}} \\caption{\\footnotesize Sketch of the central regions of NGC 3379, oriented with North at the top and East at left. A typical stellar isophote is shown by the big ellipse. The apparent rotation axis of the stars coincides very closely with the morphological minor axis, shown as the thick dotted line. The convoluted outline shows the outermost H$\\alpha$+\\ion{N}{2} isophote plotted by Machetto et al.\\ (1996). The gas rotation is measured only within $3\\arcsec$ of the center, and the rotation axis is poorly determined. Gas kinematics are consistent with the rotation axis lying along the apparent minor axis (light dotted line) of the central dust ring (thick ellipse). Both the stellar and gas velocity fields are such that the right side of the figure is at positive velocities and the left side at negative. \\label{f.sketch}} \\end{figure}  In this paper, I revisit the problem of constraining the shape and orientation of the main body of NGC 3379, using the newest stellar kinematic data (\\markcite{SS99}SS99, Gebhardt \\& Richstone \\markcite{GeR00}2000). The approach follows that of \\markcite{Sta94}S94, as modified by Statler, Dejonghe, \\& Smecker-Hane \\markcite{SDS99}(1999, hereafter SDS). The data are of sufficiently high quality to attempt to recover not just an average shape, but the shape {\\em profile\\/} over a full decade in radius. The presentation is arranged as follows: Section 2 describes the photometric and kinematic data used in the modeling. Section 3 gives an overview of the method, and details where the present treatment differs from previous applications. Sections 4 and 5 describe the results for the estimates of the intrinsic shape profile and the orientation, respectively. Section 6 discusses the implications of the results for the structure and evolution of NGC 3379's nuclear regions, and Section 7 restates the major conclusions. ",
        "conclusions": "}  Dynamical models have been fit to the stellar kinematics and surface photometry of NGC 3379, in order to constrain the intrinsic shape profile and orientation of the main body of the galaxy. The Bayesian ``maximal ignorance'' shape estimate, which reflects an unweighted average over the dynamical parameter space, indicates that the galaxy is most probably axisymmetric and oblate in its inner parts, with a slight but significant outward triaxiality gradient. The formal $1\\sigma$ limits on triaxiality $T$ are $T<0.13$ at $R=6\\farcs 6$ ($0.33 \\kpc$) and $T=0.08\\pm 0.07$ at $R=69\\arcsec$ ($3.5\\kpc$). The most probable short-to-long axis ratio of the luminosity distribution is $c_L=0.79 {+0.05 \\atop -0.1}$ for $R<16\\arcsec$ ($0.82\\kpc$), flattening beyond this point and reaching $c_L=0.66 {+0.07 \\atop -0.08}$ at $R=38\\arcsec$ ($1.9\\kpc$). The results are similar, with slightly steeper gradients in $T$ and $c_L$, if the galaxy is assumed {\\it a priori\\/} to rotate about its short axis. While the estimates for $c_L$ are robust, those for $T$ are dependent on the degree to which the internal rotation field is disklike or spheroid-like. There are regions in the dynamical parameter space where the preferred shapes are prolate-triaxial. Thus the triaxiality estimate should not be considered robust until the dynamical prior can be better constrained on physical grounds.  The inclination of the short axis to the line of sight is not well determined, because the triaxiality is not well determined. Inclinations between $30\\arcdeg$ and $50\\arcdeg$ are preferred. However, highly triaxial models in high inclination are also allowed, which can affect central black hole mass estimates (Magorrian et al. \\markcite{Mag98}1998, \\markcite{Geb00a}G00) by factors of a few. The position angle of the projected short axis on the sky {\\it is\\/} well constrained, and lies within $10\\arcdeg$ of the photometric minor axis at better than $3\\sigma$ confidence. Similar, though weaker, constraints can be placed on the P.A.\\ and inclination of the long axis. These constraints rule out the possibility that the nuclear dust ring at $R\\approx 1\\farcs 5$ lies in one of the planes normal to either axis. Stable equilibria for planar gas disks in triaxial potentials exist only in these planes. Alternatively, if $T\\equiv 0$ and the short axis inclination is $<43\\arcdeg$ ($3\\sigma$ limit), the dust could be in a neutrally stable equilibrium in a polar ring. However, even in this configuration the ring would be torqued by dynamical friction against the rotating stellar population, which would act toward aligning the ring with the stellar rotation axis. The ring is clearly a {\\em decoupled nuclear component\\/} whose orientation is not linked to the main body of the galaxy.  The dust ring may be connected with the ionized gas that extends out to $R\\approx 8\\arcsec$ (Macchetto et al. 1996, \\markcite{Pas00}P00). The projected gas rotation axis is near the minor axis of the ring, though the H$\\alpha +$[\\ion{N}{2}] isophotes are very round. It is possible that the gas and dust are both part of a strongly warped disk. However, the evolution and expected lifetime of such a structure are problematic. If the warp is caused by differential precession, one should expect the entire structure to wind up on itself in $\\sim10^{7}$ years. The decoupling with the stellar component would suggest that the gas has an external origin. But no trace of recent interaction or accretion is seen at larger radii, where it would presumably be much longer lived. At very large radii ($\\gtrsim 10\\arcmin$), NGC 3379 and NGC 3384 are collectively surrounded by a ring of \\ion{H}{1} (Schneider \\markcite{Sch85}1985) whose connection with these and neighboring galaxies is far from clear (Schneider \\markcite{Sch89}1989). Whether the intergalactic ring could be a source of material for occasional accretion events, and how this material might evolve dynamically as it accreted into the nuclear regions, are topics for future work."
    },
    "0010/astro-ph0010468_arXiv.txt": {
        "abstract": "The formation of the first stars and quasars marks the transformation of the universe from its smooth initial state to its clumpy current state. In popular cosmological models, the first sources of light began to form at a redshift $z=30$ and reionized most of the hydrogen in the universe by $z=7$. Current observations are at the threshold of probing the hydrogen reionization epoch. The study of high-redshift sources is likely to attract major attention in observational and theoretical cosmology over the next decade. ",
        "introduction": "\\label{sec1}  The detection of cosmic microwave background (CMB) anisotropies (Bennett et al.\\  1996; de Bernardis et al.\\  2000; Hanany et al.\\  2000) confirmed the notion that the present large-scale structure in the universe originated from small-amplitude density fluctuations at early times. Due to the natural instability of gravity, regions that were denser than average collapsed and formed bound objects, first on small spatial scales and later on larger and larger scales. The present-day abundance of bound objects, such as galaxies and X-ray clusters, can be explained based on an appropriate extrapolation of the detected anisotropies to smaller scales. Existing observations with the {\\it Hubble Space Telescope}\\, (e.g., Steidel et al.\\ 1996; Madau et al.\\ 1996; Chen et al.\\ 1999; Clements et al.\\ 1999) and ground-based telescopes (Lowenthal et al.\\ 1997; Dey et al.\\ 1999; Hu et al.\\ 1998, 1999; Spinrad et al.\\ 1999; Steidel et al.\\ 1999), have constrained the evolution of galaxies and their stellar content at $z\\la 6$.  However, in the bottom-up hierarchy of the popular Cold Dark Matter (CDM) cosmologies, galaxies were assembled out of building blocks of smaller mass. The elementary building blocks, i.e., the first gaseous objects to form, acquired a total mass of order the Jeans mass ($\\sim 10^4 M_\\odot$), below which gas pressure opposed gravity and prevented collapse (Couchman \\& Rees 1986; Haiman \\& Loeb 1997; Ostriker \\& Gnedin 1996). In variants of the standard CDM model, these basic building blocks first formed at $z\\sim 15$--$30$.  An important qualitative outcome of the microwave anisotropy data is the confirmation that the universe started out simple. It was by and large homogeneous and isotropic with small fluctuations that can be described by linear perturbation analysis. The current universe is clumpy and complicated.  Hence, the arrow of time in cosmic history also describes the progression from simplicity to complexity (see Figure~\\ref{fig1a}). While the conditions in the early universe can be summarized on a single sheet of paper, the mere description of the physical and biological structures found in the present-day universe cannot be captured by thousands of books in our libraries.  The formation of the first bound objects marks the central milestone in the transition from simplicity to complexity.  Pedagogically, it would seem only natural to attempt to understand this epoch before we try to explain the present-day universe. Historically, however, most of the astronomical literature focused on the local universe and has only been shifting recently to the early universe. This violation of the pedagogical rule was forced upon us by the limited state of our technology; observation of earlier cosmic times requires detection of distant sources, which is feasible only with large telescopes and highly-sensitive instrumentation.  \\begin{figure}[htbp] \\special{psfile=figI1.ps hoffset=0 voffset=-500 hscale=65 vscale=65} \\vspace{6.2in} \\caption{Milestones in the evolution of the universe from simplicity to complexity. The ``end of the dark ages'' bridges between the recombination epoch probed by microwave anisotropy experiments ($z\\sim 10^3$) and the horizon of current observations ($z\\sim 5$--6).} \\label{fig1a} \\end{figure}  For these reasons, advances in technology are likely to make the high redshift universe an important frontier of cosmology over the coming decade. This effort will involve large (30 meter) ground-based telescopes and will culminate in the launch of the successor to the {\\it Hubble Space Telescope},\\, called {\\it Next Generation Space Telescope} (\\NGST\\,). Figure~\\ref{fig1b} shows an artist's illustration of this telescope which is currently planned for launch in 2009. \\NGST\\, will image the first sources of light that formed in the universe. With its exceptional sub-nJy ($1~{\\rm nJy}=10^{-32}{\\rm erg~cm^{-2}~s^{-1}~Hz^{-1}}$) sensitivity in the 1--3.5$\\mu$m infrared regime, \\NGST\\, is ideally suited for probing optical-UV emission from sources at redshifts $\\ga 10$, just when popular Cold Dark Matter models for structure formation predict the first baryonic objects to have collapsed.  \\begin{figure}[htbp] \\special{psfile=figI2jpg.ps hoffset=25 voffset=-490 hscale=65 vscale=65} \\vspace{5.5in} \\caption{Artist's illustration of one of the current designs (GSFC) of the Next Generation Space Telescope. More details about the telescope can be found at http://ngst.gsfc.nasa.gov/} \\label{fig1b} \\end{figure}  The study of the formation of the first generation of sources at early cosmic times (high redshifts) holds the key to constraining the power-spectrum of density fluctuations on small scales. Previous research in cosmology has been dominated by studies of {\\it Large Scale Structure} (LSS); future studies are likely to focus on {\\it Small Scale Structure} (SSS).  The first sources are a direct consequence of the growth of linear density fluctuations. As such, they emerge from a well-defined set of initial conditions and the physics of their formation can be followed precisely by computer simulation. The cosmic initial conditions for the formation of the first generation of stars are much simpler than those responsible for star formation in the Galactic interstellar medium at present. The cosmic conditions are fully specified by the primordial power spectrum of Gaussian density fluctuations, the mean density of dark matter, the initial temperature and density of the cosmic gas, and the primordial composition according to Big-Bang nucleosynthesis. The chemistry is much simpler in the absence of metals and the gas dynamics is much simpler in the absence of both dynamically-significant magnetic fields and feedback from luminous objects.  The initial mass function of the first stars and black holes is therefore determined by a simple set of initial conditions (although subsequent generations of stars are affected by feedback from photoionization heating and metal enrichment).  While the early evolution of the seed density fluctuations can be fully described analytically, the collapse and fragmentation of nonlinear structure must be simulated numerically.  The first baryonic objects connect the simple initial state of the universe to its complex current state, and their study with hydrodynamic simulations (e.g., Abel et al.\\ 1998a, Abel, Bryan, \\& Norman 2000; Bromm, Coppi, \\& Larson 1999) and with future telescopes such as \\NGST\\, offers the key to advancing our knowledge on the formation physics of stars and massive black holes.  The {\\it first light} from stars and quasars ended the ``dark ages'' \\footnote{The use of this term in the cosmological context was coined by Sir Martin Rees.} of the universe and initiated a ``renaissance of enlightenment'' in the otherwise fading glow of the microwave background (see Figure~\\ref{fig1a}).  It is easy to see why the mere conversion of trace amounts of gas into stars or black holes at this early epoch could have had a dramatic effect on the ionization state and temperature of the rest of the gas in the universe. Nuclear fusion releases $\\sim 7\\times 10^6$ eV per hydrogen atom, and thin-disk accretion onto a Schwarzschild black hole releases ten times more energy; however, the ionization of hydrogen requires only 13.6 eV. It is therefore sufficient to convert a small fraction, $\\sim 10^{-5}$ of the total baryonic mass into stars or black holes in order to ionize the rest of the universe. (The actual required fraction is higher by at least an order of magnitude [Bromm, Kudritzky, \\& Loeb 2000] because only some of the emitted photons are above the ionization threshold of 13.6 eV and because each hydrogen atom recombines more than once at redshifts $z\\ga 7$). Recent calculations of structure formation in popular CDM cosmologies imply that the universe was ionized at $z\\sim 7$--12 (Haiman \\& Loeb 1998, 1999b,c; Gnedin \\& Ostriker 1997; Chiu \\& Ostriker 2000; Gnedin 2000a), and has remained ionized ever since. Current observations are at the threshold of probing this epoch of reionization, given the fact that galaxies and quasars at redshifts $\\sim 6$ are being discovered (Fan et al.\\ 2000; Stern et al.\\ 2000). One of these sources is a bright quasar at $z=5.8$ whose spectrum is shown in Figure~\\ref{fig1c}. The plot indicates that there is transmitted flux shortward of the Ly$\\alpha$ wavelength at the quasar redshift.  The optical depth at these wavelengths of the uniform cosmic gas in the intergalactic medium is however (Gunn \\& Peterson 1965), \\begin{equation} \\tau_{s}={\\pi e^2 f_\\alpha \\lambda_\\alpha n_{\\HI}(z_s) \\over m_e cH(z_s)} \\approx 6.45\\times 10^5 x_{\\HI} \\left({\\Omega_bh\\over 0.03}\\right)\\left({\\Omega_m\\over 0.3}\\right)^{-1/2} \\left({1+z_s\\over 10}\\right)^{3/2} \\label{G-P} \\end{equation} where $H\\approx 100h~{\\rm km~s^{-1}~Mpc^{-1}}\\Omega_m^{1/2}(1+z_s)^{3/2}$ is the Hubble parameter at the source redshift $z_s$, $f_\\alpha=0.4162$ and $\\lambda_\\alpha=1216$\\AA\\, are the oscillator strength and the wavelength of the Ly$\\alpha$ transition; $n_{\\HI}(z_s)$ is the neutral hydrogen density at the source redshift (assuming primordial abundances); $\\Omega_m$ and $\\Omega_b$ are the present-day density parameters of all matter and of baryons, respectively; and $x_{\\HI}$ is the average fraction of neutral hydrogen. In the second equality we have implicitly considered high redshifts (see equations~(\\ref{highz1}) and (\\ref{highz2}) in \\S \\ref{sec2.1}). Modeling of the transmitted flux (Fan et al.\\ 2000) implies $\\tau_s<0.5$ or $x_{\\HI}\\la 10^{-6}$, i.e., the low-density gas throughout the universe is fully ionized at $z=5.8$!  One of the important challenges for future observations will be to identify {\\it when and how the intergalactic medium was ionized}. Theoretical calculations (see \\S \\ref{sec6.3.1}) imply that such observations are just around the corner.  \\begin{figure}[htbp] \\special{psfile=figI3.ps hoffset=20 voffset=-400 hscale=72 vscale=72} \\vspace{6.2in} \\caption{Optical spectrum of the highest-redshift known quasar at $z=5.8$, discovered by the Sloan Digital Sky Survey (Fan et al.\\  2000). } \\label{fig1c} \\end{figure}  Figure~\\ref{fig1d} shows schematically the various stages in a theoretical scenario for the history of hydrogen reionization in the intergalactic medium. The first gaseous clouds collapse at redshifts $\\sim 20$--$30$ and fragment into stars due to molecular hydrogen (H$_2$) cooling. However, H$_2$ is fragile and can be easily dissociated by a small flux of UV radiation.  Hence the bulk of the radiation that ionized the universe is emitted from galaxies with a virial temperature $\\ga 10^4$ K, where atomic cooling is effective and allows the gas to fragment (see the end of \\S \\ref{sec3.3} for an alternative scenario).  \\begin{figure}[htbp] \\special{psfile=figI4new.ps hoffset=-40 voffset=-500 hscale=80 vscale=80} \\vspace{6.2in} \\caption{Stages in the reionization of hydrogen in the intergalactic medium. } \\label{fig1d} \\end{figure}  Since recent observations confine the standard set of cosmological parameters to a relatively narrow range, we assume a $\\Lambda$CDM cosmology with a particular standard set of parameters in the quantitative results in this review. For the contributions to the energy density, we assume ratios relative to the critical density of $\\Omm=0.3$, $\\Oml=0.7$, and $\\Omega_b=0.045$, for matter, vacuum (cosmological constant), and baryons, respectively. We also assume a Hubble constant $H_0=100\\,h\\mbox{ km s}^{-1}\\mbox{Mpc}^{-1}$ with $h=0.7$, and a primordial scale invariant ($n=1$) power spectrum with $\\sigma_8=0.9$, where $\\sigma_8$ is the root-mean-square amplitude of mass fluctuations in spheres of radius $8\\ h^{-1}$ Mpc. These parameter values are based primarily on the following observational results: CMB temperature anisotropy measurements on large scales (Bennett et al.\\ 1996) and on the scale of $\\sim 1^\\circ$ (Lange et al.\\ 2000; Balbi et al.\\ 2000); the abundance of galaxy clusters locally (Viana \\& Liddle 1999; Pen 1998; Eke, Cole, \\& Frenk 1996) and as a function of redshift (Bahcall \\& Fan 1998; Eke, Cole, Frenk, \\& Henry 1998); the baryon density inferred from big bang nucleosynthesis (see the review by Tytler et al.\\ 2000); distance measurements used to derive the Hubble constant (Mould et al.\\ 2000; Jha et al.\\ 1999; Tonry et al.\\ 1997; but see Theureau et al.\\ 1997; Parodi et al.\\ 2000); and indications of cosmic acceleration from distances based on type Ia supernovae (Perlmutter et al.\\ 1999; Riess et al.\\ 1998).  This review summarizes recent theoretical advances in understanding the physics of the first generation of cosmic structures. Although the literature on this subject extends all the way back to the sixties (Saslaw \\& Zipoy 1967, Peebles \\& Dicke 1968, Hirasawa 1969, Matsuda et al.\\ 1969, Hutchins 1976, Silk 1983, Palla et al.\\ 1983, Lepp \\& Shull 1984, Couchman 1985, Couchman \\& Rees 1986, Lahav 1986), this review focuses on the progress made over the past decade in the modern context of CDM cosmologies. ",
        "conclusions": ""
    },
    "0010/astro-ph0010374_arXiv.txt": {
        "abstract": "The pulsar velocity distribution provides information about the binary history of pulsar progenitors as well as asymmetries of the supernova events in which pulsars are born.  Studies of local pulsars present a biased view of this distribution, since they preferentially select low velocity pulsars that have remained near their birthplaces in the Galactic plane. Using the VLA, we have studied the proper motions of a large sample of distant pulsars.  These pulsars are generally faint, and the expected proper motions are small.  In this paper, we describe the data analysis techniques that we have developed to allow precise astrometric measurements of faint sources with the VLA.  These techniques include ``gating'' the VLA correlator to increase the signal-to-noise ratio of the pulsar by gathering data only during the pulse.  Wide-field imaging techniques, including multiband imaging to account for bandwidth smearing, were used to allow identification of multiple in-beam background sources for astrometric calibration.  We present the analysis of three pulsars, and demonstrate that astrometric accuracy of about ten milliarcseconds can be obtained for individual sources with our technique, allowing measurement of proper motions with errors of only a few milliarcseconds per year over our seven year baseline. ",
        "introduction": "Interferometric proper motions have always been hindered by the low flux densities of many pulsars.  This restriction results in a bias in which only the closest  pulsars are observed.  Although these pulsars move quickly on angular scales, their absolute motion must be small for them to still be near the plane of the Galaxy. The fastest pulsars, however, have traveled the greatest distances and may be more than a kpc above the Galactic plane. Previous interferometric proper motion projects have been unable to include these distant pulsars due to their low flux densities.  The study by Fomalont et al., begun in 1984, was limited to pulsars with $S>2$ mJy so that an adequate signal to noise could be achieved \\citep{fom92}.  Few faint pulsars have proper motion measurements as a result of these selection effects.  This study of proper motions of distant, high $z$ pulsars using the Very Large Array (VLA) of the National Radio Astronomy Observatory\\footnote{The National Radio Astronomy Observatory is a facility of the National Science Foundation operated under cooperative agreement by Associated Universities, Inc.} began in 1992.  We have chosen a group of 28 pulsars, 15 of which have $|z|>$ 400 pc, $d\\le 4000$ pc, and $|b| < 30^\\circ$.  This selection criteria minimizes the uncertainty in the determination of the proper motion in the $z$ direction which enables the study of proper motions of pulsars relative to the Galactic plane.  Upon completion, this project will double the number of proper motions measured for distant, high $z$ pulsars.  The flux density restriction is overcome by gating the VLA.  In order to have the pulsar in the same field as stationary sources, we must create wide-field images with position accuracies of a few mas.  This requirement forces us to account for small effects such as annual aberration and Lorentz contraction over the fields.  This paper focuses on the method of data reduction and error analysis used to calculate the proper motions.  Three pulsars (B0919+06, B1237+25 and B1937--26) are used as ``case studies'' demonstrating the accuracy of the method over a range of declinations and flux densities. Proper motions for all 28 pulsars will be presented by Brisken et al. in Paper II. ",
        "conclusions": "We have developed a new technique to calculate proper motions of weak pulsars at the VLA.  By gating the VLA in one polarization, distant, high $z$ pulsars with flux densities as low as $\\sim$1 mJy can be detected with a high signal-to-noise ratio. Previous work by \\citet{lyn82} and \\citet{har93} also utilized the advantages of in-beam reference sources and gating.  These observations were limited by a small number of baselines (one baseline in \\citet{lyn82} and two baselines in \\citet{har93}) and the positions determined using fringe rate mapping rather than imaging could not correct for extended structure in reference sources.  In addition, their data was degraded by the ionosphere which is worse at 408 MHz than in our observations at 1452.4 MHz.  The 351 baselines and larger total collecting area of the VLA enable detection of much weaker sources and improved imaging.  The addition of wide field imaging and a detailed understanding of systematic effects at the VLA also makes this technique more useful. Wide-field imaging enables sources located up to $\\sim20'$ from the field center to be used as reference sources.  By imaging the reference sources, we can exclude sources which show extended structure.  These sources are very important in the calculation of plate solutions as well as the pulsars' motions.  Proper motions of the pulsars are fit simultaneously to all reference sources in the field.  Both B0919+06 and B1237+25 agree well with previously published proper motions confirming the accuracy of this technique.  The measurement of the proper motion of B1937--26 shows the real success of the method. Ungated, this pulsar has a flux density of only $\\sim$1 mJy at 20 cm. This weak flux density excluded it from all previous proper motion projects.  By gating the VLA, proper motions of faint pulsars can now be obtained with accuracies of a few mas yr$^{-1}$ in less than ten years. These accuracies are comparable to those obtained with VLBA observations.  Although the larger synthesized beam of the VLA reduces positional accuracies, the VLA has the advantage of more straightforward data reduction and more thoroughly understood systematic effects.  The presence of reference sources in the field of view of the VLA images also makes this a favorable technique since VLBA observations generally require out-of-beam reference sources.  Therefore, proper motions of faint pulsars can now be obtained at the VLA in just a few years.  Proper motions for all 28 pulsars in this study will be reported in Paper II by Brisken et al."
    },
    "0010/astro-ph0010138_arXiv.txt": {
        "abstract": " ",
        "introduction": " ",
        "conclusions": "I have emphasized in this paper a number of puzzling astronomical observations of aromatics in the ISM, and related molecules, where understanding of these problems would greatly benefit from new experiments in laboratory astrochemistry and progress in chemical theory. The study of the ISM is an area where the interaction of astronomy, physics, and chemistry has proven particular fruitful in reaching new conclusions about the universe that surrounds us, and I strongly encourage astrochemistry groups to tackle these experimental and theoretical problems as part of this on-going interaction."
    },
    "0010/astro-ph0010412_arXiv.txt": {
        "abstract": "The spectral distortion of the cosmic background radiation produced by the inverse Compton scattering on hot electrons in clusters of galaxies (thermal Sunyaev--Zel'dovich effect) is calculated for arbitrary optical depth and electron temperature. The distortion is found by a numerical solution of the exact Boltzmann equation for the photon distribution function. In the limit of small optical depth and low electron temperature our results confirm the previous analyses. In the opposite limits, our method is the only one that permits to make accurate calculations. ",
        "introduction": "The frequency spectrum of the Cosmic Microwave Background Radiation (CMBR) is known to have the equilibrium black body form (in natural units in which $\\hbar=c=k_B=1$) \\beq f_0 (p_\\gamma) = \\left[ \\exp (p_\\gamma /T_\\gamma) -1\\right]^{-1} \\, , \\label{f0} \\eeq with the temperature $T_\\gamma = 2.725 \\pm 0.002$ K \\citep{cobelast}. No deviation from this perfect Planck distribution is observed with the accuracy of $10^{-4}$.  These data present a strong evidence in support of Big-Bang cosmology. It is established that the cosmic plasma in the early universe was in a thermal equilibrium state and the spectrum remained practically undisturbed to the present epoch. Nevertheless, small deviations from the perfect Planck spectrum are possible, and they could provide interesting information about physical processes in the early universe that might take place at relatively small red-shifts, $z \\leq 10^7$. At higher red-shifts all distortions of the thermal equilibrium spectrum would be efficiently smoothed down by the Compton scattering, $\\gamma +e \\lrar \\gamma +e$ and by the inelastic photon producing reactions, double Compton, $\\gamma + e \\lrar 2\\gamma +e$ or Bremsstrahlung, $e + A \\lrar e + A + \\gamma$. The first elastic process restores kinetic equilibrium, i.e. it forces the photon distribution function, $f_\\gamma$, to take the Bose-Einstein form with a possible non-zero chemical potential, $\\mu$, while the other two reactions push $\\mu$ down to zero.  There are several possible sources and mechanisms that could give rise to spectral distortions of the CMBR both in the early and the present day universe. In the early universe there could be electromagnetic decays of long-lived particles with life-time larger than $\\sim 100$ sec, in particular, $\\nu_H \\rar \\nu_L + \\gamma$. A study of a possible distortion of the CMBR spectrum permits to obtain strong bounds on the probability of such decays (for a review, see \\citet{cmb-limits}). In the present-day universe a very interesting spectral distortion can be induced by the scattering of cosmic background radiation on hot electrons in galactic clusters, known as the Sunyaev-Zel'dovich (SZ) effect~\\citep{zs1,zs2,zs3,zs4}. Observations of such a distortion, combined with those of thermal X-ray emission of the cluster gas, could help to extract important astrophysical information, in particular to measure the Hubble constant, or to study the evolution of clusters (for reviews see \\citet{review-zs1,review-zs2,revReph, Birkinshaw99}, and the updates in \\citet{3Kb} and \\citet{3Kr}).  Compton scattering conserves the number of photons, which implies that the SZ effect produces a systematic shift of photons from the low-energy part to the Wien side of the Planckian CMBR spectrum. To calculate the corresponding distortion of the spectral distribution one should use the Boltzmann kinetic equation for the distribution function of photons, $f_\\gamma$, \\beq {d f_\\gamma \\over dt} = I_{coll} +S \\, , \\label{kineq} \\eeq with properly taken collision integral $I_{coll}$ and a possible source term, $S$.  In the early universe case one has to solve the system of such two coupled equations for the photon and electron/positron distributions. In the lowest order in the fine-structure constant, $\\alpha \\simeq 1/137$, only Compton and Coulomb scattering need to be taken into account.  They have cross-sections of the order $\\alpha^2$, while inelastic reactions contribute at most to the order $\\alpha^3$. Because of technical difficulties this problem was treated only approximately especially for relativistic electrons. In the non-relativistic case the problem can be reduced to the partial differential equation describing diffusion in photon momentum space. For the particular case of Compton scattering such a reduction was done in 1957 by Kompaneets \\citep{kompaneets57} and for inelastic processes (double Compton scattering and Bremsstrahlung) by Lightman \\citep{lightman81}. Solutions to these equations in cosmological situations were analyzed in \\citet{bernstein90} and \\citet{hu93}.  In the case of scattering of the CMB radiation on hot electrons in galactic clusters (thermal SZ effect) the Kompaneets equation has the form \\beq {\\partial f_\\gamma (X,t) \\over \\partial t} = n_e \\sigma_T\\, {T_e \\over m}\\, {\\partial \\over X^2 \\partial X} \\left[ X^4 \\left( {\\partial \\fg \\over \\partial X} +\\frac{T_\\gamma}{T_e}\\fg(1 +\\fg) \\right)\\right] \\, , \\label{komp} \\eeq where $T_e$ and $T_\\gamma$ are the electron and photon temperatures, $m$ is the electron mass, \\be X=\\frac{p_\\gamma}{T_\\gamma} \\label{defX} \\ee is the dimensionless photon momentum, $n_e$ is the electron number density, and $\\sigma_T = 8\\pi\\alpha^2 /3m^2 = 6.65\\times 10^{-25}\\, {\\rm cm}^2$ is the Thomson cross-section. However, the electrons in the clusters are hot, with temperatures that can be larger than $15$ keV, which means that the energy change of Compton scattered photons is not small enough to be accurately described by the Kompaneets equation.  There are basically two different approaches in the literature to extend the validity of this equation. In ref.~\\citet{Rephaeli95} the relativistic form of the Maxwell velocity distribution of electrons is used and the frequency redistribution through Compton scattering is calculated (see also \\citet{Wright79,Yankovitch97,Sazonov98,Molnar99}); this method is also known as the Radiative Transfer Approach. On the other hand, one can obtain a relativistic generalization of the Kompaneets equation, by expanding in series of the parameter $\\theta_e \\equiv T_e/m$.  Low orders of this expansion were considered in the papers \\citet{Stebbins97} and \\citet{Challinor98}, while \\citet{ItohI} took into account relativistic corrections up to ${\\cal O}(\\theta_e^5)$. These two different methods are essentially consistent.  It has, however, been pointed out \\citep{Challinor98} that these results are not guaranteed to be accurate because the series approximation to the solution of the Boltzmann kinetic equation involves expansions in parameters which are not  small. In particular it was recognized that the convergence is slow and possibly even asymptotic (for instance, to calculate the crossover frequency, where the thermal distortion vanishes, it is better to use a linear approximation in $\\theta_e$ than the expansion up to ${\\cal O} (\\theta_e^3)$ or ${\\cal O} (\\theta_e^5)$). In view of that it is desirable to solve the original kinetic equation (\\ref{kineq}) directly without any specific approximation. This problem was partly addressed in \\citet{ItohI,ItohIV} where the collision integral was numerically calculated for the unperturbed photon distribution eq.~(\\ref{f0}).  This integral determines the first time derivative of the distribution function $f_\\gamma (X, 0)$ and in the limit of a small optical depth, \\be \\tau = \\int dl~n_e \\sigma_T \\ll 1 \\, , \\label{tau} \\ee (the integral is taken along the line of sight through the cluster) gives the solution, $\\fg (X,\\tau) \\approx f_0 + \\tau \\,I_{coll} [\\fg (X,0)]$.  In a subsequent paper \\citep{ItohV}, the second order corrections in $\\tau$ were calculated as a series in $\\theta_e$. The results show a good agreement with solutions of the generalized Kompaneets equation in the limit of small optical depth and low electron temperatures, $\\theta_e =T_e /m \\la 0.03$ (i.e. $T_e \\la 15$ keV).  In this paper we present an accurate numerical solution of the exact Boltzmann kinetic equation for arbitrary optical depth and electron temperature. In the limit of small $\\tau$ and $\\theta_e$ our results confirm those of the previous papers, and in particular of \\citet{ItohI,ItohV}.  However, the method presented here permits to solve the equation precisely for an arbitrary optical depth and any isotropic electron distribution function. We use essentially the same method as the one we have developed for the calculations of the spectral distortion of light \\citep{dolgov97,dolgov99a} or heavy \\citep{dolgov98,dolgov99b} neutrinos at Big Bang Nucleosynthesis.  An essential point of the calculations is an analytical reduction of the exact collision integral down to two dimensions \\citep{TkachevI,TkachevII}. In the case of Compton scattering the problem is much more complicated because the squared amplitude of this process is not a simple polynomial function of particle momenta as was the case for weak interaction in the low energy limit.  A reduction of the collision integral down to one dimension in the direct reaction term and down to two dimensions in the inverse reaction term was done in \\cite{Poutanen96} for the case of Boltzmann statistics (see also the review \\cite{Revpou} for a discussion and relevant references). The method of these works is somewhat different from ours because we specially use the procedure of the integration suitable for the quantum statistics case when the direct reaction term (as well as the inverse one) contains the product of the distribution functions not only in the initial state but also in the final state, $f_1 f_2 (1+ f_3)(1-f_4)$ (see eq.~\\ref{Ff}), that makes the reduction down to one dimension impossible in principle if all the functions are considered as unknown.  The paper is organized as follows. In the next rather technical section we present the reduction of the 9-dimensional collision integral down to a 2-dimensional one. In section 3 the numerical solution of the integro-differential Boltzmann equation is described. In section 4 our results are presented and discussed. Finally in section 5 we give the conclusions. ",
        "conclusions": "We have numerically calculated the spectral distortion of the cosmic background radiation produced by Compton scattering on hot electrons in galaxy clusters without any simplifying assumptions.  We have analytically reduced the collision term in the Boltzmann equation to two dimensional integrals, and accurately solved the kinetic equation. Our analytical expression for the collision integral is exact, and can be used for any temperatures, $T_\\gamma$, $T_e$, and optical depth, $\\tau$. The results are in good agreement with the previous works for small temperature and optical depth, but considerably different for large $\\tau$. Our method can also be applied in the case of an isotropic electron distribution function, not necessarily in equilibrium."
    },
    "0010/astro-ph0010554_arXiv.txt": {
        "abstract": "The accurate determination of the elemental composition of cosmic rays at high energies is expected to provide crucial clues on the origin of these particles. Previous direct measurements of composition have been limited by experiment collecting power, resulting in marginal statistics above $10^{14}$ eV, precisely the region where the ``knee'' of the cosmic-ray energy spectrum is starting to develop. In contrast, indirect measurements using extensive air showers can produce sufficient statistics in this region but generate elemental measurements which have relatively large uncertainties. Here we discuss a technique which has become possible through the use of modern ground-based \\cer imaging detectors. We combine a measurement of the \\cer light produced by the incoming cosmic-ray nucleus in the upper atmosphere with an estimate of the total nucleus energy produced by the extensive air shower initiated when the particle interacts deeper in the atmosphere. The emission regions prior to and after the first hadronic interaction can be separated by an imaging \\cer system with sufficient angular and temporal resolution.  Monte Carlo simulations indicate an expected charge resolution of $\\Delta Z/Z <5\\%$ for incident iron nuclei in the region of the ``knee'' of the cosmic-ray energy spectrum. This technique also has the intriguing possibility to unambiguously discover nuclei heavier than iron at energies above 10$^{14}$ eV. The identification and rejection of background produced by charged particles in ground based gamma-ray telescopes is also discussed. ",
        "introduction": "The origin of cosmic rays remains a central unresolved question in astrophysics. Nearly 90 years after the first observations, this population of charged particles remains an enigma; the heart of which is the huge dynamic range of fluxes and energies over which they have been observed. In the theoretical arena, a relatively recent paradigm has emerged, which involves the acceleration of cosmic rays via diffusive shock processes in supernovae remnants (SNR). This idea has been fueled both by a viable physical model of the acceleration process \\cite{Bell:1978a,Bell:1978b,Blandford:1978,Krymsky:1977} and a simple energetics argument in which galactic supernovae are the only galactic candidate with sufficient energy output to supply the staggering amount of power needed to sustain the cosmic-ray population (see \\eg, \\cite{Schlickeiser:1994}).  A key issue with the SNR idea is that supernova diffusive shock acceleration can only produce particles up to some maximum energy, limited either by the lifetime of the strong shock or by the particles becoming so energetic they can no longer be confined in the acceleration region \\cite{Lagage:1983}. Estimates of this upper energy limit vary, with typical values in the region of $10^{14}$ eV.  However, the observed flux of cosmic rays extends more or less continuously for another five orders of magnitude beyond this; no plausible extrapolation of the standard parameters of SNR can generate cosmic rays of such energies.  Another aspect of the cosmic ray riddle is the existence of an observational ``knee'' at $\\sim 10^{15}$ eV in the cosmic ray energy spectrum. The coincidence of this ``knee'' with the theoretical energy limit of SNR diffusive shock acceleration is intriguing and has actually served as evidence to support the theory. However the ``knee'' represents only a small change in the spectral slope of the overall flux, with the energy dependence changing from $E^{-2.75}$ below the ``knee'' to $E^{-3.0}$ above it.  We are faced with a simple observational fact that the cosmic rays have an essentially continuous spectral slope for nearly 11 orders of magnitude.  In principle, additional mechanisms could provide the flux at high energies.  The power budget for ``post-knee'' cosmic rays is only a fraction that of the total budget, so there is some freedom in selecting models.  However, to result in an energy spectrum as smooth as is observed, these mechanisms would have to generate fluxes which are remarkably, perhaps implausibly, close to that of the SNR mechanism. Another problem exists at the highest energies, around $\\sim 10^{19}$ eV, where an additional spectral break occurs (the ``ankle'').  It has been argued that particles above the ``ankle'' could be extragalactic.  If so, they can be expected to exhibit a characteristic cutoff due to photo-pion production with the cosmic microwave background.  This cutoff has yet to be observed \\cite{Hayashida:2000}.  An overall scheme which credibly addresses and unifies these issues remains elusive.  More reliable and accurate measurements in the ``knee'' region, of abundant cosmic ray  nuclei (1 $\\le Z \\le$ 26)  could drastically revise our current ideas and provide a basis for such a unification. The accurate determination of the composition of cosmic rays has provided some of the key advances in this field at lower energies, where direct measurements are possible with detectors above the atmosphere. For example, the realization that the observed spectral slope of cosmic rays is significantly steeper than that produced in the cosmic ray sources themselves resulted from measurements with sufficient elemental resolution to separate primary source cosmic ray elements from those produced in the interstellar medium at $10^{11}$ eV \\cite{Juliusson:1972}.  In this paper we propose an idea which has become possible through advances in the imaging atmospheric \\cer technique.  It is based on the concept that a detector of fine enough pixelation will be capable of observing \\cer light emitted directly from cosmic ray nuclei prior to their first interaction in the atmosphere.  In general, this light is overwhelmed by the \\cer emission from the subsequent extensive air shower (EAS).  However, with an appropriate detector, and within certain geometric constraints, this ``direct \\v{C}erenkov\", or D\\v{C}, light can be sufficiently well separated from the background of EAS \\cer light to make relatively high-precision measurements of the traditional cosmic ray composition ($1 \\le Z \\le 26$). This technique may also provide new opportunities for the first measurement of higher charge nuclei (Z $\\gg 26$) as well as possible improvements in ground based gamma-ray telescope sensitivity. ",
        "conclusions": "\\label{sec:Conclusion}  We have discussed a new experimental technique which can potentially yield excellent charge resolution measurements ($\\Delta Z/Z<5\\%$ for $Z=26$) for ground-based observations of high energy cosmic rays. The technique relies upon the observation of the direct \\cer light emitted by the primary nucleus before the first nuclear interaction with the Earth's atmosphere. The experimental technique works over an energy range in the TeV-PeV energy range, with the width of the energy window growing like $Z$ for heavy nuclei. A dedicated observatory could be built to observe cosmic rays using his technique, while the next-generation ground-based imaging \\cer telescopes such as VERITAS, HESS, and CANGAROO will begin to approach the sensitivity required to begin initial observations.  The average intensity and fluctuations in the D\\v{C} light yield have been examined using Monte Carlo simulations. The yield contains the expected $Z^2$ dependence with fluctuations that are easily understandable in terms of the radial distance of observation, variations in the depth of first hadronic interaction, and the details of the subsequent fragmentation process.  Indeed, D\\v{C} light measurement appears to provide a method to measure the primary charge independent of any hadronic interaction or fragmentation model. This could potentially solve the most difficult systematic problem in the ground-based studies of TeV/PeV cosmic rays: the difficulty in distinguishing between actual changes in composition and systematic shifts of nuclear interaction models with energy. The high resolution nature of this measurement might allow the detection of changes in the propagation pathlength distribution at high energy. The pathlength distribution at PeV energies is unknown, but has a extremely strong effect on the predicted composition at the knee of the all-particle spectrum \\cite{Swordy:1995icrc}.  The D\\v{C} light measurement technique could provide researchers with a ``tagged particle beam'' of known energy and composition. By examining D\\v{C} light yield at various radial bins, one may be able to extract nucleon-air cross-section information as a function of primary mass and energy, independent of an interaction model. Fluctuations in the D\\v{C} light yield for small interaction depths could provide information concerning the fluctuations in the fragmentation processes of nucleus-air interactions.  The D\\v{C} light measurement technique may provide additional sensitivity to ground-based imaging air \\cer gamma-ray detector for the rejection of the dominant electron background in the $10-500$ GeV energy range, and residual proton background in the $1-10$ TeV energy range. Such an improvement could potentially increase the point source sensitivity of these telescopes by an order of magnitude or more.  The unique properties of D\\v{C} light (versus EAS light) may allow the observation of exotic cosmic ray particles, such as ``strange quark matter'' (see, \\eg, \\cite{Banerjee:1998wk,Banerjee:2000ye}) or magnetic monopoles \\cite{Wick:2000}.  These particles, which are thought to have large effective charges, would provide an extremely strong \\DC signature due to the $Z^2$ dependence of the light yield.  A technique which combines the collection power and logistical advantages of ground-based cosmic ray detectors with the high-precision charge resolution of balloon or satellite-borne experiments would be a valuable asset in the field of high-energy astrophysics. Such a technique, if scientifically and financially viable, has the potential to make great strides in the determination of cosmic ray composition across the ``knee\", and subsequently, foster advances in the theory of cosmic ray origins.  We believe that the development of imaging atmospheric \\cer telescopes has provided the means to target the \\cer light emitted directly from primary nuclei prior to their hadronic interaction in the atmosphere. A direct \\cer experiment, implemented as an array of 10 m reflectors imaged by high-resolution cameras with fast ADC systems may prove to be the next step towards understanding the nearly century-old mystery of cosmic rays.   \\vspace{1ex} {\\bf Acknowledgements} The authors express their appreciation for useful discussions with members of the VERITAS collaboration during the preparation of this paper. DBK acknowledges  computational support from the Utah High Energy Astrophysics Institute."
    },
    "0010/astro-ph0010248_arXiv.txt": {
        "abstract": "On the basis of a new convergence study of high-resolution N-body simulations, my colleagues and I now agree that the Navarro, Frenk, \\& White (1996) density profile $\\rho_{NFW}(r) \\propto r^{-1} (r+r_s)^{-2}$ is a good representation of typical dark matter halos of galactic mass.  Comparing simulations of the same halo with numbers of particles ranging from $\\sim10^3$ to $\\sim10^6$, we have also shown that $r_s$, the radius where the log-slope is -2, can be determined accurately for halos with as few as $\\sim10^3$ particles.  Based on a study of thousands of halos at many redshifts in an Adaptive Refinement Tree (ART) simulation of a cosmological volume in a $\\Lambda$CDM cosmology, we have found that the concentration $\\cvir \\equiv \\Rvir/\\rs$ has a log-normal distribution, with $1\\sigma$ $\\Delta (\\log \\cvir) = 0.18$ at a given mass, corresponding to a scatter in maximum rotation velocities of $\\Delta \\Vmax/\\Vmax = 0.12$. The average concentration declines with redshift at fixed mass as $\\cvir(z) \\propto (1+z)^{-1}$.  This may have important implications for galaxy rotation curves.  Finally, we have found that the velocity function determined from galaxy luminosity functions plus luminosity-velocity relations agrees with the predictions from our $\\Lambda$CDM simulations.  But we also note that the very limited evolution with redshift of the velocity function predicted by $\\Lambda$CDM conflicts with the data that is becoming available on the number density of bright galaxies unless there is significant evolution of the luminosity-velocity relation at $z>1$. ",
        "introduction": "In this talk, I review some of the recent work by my collaborators (especially my former PhD student James Bullock) and me on the distribution of dark matter in galaxy-size halos and its evolution with redshift.  I summarize results from several recent papers, in particular Bullock et al. (1999), Gonzalez et al. (2000), Sigad et al. (2000), Klypin et al. (2000), and Bullock et al. (2000).  Avishai Dekel in his talk summarized related work by our group on the distribution of angular momentum in dark matter halos. ",
        "conclusions": ""
    },
    "0010/astro-ph0010281_arXiv.txt": {
        "abstract": "The galaxy-type CSO (compact symmetric objects) rate from an intermediate large sample PR+CJ1 is found to be consistent with the Sb-Sb merger rate of 4.6\\% in redshift z$<$1. The CSO detection rate seems going down when the sample size is increasing, which could not reach the high merger rate in z$>$1, yet no deep VLBI sample in z$>$1 is available now. ",
        "introduction": "The complete PR+CJ1 sample consists of 200 radio sources with flux $>$0.7Jy at 5 GHz and declination $>$35 degrees. The 8/65 sources in PR and the 15/200 sources in PR+CJ1 sample are identified as CSOs (Polatidis et al. 1999), so the CSO rate is 12\\% in PR sample (flux $>$1.3Jy at 5 GHz), 7.5\\% in PR+CJ1 sample. The CSO rate from PR+CJ (including flat-spectrum samples CJ2 and CJF, with flux $>$0.35 Jy at 5 GHz) is 4.4\\%(18/411) (Peck \\& Taylor 2000). The COINS sample (CSOs observed in the northern sky, based on VLBI continuum surveys of nearly 2000 compact radio sources, i.e. PR+CJ+VCS, VCS:VLBA Calibrator surveys, with flux $>$100 mJy at 5 GHz, see Peck \\& Taylor 2000) gives a CSO rate of 2.1\\%(39/1900). From the JVAS and CLASS flat-spectrum sample (with flux $>$100 mJy at 8.4 GHz) the CSO rate is 0.8\\%(14/1665) (Augusto et al. 1999). The CSO detection rate does appear going down when the sample size is increasing. However, both the uv-coverage and the sensitivity of the VCS are considerably worse than the PR and CJ surveys, it is possible that some CSOs have been missed in the VCS (Peck \\& Taylor 2000). The flat-spectrum samples could also miss some CSOs, since most of CSOs from PR+CJ1 sample show steep spectra (they are usually GPS sources).  \\begin{table*} \\scriptsize \\begin{center} \\end{center} \\vspace{6pt} \\begin{tabular*}{\\textwidth}{@{\\hspace{\\tabcolsep} \\extracolsep{\\fill}}p{5pc}llllll} \\hline\\hline\\\\[-6pt] Name & z & ID & GPS & m$_{r}$ & Comments for the optical hosts\\\\ [4pt]\\hline\\\\[-6pt] 0108+388 PR & 0.669 & G & Yes & 21.8 & very red color r-i=1.2 galaxy with an asymmetric\\\\ &  &  &  &  &  morphology, the nucleus is more prominent in\\\\ &  &  &  &  &  i band, suggesting a nuclear obscuration -[1]\\\\  0404+768 PR & 0.5985 & G &  Y &  20.7 & has weak diffuse structure, possible companion,\\\\ &  &  &  &  & and very red color r-i=1.7 -[1] \\\\  0646+600 CJ1 & 0.455 & Q & Y-[2] & 19.9-[3] & a number of faint galaxies surrounded and a\\\\ &  &  &  &  & possible companion -[3]\\\\  0710+439 PR & 0.518 & G & Y & 20.2 & very irregular morphology with very red (r-i=1.9)\\\\ &  &  &  &  & color, there is evidence of strong interaction and\\\\ &  &  &  &  & presence of large amount of obscuring matter;\\\\ &  &  &  &  & many other resolved objects are present around\\\\ &  &  &  &  & the parent object suggesting it is the dominant\\\\ &  &  &  &  & galaxy in a cluster -[1]\\\\ 1031+567 PR &  0.45 &  G & Y & 20.2  & elliptical-like galaxy showing a possible asymmetric\\\\ &  &  &  &  & morphology, weak resolved objects are also present\\\\ &  &  &  &  & in the field suggest it is the domination in a cluster -[1]\\\\   1225+368 CJ1 & 1.974 & Q & Y-[2] & 21.6-[2]\\\\   1242+410 CJ1 & 0.813 & Q & & 19.6-[4] & very red compact object -[5]\\\\   1311+678 CJ1 &      & EF &   & & steep spectrum with $\\alpha_{(11-6)cm}=-0.73$ -[6]\\\\   1358+624 PR & 0.431 & G &  Y & 19.8 & a boxing morphology showing the obscuration\\\\ &  &  &  &  & by dust, more visible in r band and a resolved\\\\ &  &  &  &  & object is visible at 5\" north, with r-i=0.3 -[1]\\\\  1437+624 CJ1 & 1.090 &  Q &   &   19.0-[7] &   \\\\  1843+356 CJ1 &  0.763 & G &  Y-[1] & 21.7 &       faint galaxy -[1]\\\\  1943+546 CJ1 & 0.263 &  G &  &  17.6-[8] & apparently in a faint cluster, faint objects\\\\ &  &  &  &  & surrounded -[8] and with $\\alpha_{(11-6)cm}=-0.55$ -[6]\\\\  2021+614 PR & 0.2266 & G &  Y-[9] &  17.3 & a low level extended halo around and the presence\\\\ &  &  &  &  & of at least two candidate companions within\\\\ &  &  &  &  & 12\" of the galaxy -[2]\\\\  2342+821 PR & 0.735 & Q & Y-[1] & 20.1-[1] & weak secondary emission with r-i=0.8 -[1]\\\\  2352+495 PR & 0.2383 & G & Y-[9] & 18.5 & an elliptical galaxy is located in a distant cluster, \\\\ &  &  &  &  & nearby faint objects are candidate companions -[2]\\\\ \\hline \\end{tabular*} \\vspace{6pt}  {\\sl{Notes:} Source parameters if not referenced are from Polatidis et al. 1999 and Snellen et al. 1996 (for m$_r$). References: [1] Stanghellini et al. 1993, 1998. [2] O'Dea et al. 1990. [3] Stickel \\& K\\\"uhr 1993a. [4] Carballo et al. 1999. [5] Vigotti et al. 1989. [6] Stickel \\& K\\\"uhr 1994. [7] Aldcroft et al. 1994. [8] Stickel \\& K\\\"uhr 1993b. [9] Snellen et al. 1996} \\end{table*} ",
        "conclusions": "Gravitional interaction and mergers affect the the morphologies and dynamics of galaxies from our Local Group to the limits of the observable universe. Observations of interacting galaxies at low redshifts (z$<$0.2) yield detailed information about many of the processes at work, these processes and the growing evidence indicate that mergers play a major role in the delayed formation of elliptical and early-type disk galaxies both in the field and in clusters (Schweizer 1999). Merger ages from 65 E and S0 galaxies based on UBV colors and a simple two-burst model of star formation spread over most of the age of universe (Schweizer 1999). We measure the Sb-Sb merging (to E+S0, efficiency=0.10) rate from Schweizer \\& Seitzer (1992, hereafter SS) assuming that the epochs of 3, 6, 9, 12 and 15 Gyr ago are equal to the time at redshift z=1,2,3,4,5 correspondingly. The result (i.e.the cumulative rate in per z/3 range) is (\\%) 3.1, 0, 1.5, 4.6, 12.3, 9.2, 15.4, 26.2, 16.9, 7.7, 1.5, 1.5, 0 in redshift from 0 to 4.33 as shown in Fig.1. The Gaussion fitting formula: y=1.93+20.63exp(-2.16(z-2.45)$^2$), the fitting is not good for the data in range z=0-2, we can see that the data slope is less steep than that of in z$>$2.4. We then fit the data in z=0-2 with a power law (1+z)$^{2.28}$ . The evolution of the merger(or pair) rate in z$<$2 has been studied by several authors (see Gottl\\\"ober et al. and references therein), which follows a power law (1+z)$^{3\\pm1}$. For comparing to the CSO rate (the CSO rate is indeed the cumulative rate in their corresponding z range as in Fig.1), we compute the cumulative merger rate in range z=0-1, z=1-2, z=2-3, z=3-4 and draw the dash line as in Fig.1. We find that the galaxy CSO rate of 4.5\\% in PR+CJ1 is close to the cumulative Sb-Sb merger rate of 4.6\\% in z$<$1. The CSO rate 7.5\\% of PR+CJ1 (line 2 in Fig.1) in range z=0-2 , we think, is not true for z=1-2, because only 2 CSOs (quasars with z=1.09 and 1.974) are detected in z=1-2. The CSO rate of 12\\% in PR is much higher than the cumulative merger rate in z$<$1, this could be due to the PR sample is still not a deep sample in z$<$1. It suggests that in the intermediate deep sample PR+CJ1 (flux $>$0.7 Jy), the cumulative galaxy CSO rate is consistent with the cumulative Sb-Sb merger rate in z$<$1. This should be investigated further in more deep and complete sample. On the other hand, we collect the properties of the CSO hosts from literature as in the table above, the CSO hosts seem having the symptoms of merging or merging environment, however, this can not confirm if they are undergoing merging.   In the PR sample, 30/65 sources are identified as quasars, 29/65 are galaxies, and 6/65 are BL Lac sources. The sources in PR+CJ1 are not completely identified. From a large optical galaxy sample, 41\\% of ellipticals, 65\\% of galaxies and 75\\% of peculiars are detected in NVSS with $>$2.5 mJy (Cotton and Condon 1998). The ellipticals and peculiars mostly are due to AGN activities, as well as the spiral galaxies detected in NVSS are usually starburst ones. At flux densities greater than 10 mJy, more than 95\\% of the optical identifications of radio sources are red ellipticals, Seyferts, and quasars, but this fraction drops to 30\\% at 5 mJy; the majority of sources with 200 $\\mu$Jy$<$ S$_{20cm}$$<$5 mJy have optically blue, disk-galaxy counterparts (Richards 1998). So for CSO rate detection, VLBI samples can be as deep as $>$10 mJy without considering of starburst galaxies.   It is proposed that all elliptical galaxies are originated from spiral-spiral galaxy merging (Schweizer 1997). We searched in NVSS for 32 ellipticals of SS sample, 15/32(47\\%) ellipticals show radio emission with $>$2.5 mJy. More than 50\\% ellipticals have not been detected in radio indicates that the radio phase of merger (which form elliptical eventually) is very short comparing with the lifetime of elliptical. For example, the radio lobes of Cygnus A is aging at 0.01 Gyr (Kaiser 2000), while the ages of young ellipticals in SS sample are greater than 0.9 Gyr. Indeed, 2 merger remnants (0.3-0.6 Gyr) in SS sample are point-like radio sources in NVSS. The CSO lifetime is less than 10$^{3-4}$years (Owsianik et al. 1999 and references therein; Xiang et al. 2000). CSOs may be triggered by the merging of nuclei of spirals, since strongly merging pairs we known have not shown jet-like radio emission, and their nuclei have not yet merged. So the CSOs could occur at late merging stage (say at the elliptical-forming stage, in that time, the nuclei of spirals is merging and the arms of spirals are already merged).   Quasar-type CSOs in PR+CJ1 show a large deviation from galaxy-type CSOs in Hubble diagram of Fig.2 . Radio selected QSOs are usually hosted by giant elliptical galaxies (Benn et al. 1998). So they may be the mergers from giant spirals. We think their unusual luminosities and relatively less data points in Fig.2 result the flatter slope than that of galaxy CSOs. The scattering spectrum of QSOs in Fig.2 suggests the QSOs are not good standard candles, they may be more bright virtually with increasing redshift.   It is expected that radio AGNs will live longer period in the MSO phase because MSOs are defined to have broad linear scales (1-15 kpc), we should detect more MSOs in a complete sample e.g. PR+CJ1. However, the MSO detection rate is 2\\% in PR (one quasar) and 5\\% in PR+CJ1 sample (Polatidis et al. 1999). One reason may be that the symmetric two-side jets are much disturbed by the remnant of spiral arms, we assumed that the merging of spiral arms is finished when a CSO is triggered. At this point, most of CSS and some of GPS sources might be actually MSOs.\\\\ Liu Xiang acknowledge P. N. Wilkinson for illuminating discussions, thank Zhang Haiyan for help on the figures.   \\begin{figure}[!h] \\centerline{\\psfig{figure=xiang.fig1.ps,width=11cm} } \\caption{The cumulative CSO rate and the Sb-Sb spiral merging (into E or S0) rate} \\end{figure}   \\begin{figure}[!h] \\centerline{\\psfig{figure=xiang.fig2.eps,width=9cm} } \\caption{Hubble diagram for CSOs in PR+CJ1 sample} \\end{figure}"
    },
    "0010/astro-ph0010039_arXiv.txt": {
        "abstract": "The number density of rich galaxy clusters still provides the most robust way of normalizing the power spectrum of dark matter perturbations on scales relevant to large-scale structure. We revisit this constraint in light of several recent developments: (1) the availability of well-defined samples of local clusters with relatively accurate X-ray temperatures; (2) new theoretical mass functions for dark matter haloes which provide a good fit to large numerical simulations; (3) more accurate mass-temperature relations from larger catalogues of hydrodynamical simulations; (4) the requirement to consider closed as well as open and flat cosmologies to obtain full multi-parameter likelihood constraints for CMB and SNe studies. We present a new sample of clusters drawn from the literature and use this sample to obtain improved results on $\\sigma_8$, the normalization of the matter power spectrum on scales of $8\\,h^{-1}\\,$Mpc, as a function of the matter density and cosmological constant in a Universe with general curvature. We discuss our differences with previous work, and the remaining major sources of uncertainty.  Final results on the normalization, approximately independent of power spectrum shape, can be expressed as constraints on $\\si$ at an appropriate cluster normalization scale $R_{\\rm Cl}$.  We provide fitting formulas for $R_{\\rm Cl}$ and $\\si(R_{\\rm Cl})$ for general cosmologies, as well as for $\\sigma_8$ as a function of cosmology and shape parameter $\\Gamma$. For flat models we find approximately $\\sigma_8\\simeq (0.495^{+0.034}_{-0.037})\\,\\Omega_{\\rm M}^{-0.60}$ for $\\Gamma=0.23$, where the error bar is dominated by uncertainty in the mass-temperature relation. ",
        "introduction": "In theories of hierarchical structure formation the class of objects most recently formed holds a special significance.  Observationally this class of objects is clusters of galaxies -- the largest virialized structures in the present day universe. The local abundance of rich clusters of galaxies provides a strong constraint on the fluctuations in the matter density on scales of order 10\\,Mpc \\cite{Evr89,FWED,BonMye91,HA91,Kaiser,Lil,OukBla,BahCen,Han,WEF}. Consistency with this constraint is one of the most important tests a model can pass, since the constraint is directly on the linear theory power spectrum, at a scale where there is an abundance of data. By fixing the normalization at wavelengths much smaller than those probed by {\\sl COBE\\/} one obtains an accurate local normalization on scales relevant to much of structure formation, a long lever arm for constraining the shape of the power spectrum, and a normalization to matter fluctuations which is independent of galaxy bias.  There have been several recent and detailed studies of the cluster abundance, including Bond \\& Myers \\shortcite{BonMye96}, Eke, Cole \\& Frenk~\\shortcite{ECF}, Viana \\& Liddle \\shortcite{VL96,VL98}, Colafrancesco, Mazzotta \\& Vittorio \\shortcite{ColMV}, Kitayama \\& Suto \\shortcite{KitSut}, Eke et al.~\\shortcite{ECFH}, Pen \\shortcite{Pen}, Wang \\& Steinhardt \\shortcite{WanSte}, Donahue \\& Voit \\shortcite{DoVoi99} and Henry \\shortcite{Henry}.  However, even more recently, there have been several developments in terms of both theory and observation which suggest it would be useful to revisit this constraint. Firstly the addition of \\ASCA\\ temperatures \\cite{ASCA} means that there is now a well defined local temperature function for clusters, with relatively small errors in temperature. Secondly a number of large N-body simulations have accurately determined the mass function of virialized haloes (e.g.~Governato et al.~1999), finding non-negligible deviations from the old Press-Schechter~(1974; hereafter PS) theory. For example the extremely large N-body simulations of the Virgo consortium have highlighted systematic departures from the PS predicted mass functions \\cite{JFWCCEY}, which alter the constraints on the power spectrum normalization coming from the cluster abundance. Thirdly, more ambitious hydrodynamical simulations of cluster formation (e.g.~Frenk et al.~1999, and references therein) have resulted in improvements in the relationship between mass and temperature and a better estimate of its scatter.  Finally, the increased sophistication of multi-parameter cosmological studies (in particular driven by recent CMB anisotropy measurements) requires that cosmological models with general curvature be considered (see e.g.~White \\& Scott 1996).  With these refinements our results are an improvement over other studies of the past few years.  We point out explicitly where we differ from other work, and also where we think things could be further improved in the future. The outline of the paper is as follows: we review some of the appropriate theory in \\S\\ref{sec:theory}; our local sample of clusters is presented in \\S\\ref{sec:data}; we describe our statistical method in \\S\\ref{sec:stats}; and we present our results and conclusions in \\S\\ref{sec:results} and \\S\\ref{sec:conclusions}. ",
        "conclusions": "\\label{sec:conclusions}  \\begin{table} \\caption{\\footnotesize% For a flat cosmology with $\\omm = 0.3$ we estimate how each source of uncertainty affects the mean and the error of our $\\sigma_{\\rm Cl}$ estimates.  $M{-}T|_{\\rm sys}$ and $M{-}T|_{\\rm stat}$ represent the systematic and statistical errors in the $M{-}T$ relation, while $T$ represents the error in the temperature of the clusters.  The $+1\\si$ and $-1\\si$ limits are found from integrating the distribution of 1000 Monte Carlos in each case.} \\label{tab:unc} \\begin{center} \\begin{tabular}{l|ccc} \\hline Source of error           & mean $\\si(R_{\\rm Cl})$  & $+1\\si$ & $-1\\si$ \\\\ \\hline all & 0.581 & 0.049 & 0.050 \\\\ $M{-}T|_{\\rm sys}$                    & 0.575 & 0.047 & 0.047 \\\\ $M{-}T|_{\\rm stat}$                   & 0.570 & 0.002 & 0.002 \\\\ $T$                                  & 0.586 & 0.005 & 0.018 \\\\ \\hline \\end{tabular} \\end{center} \\end{table}  The local abundance of rich clusters of galaxies currently provides one of the strongest constraints on the normalization of the present day dark matter power spectrum on a scale of ${\\sim}\\,10\\,$Mpc. While there have been numerous detailed studies of this constraint in the past, recent developments in both theory and observation have made it worthwhile to revisit this quantity.  We have calculated the constraint on $\\sigma_8$ arising from a new local sample of X-ray clusters with \\ASCA\\ temperatures.  We have incorporated the universal mass function determined from recent large N-body simulations, which is sufficiently different from the PS theory that the cosmological constraints inferred from cluster abundances change. We tried to carefully define the relation between mass and X-ray temperature for galaxy clusters, based on the results of many different hydrodynamical simulations.  Another major difference with previous work was that we performed all the comparisons at the `observed epoch' rather than carrying out an integration over `formation times'.  We also considered general combinations of $\\Omega_{\\rm M}$ and $\\Omega_{\\rm \\Lambda}$, including closed models.  Our results are best presented in terms of a normalization at the characteristic scale for our sample, $R_{\\rm Cl}$, where the normalization is largely independent of the power spectrum shape.  In the near future we anticipate a dramatic increase in our knowledge of the cosmological parameters from CMB anisotropy missions.  In particular it has been forecast \\cite{EHT} that \\MAP\\ will be able to determine $\\sigma_8$ to 14 per cent, at the same time as constraining a suite of other parameters.  The cluster abundance constraint is currently uncertain only at the ${\\sim}\\,10$ per cent level, and so certainly is a pivotal limit on models -- it is also an entirely independent constraint (at low $z$, in the mildly non-linear regime) compared with the CMB determinations (derived from the purely linear regime at $z\\simeq1000$). The uncertainty in the derived $\\sigma_8$ could be dramatically reduced with improvement in the $M{-}T$ relation, as well as through new X-ray data coming from \\Chandra\\ and \\XMM.  Future X-ray surveys, which go much fainter will lead to an increase in the size and quality of the X-ray data.  We showed in Fig.~\\ref{fig:vT} that there is lots of volume that was not probed at, say, $5\\,$keV. For example the \\ROSAT\\ Bright Survey \\cite{RBS} contains an approximately complete flux-limited sample of 302 clusters at $|b|\\,{>}\\,30^\\circ$, most of them lying at low redshift, which would be an excellent data-base if they all had good temperature estimates. More important than just increasing the precision of the X-ray temperature measurements is an increase in the quality of data, for the purposes of more fully understanding the comparison with models. Better spatial and spectral information from X-ray clusters should allow more reliable estimates for the value of $T_{\\rm X}$ which is most appropriate for comparing with the simulations.  In a similar vein, improved optical studies of lensing and velocity dispersions could improve mass determinations for individual clusters.  The major improvement will come through a more precise mass-temperature relation for X-ray clusters. This will require both the improved data which we can foresee from \\Chandra\\ and \\XMM, {\\it and\\/} improvements in modelling the gas physics relevant for understanding the X-ray properties of clusters. Eventually we imagine that the data will be fit more directly to much more ambitious simulations -- this will be extremely difficult in practice, since a wide range of scales needs to be modelled simultaneously. In the meantime the piece-meal approach we have taken here will continue to be useful.  With each of the ingredients, particularly the mass-temperature relation, improving with time, the cluster abundance will continue to be a strong constraint on the normalization of dark matter fluctuations on the ${\\sim}\\,10\\,$Mpc scale.  \\subsection*"
    },
    "0010/astro-ph0010513_arXiv.txt": {
        "abstract": "Double lined spectroscopic binaries in an evolved stage of evolution are expected to be extremely rare since they must consist of equally luminous and thus almost equally evolved objects, which requires an extremely similar initial mass. In this contribution we discuss such rare double evolved SB2 system: HD\\,172481. This binary includes an F-type post-AGB object and an M-type AGB companion. The spectrum shows a surprisingly strong Li\\,I 670.8\\,nm line with an equivalent width of $W_{\\lambda}$\\,=\\,54\\,m\\AA, yielding a lithium abundance of $\\log\\epsilon({\\rm Li}) = 3.6$. Several explanations for this huge lithium content are explored. ",
        "introduction": "In the course of our ongoing program to study the chemical composition of post-AGB stars (see Van Winckel, this volume), high resolution, high signal-to-noise spectra were taken of the F2Ia supergiant HD\\,172481. This star was previously selected as a candidate post-AGB star by Oudmaijer et al. (1992) because of its far infrared excess and its high galactic latitude ($b$\\,=\\,$-10.37^{\\circ}$). ",
        "conclusions": "In this contribution we shortly described our study on the very peculiar supergiant HD\\,172481. The TiO band heads superimposed on a typical F-type spectrum reveal that it is a binary object, which is further strengthened by the SED. The high radial velocities, the high galactic latitude and the moderate metal deficiency found in our chemical analysis suggest that the F-type component is an old and hence low-mass object. It is probably in its post-AGB phase of evolution as evidenced by the dust, the nature of the (low amplitude) variability and the slight s-process overabundances. Above all, this star is found to be super rich in lithium, with a Li abundance of $\\log\\epsilon({\\rm Li}) = 3.6$. A more detailed description can be found in Reyniers \\& Van Winckel (2000)."
    },
    "0010/astro-ph0010380_arXiv.txt": {
        "abstract": "The temperature distribution of the intracluster medium (ICM) in the Virgo cluster of galaxies has been derived from extensive mapping observations with ASCA covering an area of 19 deg$^2$. In the spectral analysis, the inner region within a radius of $\\sim 60'$ from M87 is characterized by an ICM temperature of $kT \\sim 2.5$ keV\\@ with little variation. On the other hand, the outer regions indicate significant variation of the temperature with an amplitude of about 1 keV\\@. The temperature map was produced from the hardness ratio (H.R.) values with a resolution of about $5'$. Besides the previously reported hot region with $kT > 4$ keV between M87 and M49, several hot regions with $kT = 3-4$ keV are detected in the cluster outskirts. The auto-correlation function for the H.R. variation shows that the temperature variation is correlated within a size of about 300 kpc, suggesting that gas blobs falling in the Virgo cluster have a typical size of groups of galaxies.  A correlation with the velocity dispersion of member galaxies shows that only the north-west region indicates an unusually large $\\beta_{\\rm spec}$ value of $2-4$. The upper limit for extended non-thermal emission in the Virgo cluster is obtained to be $L_X \\sim 9 \\times 10^{41}$ ergs s$^{-1}$ in the 2 $-$ 10 keV band. We discuss that these features consistently indicate that the Virgo cluster is in a relatively early stage of the cluster evolution. ",
        "introduction": "Clusters of galaxies are believed to be evolving over cosmological time scales. Studies of the redshift evolution of the X-ray luminosity function of clusters have been carried out, and the observed results indicate that the most luminous clusters tend to be sparse in the redshift range greater than about 0.3 (Rosati et al.\\ 1998). Recently, evidence for the evolution of individual clusters have been obtained in the form of temperature variations in the intracluster medium (ICM) from more than 10 systems (Henrikson \\& Markevitch 1996; Henry \\& Briel 1996; Honda, et al.\\ 1996; Davis \\& White 1998; Donnelly et al. 1998; Markevitch et al.\\ 1998; Churazov et al.\\ 1999; Markevitch et al. 1999; Molendi et al.\\ 1999; Shibata et al.\\ 1999; Watanabe et al. 1999). It is considered that gas heating is caused by collisions between subclusters and main clusters, and the temperatures are raised by several keV\\@. Detailed mappings of the temperature distribution in nearby clusters would give us information about the size and number distribution of the colliding blobs, which can be used to constrain theoretical models of structure formation.  The Virgo cluster of galaxies is the nearest cluster, with the enormously large angular extent of about $6^{\\circ}$ in radius. It is classified as Bautz-Morgan type III (Bautz \\& Morgan 1970), and Rood-Sastry type F (Rood \\& Sastry 1971) and shows a highly irregular morphology (Abell, Corwin, \\& Olowin 1989). This feature suggests that the cluster is undergoing dynamical evolution.  In the optical band, a catalogue containing 2096 galaxies within an area of $\\sim 140$ deg$^2$ roughly centered at $\\alpha \\sim 12^{\\rm h} 15^{\\rm m}$ and $\\delta \\sim 13^{\\circ}$ was produced by Binggeli, Sandage, \\& Tammann (1985) and Binggeli \\& Cameron (1993).  Optical properties of about 1300 member galaxies, such as morphologies, number counts, and heliocentric velocities, have been examined.  The distance to the cluster was measured with various methods and derived to be $12 - 22$ Mpc with a large uncertainty (summarized in Jacoby et al.\\ 1990).  Recently, Federspiel, Tammann, \\& Sandage (1998) derived $20.7 \\pm 2.4$ Mpc based on the 21 cm line width-absolute magnitude relation in the $B$ band.  Fukugita, Okamura, \\& Yasuda (1993) and Yasuda, Fukugita, \\& Okamura (1997), based on the Tully-Fisher relation for spiral galaxies, showed evidence that the galaxy distribution is largely elongated in the line of sight from 12 to 30 Mpc.  Jacoby et al.\\ (1990) showed that the M49 and M86 subclusters were located $\\sim 0.6 \\pm 0.7$ Mpc and $1.2 \\pm 0.6$ Mpc farther than the M87 subcluster on average.  The M86 distance is consistent with the result by Neilsen \\& Tsvetanov (2000, $2.5 \\pm 1.5$ Mpc).  In the X-ray band, the first indication of very extended emission over angular scales of $10^{\\circ}$ was obtained with non-imaging detectors on board the Ginga satellite (Takano 1990; Koyama, Takano, \\& Tawara 1991).  The X-ray emission is strongly peaked at M87, and well described by a $\\beta$ model with $\\beta = 0.436$, $R_{\\rm c} = 1'.62$, as previously obtained by the Einstein IPC data within 100$'$ from M87 (Fabricant \\& Gorenstein 1983).  In the southern outer part (2$^{\\circ}$ away from M87), the X-ray surface brightness distribution shows significant flattening with $\\beta \\sim 0.357$.  The ICM temperature around M87 was almost constant with $kT \\sim 2$ keV, but the Ginga data suggested a temperature rise toward M49\\@.  The ROSAT all-sky survey covered the Virgo cluster region between November 1990 and January 1991 in the energy range $0.1 - 2.4$ keV (B\\\"{o}hringer et al.\\ 1994).  On large scales, the X-ray morphology generally traces the galaxy distribution (Schindler, Binggeli, \\& B\\\"{o}hringer 1999).  However, the M49 halo is much smaller in the X-ray image compared to the optical appearance of the galaxies around M49\\@.  The X-ray surface brightness around M87 can be fitted by the $\\beta$ model with $\\beta = 0.45$ and $R_{\\rm c} = 2'.3$.  Although the overall temperature distribution around M87 is consistent with the Ginga result within $\\sim 1^{\\circ}$, ROSAT showed no evidence for the temperature rise around M49\\@.  Recently, mapping observations from ASCA have detected an unusual hot region between M87 and M49 with a temperature twice as high as the surrounding value of 2 keV (Kikuchi et al.\\ 2000). This provides clear evidence that the Virgo cluster is undergoing strong dynamical evolution.  In this paper, we have significantly expanded the mapped area and present a nearly complete view of the brightness and temperature distributions of the ICM\\@.  The closeness of this cluster enables us to look into detailed structures in the temperature distribution at a resolution of a few hundred kpc for the first time.  Results on the metal distribution will be reported in a separate paper (Shibata et al.\\ 2000).  We assume the distance to be 20 Mpc (e.g., Federspiel, Tammann, \\& Sandage 1998), and a $1'$ angular separation corresponds to 5.8 kpc.  The solar number abundance of Fe relative to H is taken as $4.68 \\times 10^{-5}$ (Anders \\& Grevesse 1989) throughout this paper. ",
        "conclusions": "The Virgo cluster is the nearest system and enables us to study of the temperature distribution with sufficient spatial resolution by ASCA\\@. We have obtained the full temperature map of this cluster for the first time, and found significant variation of the temperature with a spatial scale of about 300 kpc. This feature is probably an evidence that heating of the ICM proceeds with infall or mergers of gas clouds which have the size of galaxy groups, as expected from the current scenario of the structure formation. Comparison of the thermal energy of the ICM and the velocity dispersion of member galaxies have been carried out for the whole Virgo system. We have reached a picture that the M86 and M49 subclusters are in a different stage of subcluster mergers, with M86 just in the pre-heating phase and M49 already entered a shock heating phase. We have not detected extended non-thermal emission from the Virgo cluster.  Even the upper limit is much lower than those of other rich clusters (Coma cluster, A2199, and A2256), but is comparable to that of the galaxy group HCG62.  We expect that future X-ray observations would detect more evidences of such a local heating of the ICM in other clusters and give us a comprehensive view of the actual process of the cluster evolution."
    },
    "0010/astro-ph0010455_arXiv.txt": {
        "abstract": "We report on the exploitation of the stellar content of the Hamburg/ESO objective prism survey\\index{Hamburg/ESO survey} by quantitative selection methods. ",
        "introduction": "The Hamburg/ESO survey \\citep[HES; ][]{hespaperI,heshighlights,hespaperIII} covers the total southern ($\\delta < \\decdegmm{2}{5}$) extragalactic ($|b|\\gtrsim 30^{\\circ}$) sky in the magnitude range $13.0 \\gtrsim B_J \\gtrsim 17.5$. It is primarily aiming at finding bright quasars. However, at its spectral resolution of typically 15\\,{\\AA} FWHM at H$\\gamma$, it is also possible to efficiently select an abundance of interesting \\emph{stellar} objects. These include, e.g., metal-poor halo field stars, carbon stars, cataclysmic variable stars (CVs), white dwarfs (WDs), subdwarf B stars (sdBs), subdwarf O stars (sdOs), and field horizontal branch A- and B-type stars (FHB/A).  Example spectra of some of these stars are displayed in Fig. \\ref{star_examples}.  \\begin{figure}[htbp] \\begin{center} \\includegraphics[clip=, width=11.6cm, bbllx=48, bblly=420, bburx=527, bbury=740]{stellar_hesarea.ps} \\caption{\\label{stellar_hesarea} Definition of HES area (framed) and numbers of ESO/SERC fields in which the exploitation of the stellar content of the HES is currently carried out} \\end{center} \\end{figure}  \\begin{figure}[p] \\begin{center} \\includegraphics[ clip=, height=4cm, width=11cm, bbllx=203, bblly=451, bburx=501, bbury=569]{Cstars_cover.ps}\\\\ \\includegraphics[ clip=, height=4cm, width=11cm, bbllx=203, bblly=451, bburx=501, bbury=569]{HEShotstars_cover.ps}\\\\ \\includegraphics[ clip=, height=4cm, width=5.45cm, bbllx=52, bblly=451, bburx=200, bbury=569]{CVPN_cover.ps} \\includegraphics[ clip=, height=4cm, width=5.45cm, bbllx=354, bblly=451, bburx=501, bbury=569]{DZMPHS_cover.ps}\\\\ \\includegraphics[ clip=, height=4cm, width=11cm, bbllx=52, bblly=442, bburx=351, bbury=569]{DAFHBA_cover.ps} \\caption{\\label{star_examples} HES example spectra. Top panel: DQ white dwarf (left), cool carbon star (right); second panel: DB white dwarf, PG~1159 star; third panel: cataclysmic variable star, extremely metal-poor star, showing a very weak Ca~K line; lower panel: FHB/A star, cool DA white dwarf. Ordinates are photographic densities, abscissae wavelengths in {\\AA}ngstr\"om. Note that wavelength is \\emph{de}creasing from left to right. The sharp cutoff at $\\sim 5400$\\,{\\AA} is due to the IIIa-J emulsion sensitivity cutoff (``red edge'')} \\end{center} \\end{figure}  \\cite{Christlieb:2000} has developed quantitative object selection methods, such as automatic classification\\index{Automatic spectral classification}, for the systematic exploitation of the stellar content of the HES. In this paper we describe the methods of automatic classification used, and report on results obtained so far. ",
        "conclusions": ""
    },
    "0010/astro-ph0010333_arXiv.txt": {
        "abstract": "We use simulated maps of the cosmic microwave background anisotropy to quantify the ability of different statistical tests to discriminate between Gaussian and non-Gaussian models. Despite the central limit theorem on large angular scales, both the genus and extrema correlation are able to discriminate between Gaussian models and a semi-analytic texture model selected as a physically motivated non-Gaussian model. When run on the COBE 4-year CMB maps, both tests prefer the Gaussian model. Although the bispectrum has comparable statistical power when computed on the full sky, once a Galactic cut is imposed on the data the bispectrum loses the ability to discriminate between models. Off-diagonal elements of the bispectrum are comparable to the diagonal elements for the non-Gaussian texture model and must be included to obtain maximum statistical power. ",
        "introduction": "The statistics of the cosmic microwave background (CMB)  probe physical conditions in the early universe. Most inflationary models predict the CMB to follow Gaussian statistics \\cite{BardeenEtAl83,GuthPi82,Hawking82,Starobinski82}. Searching CMB data for non-Gaussian distributions is thus an important test of inflationary cosmologies. Recent analyses of the full-sky COBE data provide mixed results. Kogut {\\it et. al.} \\cite{Kogut95,Kogut96} use a likelihood analysis to compare the COBE-DMR 2- and 4-year maps to the standard Gaussian model and a set of non-Gaussian toy models.  The COBE data lie near the mode of the statistical distributions of the genus, 3-point correlation, and correlation of extrema points for the Gaussian model, while rejecting several non-Gaussian alternatives. Other authors using similar statistics on the COBE maps find comparable results \\cite{HinshawEtAl94,Smoot94,Luo94,TorresEtAl95,ColleyEtAl96,Heavens98}. More recently,  several authors have challenged the conclusion that the COBE data are well described by Gaussian statistics.  Tests using the normalized bispectrum \\cite{FMG98,Magueijo00} and a wavelet analysis \\cite{PVF98} both show the COBE data to be incompatible with Gaussian simulations at 98\\% confidence; however, these analyses did not test any non-Gaussian alternatives. The recent work of Mukherjee (2000) brings into question the the statistical significance of the \\cite{PVF98} result. In \\cite{Mukherjee} a proper Monte-Carlo  estimation of the significance level shows the wavelet technique does not yield strong evidence for non-Gaussianity in the 4-year COBE data. Bromley and Tegmark (1999) % argue that the non-Gaussian signal (if any) must lie in the {\\em phase} information in the COBE sky maps, while Banday {\\it et al.} (2000) % ascribe the discrepancy to instrumental artifacts in one of the DMR frequency  channels.  Searches for non-Gaussianity are complicated by several factors. On large angular scales, the superposition of multiple sources within the instrument angular resolution tends to push initially non-Gaussian distributions toward the Gaussian model (the Central Limit Theorem). Sample variance can occasionally produce a non-Gaussian distribution drawn from a Gaussian parent population. Given enough independent tests of the same observable sky, some outliers are expected \\cite{BT99}. {\\it Demonstrating that a statistic calculated for a CMB data set lying far from the mode expected for a Gaussian sky is a necessary but not sufficient condition to conclude that the data are not, in fact, a sample drawn from a Gaussian parent population.}  The statistical power of a test quantifies the probability of correctly accepting or rejecting hypotheses based on the results of that test. Quantifying the power for a given statistic thus requires at least two competing hypotheses. Since there is a single Gaussian distribution but an infinite number of non-Gaussian distributions, we must necessarily restrict the scope of specific non-Gaussian hypotheses. \\cite{Kogut95,Kogut96} used non-Gaussian toy models to show that the genus and extrema correlation could discriminate between Gaussian and non-Gaussian populations. We extend this analysis to include the bispectrum, replacing the earlier toy models with a model of topological defects as a specific example of a physically-motivated alternative to the standard Gaussian model. ",
        "conclusions": "Recent work has generated new interest in statistical tests of the CMB. Comparison of the bispectrum of the COBE-DMR 4-year sky maps to simulations of Gaussian maps show the COBE data to lie far from the mode of the Gaussian models \\cite{FMG98}. Since this work tested only Gaussian models, it is not clear whether the discrepant values result from non-Gaussian signals in the data (including instrumental artifacts) or are simply an outlier drawn from a Gaussian parent population. We compute the statistical power of several popular statistics (the genus, extrema correlation, and bispectrum) to determine their ability to distinguish between a Gaussian model and a texture model chosen as a physically motivated non-Gaussian alternative.  Both the genus and extrema correlation successfully discriminate between the models with 60\\% confidence, limited by the similarity of the texture model to a Gaussian on large angular scales (the central limit theorem). The COBE 4-year data prefer the Gaussian model, although the modest statistical power prevents a strong rejection of this specific non-Gaussian alternative. In contrast, we find that the bispectrum has no power to discriminate between models, and trace this to the aliasing of power caused by the Galactic cut imposed on the data. The normalized bispectrum {\\it never} has any power, even run on noiseless simulations with no Galactic cut. Since the results presented in \\cite{FMG98} and \\cite{Magueijo00} are based on an analysis in terms of the normalized bispectrum, the results, though correct, have no statistical power. Their results provide a necessary, but not sufficient, condition to rule out an underlying Gaussian character to the CMB anisotropy.   The problem of aliased power can in principle be solved, either by windowing the multipole coverage or by construction of an explicitly orthogonal basis. With this in mind, we further explore the statistical power of the bispectrum using full sky coverage to avoid the problem of aliased power. We find significant coupling between multipoles in the bispectrum coefficients for the non-Gaussian (texture) model. Failure to include these off-diagonal terms reduces the statistical power of the bispectrum on the full sky from 67\\% (comparable to the genus or extrema correlation) to 55\\% (not much better than random guessing). The off-diagonal elements are more important for the non-Gaussian model, pointing out the importance of testing various statistics against some alternative to the standard Gaussian model.    \\newpage \\appendix \\large \\begin{center}{\\bf Appendix}\\end{center} \\normalsize"
    },
    "0010/astro-ph0010619_arXiv.txt": {
        "abstract": "Our current understanding of the low surface brightness universe is quite incomplete, not only in the optical, but also in other wavelength regimes.  As a demonstration of the type of science which is facilitated by a virtual observatory, we have undertaken a project utilizing both images and catalogs to explore the multi-wavelength, low surface brightness universe. Here, we present some initial results of this project. Our techniques are complimentary to normal data reduction pipeline techniques in that we focus on the diffuse emission that is ignored or removed by more traditional algorithms. This requires a spatial filtering which must account for objects of interest, in addition to observational artifacts (\\eg bright stellar halos). With this work we are exploring the intersection of the catalog and image domains in order to maximize the scientific information we can extract from the federation of large survey data. ",
        "introduction": "Looking at large scale images (\\ie several degrees or more), one is immediately drawn to the high density of small galaxies, especially at high Galactic latitude, which are often strongly clustered. Interestingly enough, the vast majority of these galaxies have optical surface brightness distributions that are nearly identical to the terrestrial sky (Freeman 1970, Disney 1976). This interesting point, unless it is the manifestation of a cosmic coincidence, is most easily explained by accepting that current surveys suffer from an implicit surface brightness selection effect (see Disney 1998 for a stimulating discussion). As a result, untold numbers of galaxies remain uncatalogued with many important consequences.  For example, while the theoretical predictions of models of hierarchical structure formation have been successful in predicting the observed properties of the high redshift universe, they tend to over-predict the number of observed local group galaxies (Klypin \\etal 1999). This situation can be viewed as either a failure of the models, or a failure of the observations, possibly due to selection effects. In addition, galaxies which have low surface brightness (LSB) distributions have enormous cosmological implications (\\eg Impey \\& Bothun 1997), yet, they are relatively unexplored, primarily due to the intrinsic selection effects in finding them. For example, LSB galaxies constitute an unknown fraction of mass (both baryonic and dark matter) which must be accounted for when determining $\\Omega$ or $\\Lambda$.  In order to address these questions, we have initiated a project to reprocess the Digitized Palomar Observatory Sky Survey (DPOSS, Djorgovski \\etal 1998) optical photographic plate data in effort to find previously unknown, low surface brightness objects (Brunner \\etal 2001). Our techniques are complimentary to normal data reduction pipeline techniques in that we focus on the diffuse emission that is ignored or removed by more traditional algorithms (see also, Armandroff \\etal 1998). This requires a spatial filtering which must account for objects of interest, in addition to observational artifacts (\\eg bright stellar halos).  As part of this project, we have developed a novel background enhancement technique to look for new low surface brightness sources (see Figure 1 for a demonstration). Additional aspects affecting low surface brightness galaxy research in the context of a virtual observatory are addressed elsewhere (see, \\eg Schombert, J. in this volume).  \\begin{figure}[!htb] \\plotfiddle{brunnerr1-1.eps}{1.5 in}{0}{100}{100}{-120}{0} \\caption{The left image is an approximately 17 arcminute square cutout of the DPOSS F plate image containing the dwarf Spheroidal Andromeda III. The image on the right is the background enhanced image generated by our software pipeline. The elongated object in the background image, which is Andromeda III, is clearly detected via this technique.} \\end{figure} ",
        "conclusions": ""
    },
    "0010/astro-ph0010105_arXiv.txt": {
        "abstract": "Recent studies suggest that when magnetohydrodynamic (MHD) turbulence is excited by stirring a plasma at large scales, the cascade of energy from large to small scales is anisotropic, in the sense that small-scale fluctuations satisfy the inequality $k_\\parallel \\ll k_\\perp$, where $k_\\parallel$ and $k_\\perp$ are, respectively, the components of a fluctuation's wave vector $\\parallel$ and $\\perp$ to the background magnetic field. Such anisotropic fluctuations are very inefficient at scattering cosmic rays.  Results based on the quasilinear approximation for scattering of cosmic rays by anisotropic MHD turbulence are presented and explained. The important role played by molecular-cloud magnetic mirrors in confining and isotropizing cosmic rays when scattering is weak is also discussed. ",
        "introduction": "In diffusion models of Galactic cosmic-ray propagation, cosmic rays are scattered by small-scale fluctuations in the interstellar magnetic field. For cosmic-ray energies below $\\sim 10^2 -10^3$ GeV, these small-scale fluctuations can arise from resonant waves that the cosmic rays generate themselves. At higher energies, it is believed that self-confinement is not possible, because the growth rates of the resonant modes become too small in comparison to the rates at which the modes are damped (\\citeauthor{ces80} \\citeyear{ces80}, \\citeauthor{ber90} \\citeyear{ber90}). For cosmic rays with energies above $10^2-10^3$ GeV, scattering can result from turbulence that is generated by large-scale stirring of the interstellar medium (ISM), which results in a cascade of magnetic energy from large to small scales. Recent studies of magnetohydrodynamic (MHD) turbulence, however, find that the small-scale fluctuations resulting from a turbulent cascade satisfy the inequality $k_\\parallel \\ll k_\\perp$, where $k_\\parallel$ and $k_\\perp$ are, respectively, the components of a fluctuation's wave vector $\\parallel$ and $\\perp$ to the background magnetic field ${\\bf B}_0$ (e.g., \\citeauthor{gol95} \\citeyear{gol95}). This paper explains why fluctuations with this anisotropy are very inefficient at scattering cosmic rays, and presents results from quasilinear theory for the scattering mean free paths resulting from anisotropic MHD turbulence. The role of molecular clouds in confining and isotropizing cosmic rays when scattering is weak is also discussed. ",
        "conclusions": "Recent investigations into MHD turbulence are providing new and important results on the anisotropy of the small-scale fluctuations that result from a cascade of magnetic energy from large to small scales. As discussed in this paper, anisotropic small-scale fluctuations are inefficient at scattering cosmic rays. For cosmic rays with energies less than $10^2- 10^3$ GeV, resonant waves excited by streaming cosmic rays are believed to be sufficient to confine cosmic rays to the Galaxy regardless of the nature of the cascade in MHD turbulence. At higher energies, however, it is believed that such self-generated waves are insufficient. Thus, if scattering by the turbulence that is generated by large-scale stirring of the ISM is inefficient, then some additional mechanism is needed to confine and isotropize cosmic rays at energies above $\\sim 10^2 - 10^3$ GeV. Such a mechanism may be provided by molecular-cloud magnetic mirrors."
    },
    "0010/astro-ph0010275_arXiv.txt": {
        "abstract": "We report on multi-object spectroscopy in the red spectral region of 37 candidate star clusters in an $\\sim 8 \\times 8$ arcmin$^2$ field centered on the giant early-type radio galaxy NGC~1316 (Fornax~A), the brightest galaxy in the Fornax cluster. Out of this sample, 24 targets are found to be genuine star clusters associated with NGC~1316, % and 13 targets are Galactic foreground stars. For the star cluster sample, we measure a mean heliocentric velocity $v_{\\rm hel} = 1698 \\pm 46$ \\kms\\ and a velocity dispersion $\\sigma = 227 \\pm 33$ \\kms\\ within a galactocentric radius of 24 kpc. Partly responsible for the velocity dispersion is a significant rotation in the star cluster system, with a mean velocity of $\\sim 175 \\pm 70$ \\kms\\ along a position angle of $\\sim 6\\degr \\pm 18\\degr$. Using the projected mass estimator and assuming isotropic orbits, the estimated total mass is ($6.6 \\pm 1.7$) $\\times$ 10$^{11}$ \\Mzon\\ within a radius of 24 kpc. The mass is uncertain by about a factor of two, depending on the orbital assumptions. % The implied $\\cal{M}$/$L_B$ ratio is in the range 3\\,--\\,6. Four star clusters in our sample are exceptionally luminous  % ($M_V < -12.3$). This means that (1) at least this many clusters in NGC~1316 are up to an order of magnitude more luminous than the most luminous star cluster in our Galaxy or M\\,31, and (2) that the S/N ratio of their spectra allows us to measure line strengths with good accuracy. By comparing the measured colours and equivalent widths of \\Ha\\ and the \\CaII\\ triplet ($\\lambda\\lambda$\\,8498,\\,8542,\\,8662 \\AA) absorption lines for those bright star clusters in our sample with those of single-burst population models (the Bruzual \\& Charlot 1996 models), we find that they are coeval with an age of 3.0 $\\pm$ 0.5 Gyr. Their metallicities are found to be solar to within $\\pm$ 0.15 dex. We discuss the properties of the main body of NGC~1316 and conclude they are consistent with having hosted a major merger 3 Gyr ago as well. The presence of intermediate-age globular clusters in NGC~1316 shows once again that globular clusters with near-solar metallicity do form during galactic mergers, and, moreover, that they can {\\it survive\\/} disruption processes taking place during the merger (e.g., dynamical friction, tidal disruption), as well as evaporation. In this respect, NGC~1316 provides a hitherto ``missing'' evolutionary link between young merger remnants of age $\\sim$\\,0.5 Gyr such as NGC~3597, NGC~3921 and NGC~7252 on one side, and older giant ellipticals featuring bimodal colour distributions on the other side. ",
        "introduction": "\\label{s:intro}  \\subsection{Extragalactic Globular Cluster Systems}  Recent observations with the {\\it Hubble Space Telescope (HST)\\/} and large-field ground-based CCD cameras have caused rapid advances in our knowledge of the formation and evolution of star cluster systems of galaxies. Extragalactic star clusters have in recent years established themselves as potential tracers of the formation and evolution of galaxies. The number of detailed photometric studies is increasing steadily, and these studies reveal interesting connections between star cluster systems and their host galaxies.  One important, well-known aspect of star cluster systems among galaxies is that the number of star clusters per unit galaxy luminosity (named the specific frequency $S_N$) increases systematically from late-type to early-type galaxies, being $\\sim$\\,2--3 times higher in elliptical (E) than in spiral galaxies of type Sb and later (Harris \\& van den Bergh 1981; Harris 1991). This fact has been used as an argument against the scenario for forming ellipticals through mergers (e.g., van den Bergh 1995). However, recent observations with the {\\it HST\\/} have led to a wealth of discoveries of young star clusters in merging and starburst galaxies. Indeed, it now seems that star clusters in general (and globular clusters in particular) may form preferentially in high-density regions of starbursts (e.g., Meurer et al.\\ 1995), probably from giant molecular clouds whose collapse is being triggered by a 100-- to 1000-fold increase in gas pressure due to supernova and shock heating during starbursts (e.g., Elmegreen \\& Efremov 1997). Galactic gas-rich mergers are known to produce the most energetic known starbursts (e.g., Sanders \\& Mirabel 1996), to create galaxy remnants that have surface brightness profiles consistent with those of ``old'' giant ellipticals (Wright et al.\\ 1990), and to create large numbers of young globular clusters (e.g., Whitmore 1999 and references therein). It is possible that the higher specific frequency of globular clusters in ellipticals relative to that in spirals may be accounted for by secondary populations of globular clusters created during gas-rich mergers.  In this respect, one particularly interesting feature of the star cluster systems of many giant ellipticals is the presence of bimodal colour distributions, providing clear evidence for the occurrence of a ``second event'' in the formation of these systems (e.g., Zepf \\& Ashman 1993; Whitmore et al.\\ 1995; Geisler, Lee \\& Kim 1996; Kissler--Patig et al.\\ 1997; Carlson et al.\\ 1998). While such a bimodal colour distribution was actually predicted from merger models of E galaxy formation by Ashman \\& Zepf \\shortcite{ashzep92}, opinions about the general nature of the ``second event'' differ among authors (see detailed review by Ashman \\& Zepf 1998). In particular, the ``second event'' could actually be a series of different events that lead to the formation of metal-rich globular clusters. It is therefore important to obtain any additional evidence for (or against) a merger origin of the star cluster bimodality. One important constraint to {\\it any\\/} scenario to explain the bimodality is set by the colour of the ``red'' peak in the colour distribution among giant ellipticals, which (if interpreted in terms of metallicity) typically indicates a near-solar metallicity (e.g., Forbes, Brodie \\& Grillmair 1997 and references therein). This metallicity estimate has since been confirmed by Keck/LRIS {\\it spectroscopy\\/} of globular clusters in M\\,87 (Cohen, Blakeslee \\& Rhyzov 1998) and in NGC 1399 (Kissler--Patig et al.\\ 1998). In order to render the scenario in which the second-generation ``red'' star clusters are formed during galaxy mergers generally applicable, it is therefore of great interest to find out just what the metallicities of the young, luminous star clusters in merger remnants actually are, and whether or not they can be found in merger remnants at different times after the last (major) merger.  The current evidence that young star clusters in mergers have approximately solar metallicities is still sparse. Photometry of these clusters typically does not provide enough information, due to the age-metallicity degeneracy of (especially optical) colours and the fact that most merger remnants contain extensive dust features with appreciable extinction. Progress has recently been made by using combinations of near-infrared and optical colours (Maraston et al.~2000), but this method is still in its infancy.  The only dependable evidence stems from relatively strong metallic features observed in {\\it spectra\\/} of one bright cluster in the peculiar elliptical galaxy NGC~1275 (Zepf et al.\\ 1995a; Brodie et al.\\ 1998) and of two such clusters in NGC~7252 \\cite{schsei98}, a prototypical, 0.5\\,--\\,1 Gyr old remnant of two merged spiral galaxies (Hibbard \\& Mihos 1995; Schweizer 1998). It is clear that the need for more spectroscopic metallicity estimates for young star clusters in (other) mergers is imminent.  Spectroscopy of star clusters around ellipticals can also provide one with the {\\it kinematics\\/} of the cluster system, another essential source of information for discriminating between different galaxy formation scenarios. For example, spectroscopic studies of M\\,87 and NGC 1399 revealed a significantly higher velocity dispersion for the globular clusters than for the stars (Mould et al.~1990; Brodie and Huchra 1991; Grillmair et al.~1994; Cohen \\& Ryzhov 1997; Minniti et al.\\ 1998; Kissler--Patig \\& Gebhardt 1998; Kissler-Patig et al.\\ 1999). As both M\\,87 and NGC~1399 are central cluster ellipticals, these authors suggested that the globular clusters were reacting to the gravitational potential of the galaxy cluster as a whole rather than that of the galaxy itself. In NGC~5128 (Cen~A), Hui et al.\\ \\shortcite{hui+95} reported rotation in the globular cluster system, though only for the metal--rich clusters which seemed to rotate along with the well-known dust lane. However, the situation in NGC 4472 (Sharples et al.~1998) and M87 (Kissler-Patig \\& Gebhardt 1998) seems to be different in that the metal--poor globular clusters seem to dominate the rotation.  The present paper describes new spectroscopic observations aimed at deriving kinematics, ages and metallicities for a suitable sample of star clusters in the merger remnant NGC~1316.  \\subsection{NGC 1316}  \\begin{figure} \\centerline{\\psfig{figure=n1316Bb.eps,width=8.4cm}} \\caption[]{Grey-scale plot of B-band CCD image of the central 7\\farcm5 $\\times$ 7\\farcm5 of NGC~1316 (taken from Paper~II). The plate scale is indicated by the bar at the lower left of the image. North is up and East is to the left.} \\label{f:fig1} \\end{figure}  The giant early-type galaxy NGC~1316 (= Fornax A = PKS 0320\\,$-$\\,37 = ESO 357\\,$-$\\,G022 = ARP 154) is one of the brightest and closest radio galaxies in the sky. It is located in the outskirts of the Fornax cluster, at a projected distance of 3\\fdg7 from NGC~1399, the central giant elliptical galaxy. The extensive optical observations of Schweizer (1980) showed that NGC~1316 is a typical Morgan D-type galaxy with an elliptical spheroid embedded in an extensive, box-shaped envelope. Several features of NGC~1316 establish firmly that it is a merger remnant. The outer envelope includes several non-concentric arcs, tails and loops that are most likely remnants of tidal perturbations, while the inner part of the spheroid is characterized by a surprisingly high central surface brightness and small core radius (and effective radius) for its galaxy luminosity (Schweizer 1981; Caon, Capaccioli \\& D'Onofrio 1994). These characteristics, together with a velocity dispersion which is significantly lower than that of other elliptical galaxies of similar luminosity, cause NGC~1316 to lie far off the fundamental plane of early-type galaxies or the Faber-Jackson relation (e.g., D'Onofrio et al.\\ 1997). Furthermore, the inner spheroid is marked by non-concentric ripples of 0.1--0.2 mag amplitude above a best-fitting $r^{1/4}$ law (see discussion in Sect.~\\ref{s:dyn}.3).  Figure \\ref{f:fig1} shows a $B$-band image of the inner 7\\farcm5\\,$\\times$\\,7\\farcm5 of NGC~1316 (taken from Goudfrooij, Alonso \\& Minniti 2000) in which some of the features mentioned above can be found. All of these features are consistent with NGC~1316 having undergone a relatively recent merger after which dynamical relaxation has not yet had time to complete fully.  \\begin{table} \\caption[ ]{Global properties of NGC~1316.} \\label{t:n1316} \\begin{tabular*}{8.4cm}{@{\\extracolsep{\\fill}}lll@{}} \\hline \\hline \\multi{3}{c}{~~} \\\\ [-1.8ex] Parameter & Value & Reference \\\\ [0.5ex] \\hline \\multi{3}{c}{~~} \\\\ [-1.8ex] RA~~(J2000.0)     & 03$^{\\rm d}$22$\\!^{\\rm m}$41\\fs710 & {\\tt *} \\\\ DEC~(J2000.0)     & $-$37\\degr12$'$\\,29\\farcs41     \t & {\\tt *} \\\\ Galaxy Type    & (R$'$)SAB(s)0    \t   \t& RC3 \\\\ & S0$_1$pec \t     \t   \t& RSA \\\\ $r_{\\scrm{eff}}$  & 109$''$\t\t\t& CCD94 \\\\ $\\sigma_0$        & 221 km s$^{-1}$ \t\t& K00 \\\\ $B_T$             & 9.42 \t\t\t& RC3 \\\\ $A_{B,\\,{\\rm foreground}}$ & 0.00\t\t& BH84 \\\\ $(B-V)_{\\rm eff}$ & 0.93 \t\t\t& RC3 \\\\ $(U-B)_{\\rm eff}$ & 0.47 \t\t\t& RC3 \\\\ $v_{\\rm hel}$  & 1760 km s$^{-1}$\t\t& NED \\\\ Distance\t  & 22.9 Mpc \t\t\t& {\\tt **} \\\\ $M_{B_T}^0$       & \\llap{$-$}22.38\t\t& {\\tt ***} \\\\ \\multi{3}{c}{~~} \\\\ [-1.8ex] \\hline \\multi{3}{c}{~~} \\\\ [-1.8ex] \\end{tabular*}  \\baselineskip=0.97\\normalbaselineskip {\\small \\noindent {\\sl Notes to Table~\\ref{t:n1316}.} \\\\ {\\it Parameters:\\/~}$r_{\\scrm{eff}} \\cor$ Effective radius; $\\sigma_0 \\cor$ central velocity dispersion; $B_T \\cor$ Total B magnitude; $A_{B,\\,{\\rm foreground}} \\cor$ Absorption in B band due to ISM in our Galaxy; $(B-V)_{\\rm eff}$, $(U-B)_{\\rm eff} \\cor\\;$Mean colours inside the effective radius; $v_{\\rm hel} \\cor\\;$Heliocentric velocity; $M_{B_T}^0 \\cor\\;$Total absolute B magnitude, corrected for Galactic absorption. \\\\ {\\it References:\\/}~RC3 $\\cor$ de Vaucouleurs \\etal (1991); RSA $\\cor$ Sandage \\& Tammann (1987); CCD94 $\\cor$ Caon et al.\\ (1994); K00 $\\cor$ Kuntschner (2000); BH84 $\\cor$ Burstein \\& Heiles (1984); NED $\\cor$ NASA/IPAC Extragalactic Database (http://nedwww.ipac.caltech.edu); {\\tt *} $\\cor$ derived from archival HST/WFPC2 images; {\\tt **} $\\cor$ derived from supernova light curves, see text; {\\tt ***} $\\cor$ derived from $B_T$ and $A_{B,\\,\\scrm{foreground}}$. }  \\end{table}  As to the adopted distance of NGC~1316, we take advantage of the fact that two well-observed type Ia supernovae (SNe\\,Ia) have occurred in NGC~1316 (SN1980N and SN1981D). Using the precise distance indicator for SNe\\,Ia that utilizes the tight relation between their light curve shape, luminosity, and colour (Riess et al.\\ 1998; A.\\,G.\\ Riess, private communication), we arrive at a distance of 22.9 ($\\pm$ 0.5) Mpc for NGC~1316, equivalent to $(m\\!-\\!M)_0$ = (31.80 $\\pm$ 0.05).  At this distance, 1$''$ corresponds to 111 pc. Note that this sets NGC 1316 and the sub-group of galaxies surrounding it slightly behind the core of the Fornax cluster (for which $(m\\!-\\!M)_0$ = 31.54 $\\pm$ 0.14; Ferrarese et al.~2000). Global galaxy properties of NGC~1316 are listed in Table~\\ref{t:n1316}.  This paper is built up as follows. Section~\\ref{s:sample} describes the sample selection. The observations are described in section~\\ref{s:obs}, while section~\\ref{s:redu} deals with the data reduction. The various results are presented and discussed in sections~\\ref{s:res} and \\ref{s:disc}, respectively. Finally, section~\\ref{s:concl} summarizes the main conclusions of this study. ",
        "conclusions": "\\label{s:concl}  We have obtained spectra for 37 candidate star clusters in the field of the giant early-type galaxy NGC~1316, an obvious merger remnant featuring extensive shells, tails, and dusty filaments. Our main conclusions are as follows. \\begin{itemize} \\item The radial velocity measurements reveal that 24 targets are genuine star clusters, and 13 targets are Galactic foreground stars. Almost all (22/24) genuine star clusters fall in the narrow colour range $1.6 \\la B\\!-\\!I \\la 2.1$, i.e., only 2 red targets with $B\\!-\\!I > 2.1$ are highly reddened star clusters, the rest are foreground stars. \\item For the star cluster sample, we measure a mean heliocentric velocity $v_{\\rm hel} = 1698 \\pm 46$ \\kms\\ and a velocity dispersion $\\sigma = 227 \\pm 33$ \\kms\\ within a galactocentric radius of 24 kpc. Partly responsible for the velocity dispersion is a significant rotation in the star cluster system, with a mean velocity of $\\sim 175\\pm70$ \\kms\\ along a position angle of $\\sim 6\\degr \\pm 18\\degr$. Using the projected mass estimator and assuming isotropic orbits, the estimated total mass is ($6.6 \\pm 1.7$) $\\times$ 10$^{11}$ \\Mzon\\ within a radius of 24 kpc. The mass is uncertain by about a factor of two, depending on the orbital assumptions and adopted distance. The above mass implies a $\\cal{M}$/$L_B$ ratio in the range 3\\,--\\,6, which is low for giant elliptical galaxies. This may indicate that the stellar population of the integrated light of NGC~1316 is, in a luminosity-weighted sense, younger than usual. \\item For three bright star clusters in our sample, the signal-to-noise ratio is good enough to measure equivalent widths of \\Ha\\ and the \\CaII\\ triplet absorption lines with confidence. Comparing those line strengths as well as measured broad-band colours with recent single-burst stellar population models (BC96), the metallicities of those 3 star clusters are found to be solar to within $\\pm$ 0.15 dex, and their most likely age is 3 Gyr. \\item We discuss the properties of the main body of NGC~1316 itself and conclude they are consistent with having hosted a major merger 3 Gyr ago as well. The presence of intermediate-age globular clusters in NGC~1316 shows once again that globular clusters with near-solar metallicity do form during galactic mergers, and, moreover, that they can {\\it survive\\/} disruption processes taking place during the merger (e.g., dynamical friction, tidal disruption), as well as evaporation. In this respect, NGC~1316 provides a hitherto ``missing'' evolutionary link between the recently studied young merger remnants of $\\sim$\\,0.5 Gyr age such as NGC~3921 and NGC~7252 on one side, and ``old'' giant ellipticals featuring bimodal colour distributions on the other side. \\end{itemize}  \\paragraph*{Acknowledgments.} \\ \\\\ PG is grateful to the European Southern Observatory for an invited visitorship during which part of this research was carried out, and for allocating observing time to this project. We acknowledge enlightening discussions with Fran\\c{c}ois Schweizer, Claudia Maraston, and Adam Riess. We have made use of the NASA/IPAC Extragalactic Database (NED) which is operated by the Jet Propulsion Laboratory, Caltech, under contract with the National Aeronautics and Space Administration. JM and PG would like to thank the director of STScI for financial support through the Director's Discretionary Research Fund. The work of DM is partly supported by the Chilean FONDECYT No.\\ 01990440."
    },
    "0010/astro-ph0010043_arXiv.txt": {
        "abstract": "The environs of three supernova remnants (SNR)  with associated OH 1720 MHz masers, G349.7+0.2, CTB 37A and G16.7+0.1, have been surveyed in the CO J=1-0 transition with the 12 Meter Telescope of the NRAO, using the On-The-Fly technique. These observations have revealed a number of molecular clouds interacting with the SNR shock fronts. Most of the OH 1720 MHz masers have been found to lie over CO concentrations, and the maser velocities are coincident with the CO peak velocities to an accuracy better than 2 km/s. The present data trace the interstellar medium (ISM) structures interacting with the SNRs; however, to probe the shocked molecular gas in which the OH 1720 MHz emission originates, higher excitation transitions and more complex species should be observed. In CTB 37A, where the shock velocity into the molecular cloud could be determined, it has been found to be of C-type, in agreement with theoretical predictions. Part of the rim of G16.7+0.1 appears to be flattened by a dense external cloud, yet the only associated OH 1720 MHz maser lies near the opposite region of the remnant. This behavior, also observed in IC 443 and 3C 391, seems to contradict the suggestion that OH 1720 MHz maser emission occurs mainly for transverse shocks. ",
        "introduction": " ",
        "conclusions": ""
    },
    "0010/astro-ph0010569_arXiv.txt": {
        "abstract": "We have studied a mass model for spiral galaxies in which the dark matter surface density is a scaled version of the observed HI surface density.  Applying this mass model to a sample of 24 spiral galaxies with reliable rotation curves one obtains good fits for most galaxies. The scaling factors cluster around 7, after correction for the presence of primordial helium. But for several cases different, often larger, values are found. For galaxies that can not be fitted well the discrepancy occurs at large radii and results from a fairly rapid decline of the HI surface density in the outermost regions.  Because of such imperfections and in view of possible selection effects it is not possible to conclude here that there is a real coupling between HI and dark matter in spiral galaxies. ",
        "introduction": "The first indications that spiral galaxies are surrounded by unseen matter date to the early seventies (Freeman 1970; Roberts 1975; Shostak 1973). Later 21-cm line studies, especially those of Bosma (1978,1981) and Begeman (1987), have all confirmed this discovery. Dark matter manifests itself already in the outer(most) parts of the luminous body of spiral galaxies (Rubin 1978; Kent 1986), but its main signature is the general flatness of rotation curves well outside the optical radius.  Bosma (1978, 1981) noted a curious result for the galaxies in his sample: the total surface density of matter needed to explain the observed rotation curve in the outer region, i.e. outside the optical radius, was roughly proportional to the surface density of neutral hydrogen. Later work has shown that this proportionality not only holds for the high surface brightness (HSB) galaxies studied by Bosma, but also for dwarf galaxies (Carignan 1985a; Carignan \\& Puche 1990; Jobin \\& Carignan 1990).  These results, and other considerations, have led Pfenniger et al. (1994a, b) to the hypothesis that the dark matter surrounding spiral galaxies consists of cold gas, mainly in the form of molecular hydrogen. The spatial distribution of this cold gas should be similar to that of the observed neutral hydrogen.  As the contribution of luminous matter to the total mass density is negligible in the outer parts (under the assumption that the mass-to-light ratio of the luminous matter is independent of radius), Bosma's relationship is equivalent to a proportionality of dark matter and neutral hydrogen.  This apparent coupling of dark matter and neutral hydrogen deserves closer scrutiny, because it seems at odds with the current view that dark haloes are triaxial (but not extremely flattened) and that they consist of non-baryonic material. Rather, this coupling suggests possible support for the hypothesis put forth by Pfenniger et al. (1994a).  Several questions should be considered: (i) How tight is this relation between the surface densities of HI and dark matter for the galaxies with the best available data?  (ii) To what extent does the relation depend on the use of the so-called maximum disc assumption? (iii) Does the relation also hold for the low surface brightness (LSB) galaxies?  In this paper we investigate the apparent coupling of dark matter and neutral hydrogen for a sample of 24 spiral galaxies with good rotation curves. The sample contains both HSB and dwarf galaxies. Unfortunately the quality of the rotation curves for LSB galaxies is still rather poor and we have decided not to include those galaxies in the present discussion.   We wish to stress at the outset that there is a potentially powerful selection effect that may cause a relationship between the surface densities of HI and dark matter for the galaxies in our sample. This is because the HI surface density distributions of the galaxies in our sample have the common characteristic that the highest values in the inner regions, as well as the lowest values in the outer regions are similar from galaxy to galaxy (the latter being linked to the sensitivity reached in the observations). As a result, a large radial extent implies a large radial scale length of HI and vice versa. In section~5 we discuss how this can lead to a proportionality between dark matter and HI. ",
        "conclusions": "We have fitted mass models to the observed rotation curves of 24 galaxies under the assumption that the dark matter surface density is proportional to the HI surface density. For about two thirds of the galaxies we obtain good fits to the data. After correction for primordial Helium the required scaling factor is about 7. The main discrepancies occur at large radii, with the fitted rotation curves falling below the observed ones. This results from a fairly rapid decline of the HI surface density in the outermost regions.  For the HI data used here, the radial density profiles of HI typically extend to about 3 scale lengths, i.e. close to or just beyond the radius where the rotation curve of the HI component is expected to decline. The good fits are therefore somewhat coincidental.  More sensitive HI observations would reach the declining part of the rotation curve of the HI component for several galaxies in our sample (provided the HI surface density remains exponential further out). If the observed circular velocity of the galaxy remains constant with radius it would then no longer be possible to fit such a flat rotation curve with a scaled up distribution of HI. Furthermore, in several galaxies scaling up of HI leads to rotation curves with wiggles, and in some galaxies not included in this paper the neutral hydrogen density drops rather abruptly around the optical radius.  Therefore, in many cases, there seems to be little or no relation between HI and dark matter.  Note, however, that for a number of reasons (ionisation by the intergalactic radiation field, presence of molecular gas and missing HI flux) the observed HI densities may be lower than the actual gas densities.  Because of these complications and the selection effects sketched above, it is not possible to conclude here that there is a real coupling between HI and dark matter in spiral galaxies."
    },
    "0010/astro-ph0010057_arXiv.txt": {
        "abstract": "We use milli-arcsecond accuracy astrometry (proper motions and parallaxes) from Hipparcos and from radio observations to retrace the orbits of 56 runaway stars and nine compact objects with distances less than 700 pc, to identify the parent stellar group. It is possible to deduce the specific formation scenario with near certainty for two cases. (i) We find that the runaway star $\\zeta$~Ophiuchi and the pulsar PSR~J1932$+$1059 originated about 1~Myr ago in a supernova explosion in a binary in the Upper Scorpius subgroup of the Sco OB2 association. The pulsar received a kick velocity of $\\sim$350 km s$^{-1}$ in this event, which dissociated the binary, and gave $\\zeta$~Oph its large space velocity. (ii) Blaauw \\& Morgan and Gies \\& Bolton already postulated a common origin for the runaway-pair AE~Aur and $\\mu$~Col, possibly involving the massive highly-eccentric binary $\\iota$~Ori, based on their equal and opposite velocities. We demonstrate that these three objects indeed occupied a very small volume $\\sim$2.5~Myr ago, and show that they were ejected from the nascent Trapezium cluster.  We identify the parent group for two more pulsars: both likely originate in the $\\sim 50$~Myr old association Per OB3, which contains the open cluster $\\alpha$ Persei. At least 21 of the 56 runaway stars in our sample can be linked to the nearby associations and young open clusters. These include the classical runaways 53~Arietis (Ori~OB1), $\\xi$~Persei (Per~OB2), and $\\lambda$~Cephei (Cep~OB3), and fifteen new identifications, amongst which a pair of stars running away in opposite directions from the region containing the $\\lambda$ Ori cluster. Other currently nearby runaways and pulsars originated beyond 700 pc, where our knowledge of the parent groups is very incomplete. ",
        "introduction": "\\label{sec:intro} About 10--30\\% of the O stars and 5--10\\% of the B stars (Gies \\cite{gie1987}; Stone \\cite{sto1991}) have large peculiar velocities (up to 200~km~s$^{-1}$), and are often found in isolated locations; these are the so-called `runaway stars' (Blaauw \\cite{bla1961}, hereafter Paper I). The velocity dispersion of the population of runaway stars, $\\sigma_\\mathv \\sim 30$~km~s$^{-1}$ (e.g., Stone \\cite{sto1991}), is much larger than that of the `normal' early-type stars, $\\sigma_\\mathv \\sim 10$~km~s$^{-1}$.  Besides their peculiar kinematics, runaway stars are also distinguished from the normal early-type stars by an almost complete absence of multiplicity (cf.\\ the binary fraction of normal early-type stars is $>$50\\% [e.g., Mason et al.\\ \\cite{mas1998}]). Furthermore, over 50\\% of the (massive) runaways have large rotational velocities and enhanced surface helium abundances (Blaauw \\cite{bla1993}).  Several mechanisms have been suggested for the origin of runaway stars (Zwicky \\cite{zwi1957}; Paper~I; Poveda, Ruiz \\& Allen \\cite{pra1967}; Carrasco et al.\\ \\cite{car1980}; Gies \\& Bolton \\cite{gb1986}), two of which are still viable: the binary-supernova scenario (Paper~I) and the dynamical ejection scenario (Poveda et al.\\ \\cite{pra1967}). We summarize them in turn.   \\paragraph{Binary-supernova scenario (BSS):} In this scenario a runaway star receives its velocity when the primary component of a massive binary system explodes as a supernova. When the supernova shell passes the secondary the gravitational attraction of the primary reduces considerably, and the secondary starts to move through space with a velocity comparable to its original orbital velocity (30--150~km~s$^{-1}$).  What remains of the primary after the explosion is a compact object, either a neutron star or a black hole. Depending on the details of the preceding binary evolution, the eccentricity of the orbit, and the kick velocity $\\vec{\\mathv}_\\mathrm{kick}$ due to the asymmetry of the supernova explosion (e.g., Burrows, Hayes \\& Fryxell \\cite{bhf1995}), the compact object may or may not remain bound to the runaway star (e.g., Hills \\cite{hil1983}). In most cases less than half of the total mass of the binary is expelled in the explosion. As this is insufficient to break up the binary (Paper I), most BSS runaways are expected to remain double.  Examples of such systems are provided by the high-mass-X-ray binaries (e.g., van den Heuvel et al.\\ \\cite{heu2000}). Their typical velocities of $\\sim$50 km s$^{-1}$ (Kaper et al.\\ \\cite{kap1997}; Chevalier \\& Ilovaisky \\cite{ci1998}) are the natural result of the recoil velocity acquired when the supernova shell leaves the binary system.  Several searches failed to find compact companions of classical runaway stars (Gies \\& Bolton \\cite{gb1986}; Philp et al.\\ \\cite{phi1996}; Sayer, Nice \\& Kaspi \\cite{snk1996}), suggesting that for these systems $\\vec{\\mathv}_\\mathrm{kick}$ of the neutron star was large enough (several 100~km~s$^{-1}$) to unbind the binary (e.g., Frail \\& Kulkarni \\cite{fk1991}; Cordes, Romani \\& Lundgren \\cite{crl1993}; Lai \\cite{lai1999}). The typical magnitude of this `threshold' kick velocity is uncertain (e.g., Hills \\cite{hil1983}; Lorimer, Bailes \\& Harrison \\cite{lbh1997}; Hansen \\& Phinney \\cite{hp1997}; Hartman \\cite{har1997}).  Single BSS runaways must originate in close binaries because these systems have the largest orbital velocities, and therefore they have experienced close binary evolution before being ejected as a runaway. This leads to the following observable characteristics: \\newcounter{it61} \\begin{list}{\\arabic{it61}:}{ \\usecounter{it61} \\setlength{\\topsep}{1mm} \\setlength{\\labelwidth}{3mm} \\setlength{\\labelsep}{2mm} \\setlength{\\itemindent}{0mm} \\setlength{\\parsep}{0mm} \\setlength{\\itemsep}{0mm} \\setlength{\\leftmargin}{5mm}} \\item BSS runaways are expected to have increased helium abundance\\footnote{The abundances of other elements are increased as well, in particular that of nitrogen which can be enhanced by a factor 4--5 (Van Rensbergen priv.\\ comm.)} and large rotational velocity: when the primary fills its Roche lobe, mass and angular momentum is transferred to the star that will become the runaway. The mass transfer stops when the `primary' has become a helium star (i.e., only the helium core remains). This process enriches the runaway with helium, and spins it up (e.g., Packet \\cite{pac1981}; van den Heuvel \\cite{heu1985}; Blaauw~\\cite{bla1993}). \\item A BSS runaway can become a blue straggler, because it is rejuvenated during the mass-transfer period through the fresh fuel it receives from the primary. \\item The kinematic age of a BSS runaway star (defined as the time since the runaway left its parent group) should be smaller than the age of the parent group. The primary of the original binary system first evolves for several Myr before it explodes and the runaway is ejected. \\end{list}  \\paragraph{Dynamical ejection scenario (DES):} In this scenario runaway stars are formed through gravitational interactions between stars in dense, compact clusters. Although binary-single star encounters produce runaways (e.g., Hut \\& Bahcall \\cite{hb1983}), the most efficient interaction is the encounter of two hard binary systems (Hoffer \\cite{hof1983}).  Detailed simulations show that these collisions produce runaways with velocities up to 200~km~s$^{-1}$ (Mikkola \\cite{mik1983a}, \\cite{mik1983b}; Leonard \\& Duncan \\cite{ld1988}, \\cite{ld1990}; Leonard \\cite{leo1991}). The outcome of a binary-binary collision can be (i) two binaries, (ii) one single star and a hierarchical triple system, (iii) two single stars and one binary, and (iv) four single stars (Leonard \\cite{leo1989}). In most cases the collision will result in the ejection of two single stars and one hard binary with an eccentric orbit (Hoffer \\cite{hof1983}; Mikkola \\cite{mik1983a}). Since the resulting binary is the most massive end product of the collision it is unlikely to gain a lot of speed; it might even remain within the parent cluster. This process naturally leads to a low (0--33\\%) runaway binary fraction which is in qualitative agreement with the observations (e.g, Gies \\& Bolton \\cite{gb1986}).  For the DES to be efficient the initial binary fraction in clusters needs to be large. Recent observations show that the binary fraction for massive stars in young clusters is $>$50\\% (Abt \\cite{abt1983}; Kroupa, Petr \\& McCaughrean \\cite{kpm1999}; Preibisch et al.\\ \\cite{pre1999} [approaching~100\\%]).  DES runaways have the following characteristics: \\newcounter{it62} \\begin{list}{\\arabic{it62}:}{ \\usecounter{it62} \\setlength{\\topsep}{1mm} \\setlength{\\labelwidth}{3mm} \\setlength{\\labelsep}{2mm} \\setlength{\\itemindent}{0mm} \\setlength{\\parsep}{0mm} \\setlength{\\itemsep}{0mm} \\setlength{\\leftmargin}{5mm}} \\item DES runaways are formed most efficiently in a high-density environment, e.g., in young open clusters. They may also originate in OB associations. These birth sites of massive stars are unbound stellar groups and therefore expand (e.g., Blaauw \\cite{bla1952a,bla1978}; Elmegreen \\cite{elm1983}; Kroupa \\cite{kro2000a}), so DES runaways must have been ejected very soon after their formation. The kinematic age and the age of the parent association are thus nearly equal. \\item DES runaways are not expected to show signs of binary evolution such as large rotational velocities and increased helium abundance. However, Leonard (\\cite{leo1995}) suggested that some binary-binary encounters produce runaways consisting of two stars that merged during the interaction. These would have enhanced helium abundances and large rotational velocities (Benz \\& Hills \\cite{bh1987}, but see Lombardi, Rasio \\& Shapiro \\cite{lrs1995}). \\item DES runaways are expected to be mostly single stars. \\end{list}  \\TabOne \\bigskip \\noindent Which of the two formation processes is responsible for runaway stars has been debated vigorously. Both mechanisms create stars with large peculiar velocities which enable them to travel far from their parent group: a velocity of 100~km~s$^{-1}$ corresponds to $\\sim$100~pc in only 1~Myr. The relative importance of the two scenarios can be established by (i) studying the statistical properties of the ensemble of runaway stars, or by (ii) investigating individual runaways in detail. The former approach is based on differences in the general runaway characteristics predicted by each scenario.  This requires a large, complete database of runaway stars, which is, to date, unavailable (e.g., Moffat et al.\\ \\cite{mof1998}). Here we therefore follow the latter, individual approach by retracing the orbits of runaway stars back in time. The objects encountered by a runaway along its path (e.g., an open cluster, an association, other runaways, or a neutron star), and the times at which these encounters occurred, provide information about its formation. Evidence for the BSS as the formation mechanism for single runaway stars is to find a runaway and a neutron star (pulsar) which occupied the same region of space at the same time in the past.  Evidence for the DES is to find a common site of origin for the individual components of the encounter, e.g., a pair of runaways and a binary, in a dense star cluster.  The individual approach requires highly accurate positions ($\\alpha$,$\\delta$,$\\pi$) and velocities ($\\mu_{\\alpha\\ast}$,$\\mu_\\delta$,$\\mathv_\\mathrm{rad}$).  Here $\\alpha$ denotes right ascension, $\\delta$ declination, $\\pi$ parallax, $\\mu_{\\alpha\\ast} = \\mu_\\alpha \\cos \\delta$ proper motion in right ascension, $\\mu_\\delta$ proper motion in declination, and $\\mathv_\\mathrm{rad}$ the radial velocity.  The milli-arcsecond (mas) accuracy of Hipparcos astrometry (ESA \\cite{esa1997}) allows specific investigations of the runaway stars within $\\sim$700~pc.  Positions and proper motions of similar accuracy are now available as well for some pulsars through timing measurements and VLBI observations (e.g., Taylor, Manchester \\& Lyne \\cite{tml1993}; Campbell \\cite{cam1995}). The Hipparcos data also significantly improved and extended the membership lists of the nearby OB associations (de Zeeuw et al.\\ \\cite{zee1999}), and of some nearby young open clusters.  The resulting improved distances and space velocities of these stellar aggregates make it possible to connect the runaways and pulsars to their parent group, and, in some cases, to identify the specific formation scenario (Hoogerwerf, de Bruijne \\& de Zeeuw \\cite{hbz2000}; de Zeeuw, Hoogerwerf \\& de Bruijne \\cite{zhb2000}).  Pre-Hipparcos data (e.g., Blaauw \\& Morgan \\cite{bm1954}; Paper I; Blaauw \\cite{bla1993}; van Rensbergen, Vanbeveren \\& de Loore \\cite{rvl1996}) allowed identification of the parent groups for some runaways, but generally lacked the accuracy to study the orbits of the runaways in detail (but see Blaauw \\& Morgan \\cite{bm1954} and Gies \\& Bolton \\cite{gb1986}).  We define a sample of nearby runaways and pulsars with good astrometry in \\S\\ref{sec:samples}, and then analyse two cases in depth: $\\zeta$~Oph and PSR~J1932$+$1059 in \\S\\ref{sec:SNinUS} and AE~Aur, $\\mu$~Col and $\\iota$~Ori in \\S\\ref{sec:orion}. We apply the method developed in these sections to the entire sample of runaways and pulsars in \\S\\ref{sec:class-runaways}, \\S\\ref{sec:new-runaways}, and \\S\\ref{sec:noparent}.  We discuss helium abundances, rotational velocities and the blue straggler nature of runaways in \\S\\ref{sec:helium} and \\S\\ref{sec:blue}, and summarize our conclusions in \\S\\ref{sec:end}.\\looseness=-2 ",
        "conclusions": "\\label{sec:end} We have used the Hipparcos astrometry for a sample of nearby candidate OB runaway stars to locate their parent groups, and to identify their formation mechanisms. We retraced the orbits of these candidate runaways, and determined where and when they passed through a possible parent group. We find that both mechanisms proposed for the production of runaway stars, the binary-supernova scenario (BSS) and the dynamical ejection scenario (DES), operate. Table~\\ref{tab:05} summarizes the results.  Even though the number of runaways discussed in this paper is small, and the weight of evidence varies, we find that roughly $2/3$ of the runaways is produced by the BSS and $1/3$ by the DES. This agrees with the results of binary population synthesis calculations by Portegies Zwart (\\cite{por2000}). At present it is not feasible to extend this study to other runaways. This is mainly due to large uncertainties in the velocities and distances of the runaways, and the limited knowledge of star-forming regions with distances $>$500~pc.  Tracing the runaway orbits back in time provides, for the first time, direct evidence that both scenarios produce single runaway stars (Hoogerwerf et al.\\ \\cite{hbz2000}). The orbit calculations demonstrate that the runaway $\\zeta$~Oph and the progenitor of PSR~J1932$+$1059 once formed a binary system in the Upper Scorpius association, and that the neutron star acquired a kick velocity of $\\sim$350~km~s$^{-1}$ in the supernova explosion.  The runaways AE~Aur and $\\mu$~Col, and the binary $\\iota$~Ori were involved in a dynamical interaction (a binary-binary collision) $\\sim$2.5~Myr ago, which took place in the Trapezium cluster.   \\TabFive  The current investigation is biased towards finding BSS runaways. This is mainly due to the fact that the accuracy of the available data, and our knowledge of the location and motions of star-forming regions, restrict the study to $\\sim$700~pc. The small volume implies that we are only able to identify runaway stars with small kinematic ages of 0--10~Myr (i.e., runaways which recently left their parent association). Runaways which were created at an earlier time have most likely traveled outside our sample limits. Since associations and open clusters can create BSS runaways during $\\sim$50~Myr (approximately the lifetime of a $8~M_\\odot$ star) and DES runaways only in the inital stages when the group still has a high density, we expect to find more BSS than DES runaways because there are relatively many more old parents than young parents in the Solar neighbourhood. This bias is somewhat weakened by the fact that most dynamical interactions produce two runaway stars while the binary-supernova mechanism produces only one.  The creation of runaway stars modifies the mass function of the parent group at the high-mass end, where the total number of stars is small. For example, the encounter in Orion described in \\S\\ref{sec:orion} removed four stars with a total mass of order 70 $M_\\odot$ from the Trapezium cluster, while only six stars more massive than 10~$M_\\odot$ remain.  Derivation of the initial mass function of young stellar groups from the present-day mass function without accounting for the associated runaway stars leads to erroneous results.  Our Hipparcos-based study has identified 56 runaway stars within 700 pc from the Sun, and tripled the subset of these for which a parent group is known (from 6 to 21).  As mentioned in \\S\\ref{sec:samples}, less than a third of the O--B5 stars in the Hipparcos Catalog have a measured radial velocity. Obtaining these is likely to result in another factor of three increase in the size of the sample, so that statistical studies become possible.  The next major step in our understanding of the origin of runaway stars will come when large datasets of micro-arcsecond ($\\mu$as) astrometry and accurate radial velocities (1--2~km~s$^{-1}$) become available.  Distances accurate to a few parsec will allow for a final confirmation or rejection of the genetic link between runaways and their parents (e.g., Figure~\\ref{fig:12}).  These data will become available over the next two decades with the launches of several astrometric satellites (FAME, SIM, GAIA). These aim to obtain $\\mu$as astrometry for a large number of stars, from $10\\,000$ stars with SIM to 1 billion stars with GAIA. Besides astrometry, accurate radial velocities are also required; unfortunately, there is no dedicated effort to obtain these for a large number of O and B stars.  The BSS and DES can produce runaway stars with spectral types beyond B5 (e.g., Kroupa \\cite{kro2000b}; Portegies Zwart \\cite{por2000}). These will be harder to find, as the velocity distribution of the later-type stars in the Galactic disk is broader than for the O--B5 stars, and the fractional production of low-mass runaways is small. Identifying their parent groups is also harder, because these stars may have traveled for much longer times. However, $\\mu$as accuracy astrometry complemented with accurate radial velocities will undoubtedly reveal such objects, and will provide further constraints on the binary fraction and the binary mass-ratios in open clusters and associations.  Figure~\\ref{fig:01}{\\it b} shows that there are 19 additional pulsars within one kpc for which an accurate proper motion is not available. A systematic program to measure these might allow the detection of more examples of pairs such as $\\zeta$~Oph and PSR~J1932$+$1059. It would also improve the characterisation of the pulsar population as a whole. VLBI techniques hold the promise of achieving sub-mas astrometry (positions, proper motions, and parallaxes) in the near future."
    },
    "0010/astro-ph0010261_arXiv.txt": {
        "abstract": "Some progenitor models of gamma-ray bursts (GRBs) (e.g., collapsars) may produce anisotropic jets in which the energy per unit solid angle is a power-law function of the angle ($\\propto\\theta^{-k}$). We calculate light curves and spectra for GRB afterglows when such jets expand either in the interstellar medium or in the wind medium. In particular, we take into account two kinds of wind: one ($n\\propto r^{-3/2}$) possibly from a typical red supergiant star and another ($n\\propto r^{-2}$) possibly from a Wolf-Rayet star. We find that in each type of medium, one break appears in the late-time afterglow light curve for small $k$ but becomes weaker and smoother as $k$ increases. When $k\\ge 2$, the break seems to disappear but the afterglow decays rapidly. Thus, one expects that the emission from expanding, highly anisotropic jets provides a plausible explanation for some rapidly fading afteglows whose light curves have no break. We also present good fits to the optical afterglow light curve of GRB 991208. Finally, we argue that this burst might arise from a highly anisotropic jet expanding in the wind ($n\\propto r^{-3/2}$) from a red supergiant to interpret the observed radio-to-optical-band afterglow data (spectrum and light curve). ",
        "introduction": "An energetic anisotropic jet is supposed to be ejected in an explosion. We assume that such a jet is adiabatic, i.e., the radiative energy is a negligible fraction of the total energy of the jet. This assumption is valid if the energy density of the electrons accelerated by a shock, produced by the interaction of the jet with its surrounding medium, is a small fraction $\\epsilon_e$ of the total energy density of the shocked medium or if most of the electrons are adiabatic, i.e., their radiative cooling timescale is larger than that of the jet expansion (Dai, Huang \\& Lu 1999; B\\\"ottcher \\& Dermer 2000).  \\subsection{Dynamics}  The central question in the next study is how the Lorentz factor $\\gamma$ at angle $\\theta$ within the jet evolves with the observer time $t$ because all other quantities that appear in the observed flux are functions of $\\gamma$ and of the jet radius and medium density. To obtain $\\gamma$, we assume that the kinetic energy per unit solid angle within the jet has the following distribution: \\begin{equation} \\varepsilon(\\theta)\\equiv\\frac{dE}{d\\Omega}=\\left \\{ \\begin{array}{ll} \\varepsilon_0  & {\\rm if}\\,\\, \\theta\\leq\\theta_0 \\\\ \\varepsilon_0\\left(\\frac{\\theta}{\\theta_0}\\right)^{-k} & {\\rm if}\\,\\, \\theta_0<\\theta<\\theta_m, \\end{array} \\right. \\end{equation} where $\\theta_m$ is the half opening angle of the jet. The ejected mass per unit solid angle, $M_{\\rm ej}(\\theta) =M_0$, is assumed to be independent of the angle (at least over some range of angles). A motivation for this assumption is that a different distribution of the ejected mass affects only evolution of the jet at very early times, but at later times the jet expands based on the Blandford-McKee's (1976) self-similar solution during the relativistic phase and the Sedov-Taylor self-similar solution during the non-relativistic phase. Thus, the initial Lorentz factor at angle $\\theta$ can be written as \\begin{equation} \\gamma_0(\\theta)=\\left \\{ \\begin{array}{ll} \\frac{\\varepsilon_0}{M_0c^2}+1 & {\\rm if}\\,\\, \\theta\\leq\\theta_0 \\\\\\left(\\frac{\\varepsilon_0} {M_0c^2}\\right)\\left(\\frac{\\theta}{\\theta_0} \\right)^{-k} +1               & {\\rm if}\\,\\, \\theta_0<\\theta<\\theta_m. \\end{array} \\right. \\end{equation} The evolution of the Lorentz factor $\\gamma(\\theta, r)$ can be calculated from the following equation \\begin{equation} M(r)\\gamma^2+M_0\\gamma=M(r)+M_0+ \\varepsilon(\\theta)/c^2, \\end{equation} where $M(r)$ is the mass of the swept-up medium per unit solid angle. Equation (3) expresses the conservation of energy and it applies to only the case without heating of the ejected material by the reverse shock. In this paper we neglect sideways expansion of the jet. Evidence for this consideration is that an observation of GRB 990123 on 7 February 2000 by HST (Fruchter et al. 2000), combined with the previous observations, indicates that the steepening of the R-band afterglow light curve of this burst during a period of $\\sim 378$ days is still roughly consistent with the edge effect of an ultra-relativistic jet of a fixed opening angle expanding in a homogeneous medium proposed by M\\'esz\\'aros \\& Rees (1999). Further evidence is that, as argued by Jaunsen et al. (2000), the rapid initial decline, the sharp break in the optical light curve, and the spectral properties of the GRB 980519 afterglow are best interpreted as being due to an ultra-relativistic jet of a fixed opening angle expanding in a wind medium with $n\\propto r^{-2}$.  The external medium density is assumed to be a power-law function of radius: $n(r)=Ar^{-s}$, where  $A=n_*\\times 1\\, {\\rm cm}^{-3}$ for an $s=0$ (homogeneous) medium (e.g., ISM), $A=3\\times 10^{35}A_*\\,{\\rm cm}^{-1}$ for an $s=2$  wind possibly from a Wolf-Rayet star as the GRB progenitor (Chevalier \\& Li 1999, 2000), and $A=10^{30}\\bar{A}_*\\,{\\rm cm}^{-3/2}$ for an $s=3/2$ wind possibly from a typical red supergiant star (Fransson, Lundquist \\& Chevalier 1996).  For convenience, $x$ is defined as the radius $r$ scaled to \\begin{equation} r_0=\\left[\\frac{(3-s)\\varepsilon}{m_pc^2A\\gamma_0^2} \\right]^{1/(3-s)}=\\left \\{ \\begin{array}{lll} 2.7\\times 10^{17}\\,{\\rm cm}\\,\\varepsilon_{53} ^{1/3}n_*^{-1/3}\\gamma_{0,2}^{-2/3}\\,\\,  & {\\rm if} \\,\\, s=0\\\\ 2.2\\times 10^{16}\\,{\\rm cm}\\,\\varepsilon_{53}A_*^{-1} \\gamma_{0,2}^{-2}\\,\\,  & {\\rm if}\\,\\, s=2\\\\ 4.6\\times 10^{14}\\,{\\rm cm}\\,\\varepsilon_{53}^{2/3}\\bar{A}_*^{-2/3} \\gamma_{0,2}^{-4/3}\\,\\,  & {\\rm if}\\,\\, s=3/2, \\end{array} \\right. \\end{equation} where $\\varepsilon_{53}=\\varepsilon(\\theta)/10^{53} {\\rm ergs}$ and $\\gamma_{0,2}=\\gamma_0(\\theta)/10^2$. Furthermore, we have \\begin{equation} M(r)=\\frac{m_pn(r)r^3}{3-s}, \\end{equation} where $m_p$ is the proton mass.  The solution of equation (3) is \\begin{equation} \\gamma(\\theta, r)=\\frac{1}{2M}\\left[\\sqrt{M_0^2+ 4M(M+M_0+\\varepsilon/c^2)}-M_0\\right]. \\end{equation} The dimensionless radius $x$ and the time $t'$ in the frame comoving with the jet evolve with the observer time $t$ based on \\begin{equation} \\frac{dx}{dt}=\\frac{c\\beta}{r_0(1-\\beta\\mu)}, \\end{equation} \\begin{equation} \\frac{dt'}{dt}=\\frac{1}{\\gamma(1-\\beta\\mu)}, \\end{equation} where $\\beta=\\sqrt{1-1/\\gamma^2}$, $\\mu=\\cos\\theta$, and the line of sight has been taken to be the jet axis. The solution of equations (7) and (8) combined with (6) gives the dynamics of a sub-jet at angle $\\theta$ within the jet.  \\subsection{Synchrotron Radiation}  We consider synchrotron radiation of the electrons accelerated by the shock. To calculate the spectrum and light curve, one needs to determine three crucial frequencies: the synchrotron self-absorption frequency ($\\nu_a'$), the typical synchrotron frequency ($\\nu_m'$) and the cooling frequency ($\\nu_c'$). We assume a power law distribution of the electrons accelerated by the shock: $dn_e/d\\gamma_e\\propto \\gamma_e^{-p}$ for $\\gamma_e\\ge\\gamma_{m}$, where $p$ is the index of the electron energy distribution and $\\gamma_{m}=[(p-2)/ (p-1)](m_p/m_e)\\epsilon_e\\gamma$ is the minimum Lorentz factor. If $\\epsilon_B$ is assumed to be the ratio of the magnetic to thermal energy densities of the shocked medium,  the magnetic field strength in the shocked medium can be approximated by $B'=[32\\pi\\epsilon_B\\gamma(\\gamma-1)nm_pc^2]^{1/2}$.  Under these assumptions, the typical synchrotron frequency can be written: $\\nu_m'=\\gamma_m^2eB'/(2\\pi m_ec)$. The cooling frequency, $\\nu_c'=\\gamma_c^2eB'/(2\\pi m_ec)$,  is clearly determined by the Lorentz factor $\\gamma_c$, at which an electron is cooling on the dynamical expansion timescale ($t'$, measured in the comoving frame). According to Sari, Piran \\& Narayan (1998), $\\gamma_c=6\\pi m_ec/(\\sigma_TB'^2t')$ where $\\sigma_T$ is the Thomson scattering cross-section. Scaled with the typical values of the involved parameters, these two frequencies, measured in the comoving frame, are derived, \\begin{equation} \\nu_m'=\\left \\{ \\begin{array}{lll} 3.7\\times 10^9\\,{\\rm Hz}\\left(\\frac{\\epsilon_e}{0.1} \\right)^2\\left(\\frac{\\epsilon_B}{0.01}\\right)^{1/2} \\left(\\frac{p-2}{p-1}\\right)^2n_*^{1/2}\\gamma^{1/2} (\\gamma-1)^{5/2}\\,\\,  & {\\rm if}\\,\\, s=0\\\\ 9.5\\times 10^{10} \\,{\\rm Hz}\\left(\\frac{\\epsilon_e}{0.1}\\right)^2\\left(\\frac{ \\epsilon_B}{0.01}\\right)^{1/2}\\left(\\frac{p-2}{p-1} \\right)^2\\varepsilon_{53}^{-1}A_*^{3/2}\\gamma_{0,2}^2 \\gamma^{1/2}(\\gamma-1)^{5/2}x^{-1}\\,\\,  & {\\rm if}\\,\\, s=2\\\\ 3.7\\times 10^{13} \\,{\\rm Hz}\\left(\\frac{\\epsilon_e}{0.1}\\right)^2\\left(\\frac{ \\epsilon_B}{0.01}\\right)^{1/2}\\left(\\frac{p-2}{p-1} \\right)^2\\varepsilon_{53}^{-1/2}\\bar{A}_*\\gamma_{0,2} \\gamma^{1/2}(\\gamma-1)^{5/2}x^{-3/4}\\,\\,  & {\\rm if}\\,\\, s=3/2; \\end{array} \\right. \\end{equation} \\begin{equation} \\nu_c'=\\left \\{ \\begin{array}{lll} 2.8\\times 10^{28}\\,{\\rm Hz}\\left(\\frac{\\epsilon_B}{0.01} \\right)^{-3/2}n_*^{-3/2}\\gamma^{-3/2}(\\gamma-1)^{-3/2} t'^{-2}\\,\\,  & {\\rm if}\\,\\, s=0\\\\ 1.7\\times 10^{24}\\,{\\rm Hz}\\left(\\frac{\\epsilon_B}{0.01} \\right)^{-3/2}\\varepsilon_{53}^3A_*^{-9/2}\\gamma_{0,2} ^{-6}\\gamma^{-3/2}(\\gamma-1)^{-3/2}x^3t'^{-2}\\,\\,  & {\\rm if}\\,\\, s=2\\\\ 2.8\\times 10^{16}\\,{\\rm Hz}\\left(\\frac{\\epsilon_B}{0.01} \\right)^{-3/2}\\varepsilon_{53}^{3/2}\\bar{A}_*^{-3}\\gamma_{0,2} ^{-3}\\gamma^{-3/2}(\\gamma-1)^{-3/2}x^{9/4}t'^{-2}\\,\\,  & {\\rm if}\\,\\, s=3/2, \\end{array} \\right. \\end{equation} where $t'$ is in units of 1 s. The synchrotron self-absorption frequency is usually smaller than $\\nu_m'$ and $\\nu_c'$ a few hours after the burst. In this case, the optical depth due to synchrotron self-absorption at $\\nu_a'$ can be approximated by \\begin{equation} \\tau_{ab}(\\nu_a')\\approx \\frac{5}{3-s}\\frac{enr}{B'\\gamma_p^5} \\left(\\frac{\\nu_a'}{\\nu_p'}\\right)^{-5/3}\\equiv 1, \\end{equation} where $\\gamma_p=\\min(\\gamma_m, \\gamma_c)$ and $\\nu_p'=\\min(\\nu_m', \\nu_c')$ (see, e.g., Panaitescu \\& Kumar 2000). At early times of the afterglow, all the electrons accelerated behind the shock cool in a timescale smaller than the dynamical expansion timescale of the jet at angle $\\theta$, i.e., they are in the fast cooling regime, implying $\\nu_c'<\\nu_m'$, in which case we find the self-absorption frequency, \\begin{equation} \\nu_a'=\\left \\{ \\begin{array}{lll} 0.15\\,{\\rm Hz}\\left(\\frac{\\epsilon_B}{0.01} \\right)^{6/5}\\varepsilon_{53}^{1/5}n_*^{8/5}\\gamma_{0,2}^{-2/5} \\gamma^{6/5}(\\gamma-1)^{6/5}x^{3/5}t'\\,\\,  & {\\rm if}\\,\\, s=0\\\\ 7.5\\times 10^3\\,{\\rm Hz}\\left(\\frac{\\epsilon_B}{0.01} \\right)^{6/5}\\varepsilon_{53}^{-3}A_*^{24/5}\\gamma_{0,2}^6 \\gamma^{6/5}(\\gamma-1)^{6/5}x^{-3}t'\\,\\,  & {\\rm if}\\,\\, s=2\\\\ 1.3\\times 10^{12}\\,{\\rm Hz}\\left(\\frac{\\epsilon_B}{0.01} \\right)^{6/5}\\varepsilon_{53}^{-7/5}\\bar{A}_*^{16/5}\\gamma_{0,2} ^{14/5}\\gamma^{6/5}(\\gamma-1)^{6/5}x^{-21/10}t'\\,\\,  & {\\rm if}\\,\\, s=3/2. \\end{array} \\right. \\end{equation} At later times, $\\nu_m'>\\nu_c'$, in which case the self-absorption frequency is given by \\begin{equation} \\nu_a'=\\left \\{ \\begin{array}{lll} 4.2\\times 10^8\\,{\\rm Hz}\\left(\\frac{\\epsilon_e}{0.1} \\right)^{-1}\\left(\\frac{\\epsilon_B}{0.01}\\right)^{1/5} \\left(\\frac{p-2}{p-1}\\right)^{-1}\\varepsilon_{53}^{1/5}n_*^{3/5} \\gamma_{0,2}^{-2/5}\\gamma^{1/5}(\\gamma-1)^{-4/5}x^{3/5} \\,\\,  & {\\rm if}\\,\\, s=0\\\\ 3.1\\times 10^{10} \\,{\\rm Hz}\\left(\\frac{\\epsilon_e}{0.1}\\right)^{-1}\\left(\\frac{ \\epsilon_B}{0.01}\\right)^{1/5}\\left(\\frac{p-2}{p-1} \\right)^{-1}\\varepsilon_{53}^{-1}A_*^{9/5}\\gamma_{0,2}^2 \\gamma^{1/5}(\\gamma-1)^{-4/5}x^{-1}\\,\\,  & {\\rm if}\\,\\, s=2\\\\ 3.5\\times 10^{13} \\,{\\rm Hz}\\left(\\frac{\\epsilon_e}{0.1}\\right)^{-1}\\left(\\frac{ \\epsilon_B}{0.01}\\right)^{1/5}\\left(\\frac{p-2}{p-1} \\right)^{-1}\\varepsilon_{53}^{-2/5}\\bar{A}_*^{6/5}\\gamma_{0,2}^{4/5} \\gamma^{1/5}(\\gamma-1)^{-4/5}x^{-3/2}\\,\\,  & {\\rm if}\\,\\, s=3/2. \\end{array} \\right. \\end{equation}  The synchrotron peak specific luminosity of a ring ($\\theta\\rightarrow \\theta+d\\theta$) in the comoving frame is given by \\begin{equation} dL_m'=\\left \\{ \\begin{array}{lll} 6.1\\times 10^{29}\\,{\\rm ergs}\\,\\,{\\rm s}^{-1}\\, {\\rm Hz}^{-1}\\left(\\frac{\\epsilon_B}{0.01}\\right)^{1/2} \\varepsilon_{53}n_*^{1/2}\\gamma_{0,2}^{-2} \\gamma^{1/2}(\\gamma-1)^{1/2}x^3\\sin\\theta d\\theta \\,\\,  & {\\rm if}\\,\\, s=0\\\\ 1.6\\times 10^{31}\\,{\\rm ergs}\\,\\,{\\rm s}^{-1}\\, {\\rm Hz}^{-1}\\left(\\frac{\\epsilon_B}{0.01}\\right)^{1/2} A_*^{3/2}\\gamma^{1/2}(\\gamma-1)^{1/2} \\sin\\theta d\\theta\\,\\,  & {\\rm if}\\,\\, s=2\\\\ 6.1\\times 10^{33}\\,{\\rm ergs}\\,\\,{\\rm s}^{-1}\\, {\\rm Hz}^{-1}\\left(\\frac{\\epsilon_B}{0.01}\\right)^{1/2} \\varepsilon_{53}^{1/2}\\bar{A}_*\\gamma_{0,2}^{-1}\\gamma^{1/2} (\\gamma-1)^{1/2}x^{3/4} \\sin\\theta d\\theta\\,\\,  & {\\rm if}\\,\\, s=3/2. \\end{array} \\right. \\end{equation} In the case of  $\\nu_c'<\\nu_m'$, the spectrum is (Sari et al. 1998) \\begin{equation} dL_{\\nu'}'=\\left \\{ \\begin{array}{llll} dL_m'(\\nu_a'/\\nu_c')^{1/3}(\\nu'/\\nu_a')^2\\,\\, & {\\rm if}\\,\\, \\nu'<\\nu_a' \\\\ dL_m'(\\nu'/\\nu_c')^{1/3}\\,\\, & {\\rm if}\\,\\, \\nu_a'<\\nu'<\\nu_c' \\\\ dL_m'(\\nu'/\\nu_c')^{-1/2}\\,\\, & {\\rm if}\\,\\, \\nu_c'<\\nu'<\\nu_m' \\\\ dL_m'(\\nu_m'/\\nu_c')^{-1/2}(\\nu'/\\nu_m')^{-p/2}\\,\\, & {\\rm if}\\,\\,\\nu'>\\nu_m'. \\end{array} \\right. \\end{equation} If $\\nu_c'>\\nu_m'$, the spectrum becomes (Sari et al. 1998) \\begin{equation} dL_{\\nu'}'=\\left \\{ \\begin{array}{llll} dL_m'(\\nu_a'/\\nu_c')^{1/3}(\\nu'/\\nu_a')^2\\,\\, & {\\rm if}\\,\\, \\nu'<\\nu_a' \\\\ dL_m'(\\nu'/\\nu_m')^{1/3}\\,\\, & {\\rm if}\\,\\, \\nu_a'<\\nu'<\\nu_m' \\\\ dL_m'(\\nu'/\\nu_m')^{-(p-1)/2}\\,\\, & {\\rm if}\\,\\, \\nu_m'<\\nu'<\\nu_c' \\\\ dL_m'(\\nu_c'/\\nu_m')^{-(p-1)/2}(\\nu'/\\nu_c')^{-p/2}\\,\\, & {\\rm if}\\,\\,\\nu'>\\nu_c'. \\end{array} \\right. \\end{equation}  The observed total flux density at observed frequency $\\nu$ is given by \\begin{equation} F_\\nu=\\int^{\\theta_m}_0\\frac{dL'_{\\nu\\gamma (1-\\beta\\mu)}}{4\\pi D_L^2\\gamma^3(1-\\beta\\mu)^3}, \\end{equation} where $D_L$ is the luminosity distance to the source. For a flat universe with $H_0=65\\,{\\rm km}\\,{\\rm s}^{-1}\\, {\\rm Mpc}^{-1}$, the luminosity distance to the source $D_L=2c/H_0(1+z-\\sqrt{1+z})=1.8\\times 10^{28}\\,{\\rm cm} [(1+z)/2]^{1/2}[(\\sqrt{1+z}-1)/(\\sqrt{2}-1)]$. ",
        "conclusions": "The model presented in section 2 clearly includes the following assumptions. First, the initial opening angle of a highly anisotropic jet (e.g., $k=3$), $\\theta_m$, was taken to be $0.3$, and the observer's angle ($\\theta_{\\rm obs}$) between the axis line of the jet and the line of sight to be zero. Because $dL_m\\propto \\varepsilon(\\theta)$ in the ultra-relativistic limit, the emission from the jet mainly arises from the shock-accelerated electrons moving along the axis line. Thus, the flux density of the emission is weakly dependent of $\\theta_m$, but strongly depends on $\\theta_{\\rm obs}$. For example, the flux density at $\\theta_{\\rm obs}=0.3$ is about four orders of magnitude smaller than that of the emission at $\\theta_{\\rm obs}=0$. GRB 980425 was likely such an off-axis burst surrounded by an $s=2$ wind medium (Nakamura 1999) because its gamma-ray luminosity of GRB 980425 was also approximately four orders of magnitude smaller than that of GRB  970228/980326 and its X-ray afterglow slowly declined. Second, we considered jets of a fixed opening angle and neglected the effect of sideways expansion based on two observational facts. As shown analytically by Rhoads (1999) and Sari, Piran \\& Halpern (1999), this effect can lead to a sharp break in an afterglow light curve. However, numerical studies by many authors (e.g, Panaitescu \\& M\\'esz\\'aros 1999; Moderski, Sikora \\& Bulik 2000; Huang et al. 2000a, b; Kumar \\& Panaitescu 2000; Gou et al. 2000; Wei \\& Lu 2000) show that the actual break, when two effects such as the equal-time surface and detailed dynamics of the jet are considered, is much weaker and smoother than that predicted analytically. Finally, we used the adiabatic solution for kinematics of the jet. This argument is based on the radiative efficiency of the jet, $f=\\epsilon_e[t_{\\rm syn}^{\\prime -1}/(t_{\\rm syn} ^{\\prime -1}+t_{\\rm exp}^{\\prime -1})]$, defined by Dai et al. (1999). At very early times, the cooling timescale due to synchrotron radiation, $t'_{\\rm syn}$, may be much less than the expansion timescale, $t'_{\\rm exp}$, and $f\\approx 0.1(\\epsilon_e/0.1)$; but at late times, $t'_{\\rm syn}\\gg t'_{\\rm exp}$ and thus $f\\approx 0.1(\\epsilon_e/0.1) (t'_{\\rm exp}/t'_{\\rm syn})\\ll 1$. Therefore, it can be seen that the energy losses due to synchrotron radiation are insignificant during the whole evolution stage of the jet.  Our numerical results  show that for an isotropic jet ($k=0$) expanding in the ISM, there is a sharp break in the light curve at late times, but for an isotropic jet expanding in the wind, the break of the light curve is weaker and smoother.  This result is easily understood in terms of the analytical model of Dai \\& Lu (1998) and Chevalier \\& Li (1999, 2000): at early times of the jet evolution, $\\gamma>\\theta_m^{-1}$, in which stage the jet is spherical-like and thus the decay index of the afterglow $\\alpha_1=-[s/(8-2s)+3(p-1)/4]$ in the slow cooling regime; but at later times, $\\gamma<\\theta_m^{-1}$,  the decay index becomes $\\alpha_2=-[(6-s)/(8-2s)+3(p-1)/4]$ due to the edge effect. These indexes imply the steepening of the light curve $\\Delta\\alpha\\equiv\\alpha_1- \\alpha_2=(3-s)/(4-s)=0.5$ for $s=2$, 0.6 for $s=3/2$, or 0.75 for $s=0$. Therefore, the break in the light curve of the afterglow from a jet expanding in a wind is weaker and smoother than that for a homogeneous medium case. Our results also show that for an anisotropic jet, one break seems to appear in the light curve at small $k$ but becomes weaker and smoother as  $k$ increases. This conclusion is different from that presented by M\\'esz\\'aros et al. (1998), who found that an afterglow decays as one single power-law for any $k$. The reason for this difference is that we took into account the light travel effects related to different sub-jets within the jet but M\\'esz\\'aros et al. (1998) didn't. It is easily understood that an anisotropic jet with a small $k$ can be treated as a quasi-isotropic jet with a half opening angle $\\theta_m$, and thus the temporal decay of the flux density of radiation from such a jet will begin to expedite at $t=t_{\\rm jet}$, which is given by equation (18) for the ISM case. However, for a highly anisotropic jet (e.g., $k=2$ or 3), the contribution of radiation from $\\theta>\\theta_0$ is in fact much smaller than that from $\\theta<\\theta_0$ due to the fact that the energy per unit solid angle ($dE/d\\Omega$) within this jet decreases rapidly with increasing $\\theta$, and thus, the light curve of the resulting radiation should begin to steepen at a very early time, $t\\approx 230(\\varepsilon_{53}/n_*)^{1/3} (\\theta_0/10^{-2})^{8/3}\\,{\\rm s}$, which is also estimated from equation (18) in the ISM case but for $\\theta_m$ being replaced by $\\theta_0$. Therefore, one cannot see a broken light curve at late times for such a highly anisotropic jet.  In summary, the energy per unit solid angle within a realistic jet may be a power-law function of the angle ($\\propto\\theta^{-k}$). Such anisotropic jets are expected in the collapsar model of MacFadyen et al. (2000). We numerically calculated light curves and spectra of the emission from anisotropic jets expanding either in the interstellar medium (ISM) or in the wind medium. We also took into account two kinds of wind: one ($n\\propto r^{-3/2}$) possibly from a typical red supergiant and another ($n\\propto r^{-2}$) possibly from a Wolf-Rayet star. Two of the main results of this work are that ({\\em i}) at very early times, the afterglow from a jet expanding in the ISM brightens rapidly but the afterglow flux for the wind case is approximately a constant, no matter whether the jet is isotropic or anisotropic. Based on this conclusion, future observations led by HETE-2 on very early-time optical afterglows may be able to distinguish between the environments as well as GRB progenitors. ({\\em ii}) In each type of medium, one break appears in the late-time afterglow light curve for small $k$, but becomes weaker and smoother as $k$ increases. When $k\\ge 2$, the break seems to disappear but the afterglow decays rapidly. Therefore, one expects that the emission from expanding, highly anisotropic jets provides an explanation for some rapidly fading afteglows whose light curves have no break. We also presented good fits to the optical afterglow flux data of GRB 991208. Finally, to interpret the observed radio-to-optical-band afterglow data of this burst, we argued that it might arise from a highly anisotropic jet expanding in the wind ($n\\propto r^{-3/2}$) from a red supergiant."
    },
    "0010/astro-ph0010111_arXiv.txt": {
        "abstract": "We have studied the evolution of globular cluster systems (GCS) in elliptical galaxies with a power-law initial GCS mass function (GCMF) ($f(M) \\propto M^{-\\alpha}$) similar to that predicted by some theoretical studies of globular clusters formation and to that of young cluster systems observed in merging galaxies.  We have carried out a survey over a large number of different host galaxies and we have considered different values for the index, $\\alpha$, of the initial power-law GCMF ($\\alpha=1.5,1.8,2.0$); we show the dependence of the main GCS final properties (mean mass and dispersion of the final GCMF, fraction of surviving clusters, radial gradient of the GCMF parameters) on the structure of the host galaxy and on the slope of the initial GCMF.  For a subsample of host galaxies with values of effective masses and radii equal to those determined using observational data for a number of giant, normal and dwarf galaxies our results show that the relation between the final GCMF properties and those of the host galaxies as well as the dependence of the final GCMF parameters on the galactocentric distance within individual galaxies differ from those observed in old GCS: the values of the final GCS mean mass are in general smaller ($4.2\\ltorder \\lmf\\ltorder 5.0$) than those observed and the galaxy-to-galaxy dispersion of $\\lmf$ is larger than that reported by observational analyses. The results are compared with those of a companion paper in which we investigated the evolution of GCS with a log-normal initial GCMF and in which the final GCS properties were perfectly consistent with observations. ",
        "introduction": "In a recent paper (Vesperini 2000) we have investigated the evolution of the main properties of globular clusters systems (hereafter GCS) in elliptical galaxies starting with a log-normal GCS initial mass function. The choice of such a functional form for the initial mass function was motivated  by the fact that all the globular cluster system mass functions (hereafter GCMF; we will indicate a globular cluster system luminosity function by GCLF) observed so far are well fitted by a log-normal distribution; in particular, a log-normal GCMF fits well the observed GCMF of clusters located in the external regions of galaxies where evolutionary processes are unlikely to have significantly altered the initial properties of clusters.  In Vesperini (2000) it was shown that the main final GCS properties resulting from the evolution of GCS with a log-normal initial GCMF and the relations between GCS properties and those of the host galaxies are in very good agreement with those reported by a number of observational analyses.  In this paper we study the evolution of GCS starting with a power-law initial GCMF. The motivation  for considering this functional form for the initial GCMF comes both from  some theoretical investigations of globular clusters formation (see e.g. Elmegreen \\& Efremov 1997, Harris \\& Pudritz 1994) which predict this  shape  for the initial GCMF and from several observational studies of young cluster systems in interacting and merging galaxies (see e.g. Schweizer et al. 1996, Miller et al. 1997, Johnson et al. 1999, Zepf et al. 1999; see also Whitmore 1999 for a review and references therein) showing that these systems are characterized by a power-law luminosity function. In fact, whether the observed power-law luminosity function indeed corresponds to an underlying power-law mass function or it results from the age spread of clusters with a log-normal mass function is matter of debate (Fritze-von Alvensleben 1998, 1999, Carlson et al. 1999, Zhang \\& Fall 1999); moreover a recent analysis of the GCLF of young clusters in the Antennae system seems to show a turnover in the LF which would be therefore better fitted by a two-index power-law (Whitmore et al. 1999).  Most observational analyses, on the basis of  the high luminosities and compact sizes observed, support the idea that the young clusters found in interacting galaxies are globular clusters which will eventually evolve into  systems similar to old globular clusters (see e.g. the results of  Ho \\& Filippenko 1996a, 1996b which strongly suggest that at least the brightest clusters must indeed be globular clusters; see also Carlson et al. 1998, 1999) but  this point has been questioned too: van den Bergh (1995), on the basis of the shape of the GCLF, has claimed that these objects would be more properly classified as young open clusters rather than as young globular clusters.  Several theoretical investigations have shown that evolutionary processes can lead to the disruption of a significant number of clusters and alter the parameters and shape of the initial GCMF (Fall \\& Rees 1977, Fall \\& Malkan 1978, Caputo \\& Castellani 1984, Chernoff, Kochanek \\& Shapiro 1986, Chernoff \\& Shapiro 1987, Aguilar, Hut \\& Ostriker 1988, Vesperini, 1994, 1997, 1998, Okazaki \\& Tosa 1995, Capuzzo Dolcetta \\& Tesseri 1997, Gnedin \\& Ostriker 1997, Murali \\& Weinberg 1997a, 1997b, Ostriker \\& Gnedin 1997, Baumgardt 1998); in particular a power-law initial GCMF can be turned by evolutionary processes into a log-normal GCMF  (or in general into a bell shaped GCMF). On the other hand it is not clear whether evolutionary processes, acting with different efficiency in galaxies with different structures, can turn an initial power-law into a log-normal GCMF with approximately universal parameters and with a weak radial variation of the GCMF parameters within individual galaxies as found in most current observational studies. \\begin{figure} \\centerline{\\psfig{file=vfig1.ps,height=10cm,width=10cm,angle=0}} \\caption{Set of values of the effective mass, $M_e~(M_{\\odot})$, and of the effective radius, $R_e$ (kpc), of the host galaxies considered in the paper.} \\end{figure}  In Vesperini (2000) it has been shown that, for a log-normal initial GCMF, the galaxy-to-galaxy variation of the GCMF parameters resulting from theoretical calculations  is perfectly consistent with the observed trends and that a considerable disruption of clusters does not necessarily give rise to a radial gradient of the mean mass of clusters inconsistent with observations. The goal of this paper is that of determining if the same conclusions hold for the case of GCS with a power-law initial GCMF.  The layout of the paper is the following. In \\S 2 we briefly sketch the method used for the investigation and the initial conditions adopted; in \\S 3 we describe the general results while in \\S 4 we discuss in detail the implications of our results for galaxies for which observational data are available. We summarize our conclusions in \\S 5. ",
        "conclusions": "We have investigated the evolution of the main properties of GCS in elliptical galaxies starting with an initial power-law GCMF similar to that of young cluster systems observed in merging galaxies.  We have considered a large set of different host galaxies and we have shown the dependence of the final parameters of the GCMF, of the fraction of surviving clusters and of the ratio of the final total mass in clusters to the total initial mass of clusters on the effective radius and effective mass of the host galaxy (see Figs 2, 3, and 4). Three different values (close to those determined observationally from the GCLF of young cluster systems) of the index of the initial power-law GCMF have been explored ($\\alpha=1.5,~1.8,~2.0$).  While evolutionary processes can easily turn a GCMF with an initial power-law shape into a GCMF with a log-normal shape like that observed in old cluster systems, our investigation reveals several inconsistencies between the observed properties of old GCS in elliptical galaxies and the theoretical results obtained adopting a power-law initial GCMF. In particular, studying the evolution of GCS in a sample of giant, normal and dwarf elliptical galaxies with effective radii and effective masses equal to those determined observationally, we have shown that: \\begin{itemize} \\item the theoretical values of $\\lmf$ are in general  smaller than those observed; \\item the range spanned by $\\lmf$ is  significantly larger than that reported by observational studies; in particular, for galaxies with $\\lme>10.5$, we have shown that starting with a power-law initial GCMF with $\\alpha=1.8$, $\\lmf$ ranges from 4.2 to 5.0 whereas  observations show an approximately universal value of $\\lmf$ ($\\lmf \\simeq 5.16$ for $M/L_V=2$) with a very small galaxy-to-galaxy dispersion ($\\sim 0.06$); \\item $\\lmf$ varies significantly with the effective mass of the host galaxy; specifically $\\lmf$  tends to increase with decreasing values of $\\lme$ for $\\lme>10$, it reaches a peak and then decreases again for dwarf galaxies; this does not agree with observations which show that for giant galaxies, $\\lmf$ is approximately constant and larger than $\\lmf$ of clusters in dwarf galaxies. \\item adopting a power-law initial GCMF, a significant dependence of $\\lmf$ on the galactocentric distance is produced by the effects of evolutionary processes; this is in contrast with several observational studies which fail to find a significant radial gradient of $\\lmf$ within individual galaxies. \\item Starting with an initial power-law GCMF, values of $\\lmf$ consistent with those observed in massive  galaxies can be obtained only with values of $\\alpha$ smaller ($\\alpha \\simeq 1.2-1.3$) than those observed in young cluster systems. \\end{itemize} In Vesperini (2000) it has been shown that the final properties of GCS obtained starting with a log-normal initial GCMF similar to that observed in the external regions of some elliptical galaxies (where evolutionary processes are unlikely to have significantly altered the initial conditions of clusters)  are in general good  agreement with the observed properties of old cluster systems. In this paper we have shown that, starting with a power-law initial GCMF, the final GCS properties present several discrepancies from those observed in old GCS; this would seem to rule out the possibility that old cluster systems were formed with a power-law initial GCMF similar to that observed in young cluster systems of merging galaxies and to strongly favour a log-normal initial GCMF.  One can not exclude the existence of differences in the process of globular cluster formation at the current epoch and at the time of formation of currently old clusters leading to different initial GCMF. While individual young clusters could indeed be young globular clusters and  those which survive could evolve into systems similar to old globular clusters, the future global properties of these young cluster systems could differ from those of currently old cluster systems."
    },
    "0010/astro-ph0010327_arXiv.txt": {
        "abstract": "In this work we show how galaxy clusters can be used to discriminate among different cosmological models. We have used available X-ray \\& optical cluster data to constrain the cosmological parameters as well as the cluster scaling relations, $T-M$ and $L_x-T$. We also show the power of future SZE data to constrain even more these parameters. ",
        "introduction": "In several works, authors have used different cluster data sets in an attempt to constrain the cosmology. The usuall procedure is, starting from the Press-Schechter (PS) mass function, fit the experimental mass function or by using a given $T-M$ or $L_x-M$ relation build some other cluster functions like the temperature, X-ray luminosity or flux functions, and then compare them with the corresponding data sets. This can be a dangerous process. First, when considering just one data set one ensures that its best fitting model is compatible with just that data set. Some care should be taken to check that the best fitting model is also consistent with other data sets. A second problem comes when the cluster scaling relations $T-M$ or $L_x-M$ are assumed as fixed relations. However, the scatter in these relations are known to be important and they can introduce uncertainties in the final result. In order to avoid all these difficulties we made a fit to different data sets simultaneously without doing any assumption about neither the cosmology nor the $T-M$ and  $L_x-M$ relations. For the $T-M$ relation we assume the free-parameter relation:\\\\ \\begin{equation} T_{gas} = T_0 M_{15}^{\\alpha}(1 + z)^{\\psi} \\end{equation} where $T_0, \\alpha$ and $\\psi$ are our three free parameters. $M_{15}$ is the cluster mass in $h^{-1} 10^{15} M_{\\odot}$ units. And similarly for the $L_x-M$ :\\\\ \\begin{equation} L_x^{Bol} = L_0 M_{15}^{\\beta}(1 + z)^{\\phi} \\end{equation} With these two scalings plus the PS formalism we were able to build the mass function, the temperature function and the X-ray luminosity and flux functions. As can be seen in Diego et al. (2000a) we found that only low density universes ($\\Omega \\approx 0.3$ with or without a cosmological constant) are compatible with recent determinations of these functions. In that work we also obtain some interesting limits for the parameters in the cluster scaling relations. Also important is to mention that by combining different data sets in our fit, we have reduced significantly the degeneracy in $\\sigma_8 - \\Omega$. In fact, we found a clear peak in the probability distribution at the position $\\sigma_8 = 0.8, \\Omega = 0.3$ ",
        "conclusions": ""
    },
    "0010/nucl-th0010049_arXiv.txt": {
        "abstract": "We review recent work on the phase structure of QCD at very high baryon density. We introduce the phenomenon of color superconductivity and discuss how the quark masses and chemical potentials determine the structure of the superfluid quark phase. We comment on the possibility of kaon condensation at very high baryon density and study the competition between superfluid, density wave, and chiral crystal phases at intermediate density. ",
        "introduction": "\\label{sec_intro}  In the interior of compact stars matter is compressed to densities several times larger than the density of ordinary matter. Unlike the situation in relativistic heavy ion collisions, these conditions are maintained for essentially infinite periods of time. Also, compared to QCD scales, matter inside a compact star is quite cold. At low density quarks are confined, chiral symmetry is broken, and baryonic matter is described in terms of neutrons and protons as well as their excitations. At very large density, on the other hand, we expect that baryonic matter is described more effectively in terms of quarks rather than hadrons. As we shall see, these quarks can form new condensates and the phase structure of dense quark matter is quite rich.  At very high density the natural degrees of freedom are quark excitations and holes in the vicinity of the Fermi surface. Since the Fermi momentum is large, asymptotic freedom implies that the interaction between quasi-particles is weak. In QCD, because of the presence of unscreened long range gauge forces, this is not quite true. Nevertheless, we believe that this fact does not essentially modify the argument. We know from the theory of superconductivity the Fermi surface is unstable in the presence of even an arbitrarily weak attractive interaction. At very large density, the attraction is provided by one-gluon exchange between quarks in a color anti-symmetric $\\bar 3$ state. High density quark matter is therefore expected to be a color superconductor \\cite{Frau_78,Barrois:1977xd,Bar_79,Bailin:1984bm}.  Color superconductivity is described by a pair condensate of the form \\beq \\label{csc} \\phi = \\langle \\psi^TC\\Gamma_D\\lambda_C\\tau_F\\psi\\rangle. \\eeq Here, $C$ is the charge conjugation matrix, and $\\Gamma_D, \\lambda_C,\\tau_F$ are Dirac, color, and flavor matrices. Except in the case of only two colors, the order parameter cannot be a color singlet. Color superconductivity is therefore characterized by the breakdown of color gauge invariance. As usual, this statement has to be interpreted with care because local gauge invariance cannot really be broken. Nevertheless, we can study gauge invariant consequences of a quark pair condensate, in particular the formation of a gap in the excitation spectrum.  In addition to that, color superconductivity can lead to the breakdown of global symmetries. We shall see that in some cases there is a gauge invariant order parameter for the $U(1)$ of baryon number. This corresponds to true superfluidity and the appearance of a massless phonon. We shall also find that for $N_f>2$ color superconductivity leads to chiral symmetry breaking and that quark matter may support a kaon condensate. Finally, as we move to stronger coupling we find that other forms of order can compete with color superconductivity and that quark matter may exist in the form of chiral density waves or chiral crystals. ",
        "conclusions": ""
    },
    "0010/astro-ph0010488_arXiv.txt": {
        "abstract": "The objective of this work is to report on  the influence of muon interactions on the development of air showers initiated by astroparticles. We make a comparative study of the different theoretical approaches to muon bremsstrahlung and muonic pair production interactions. A detailed algorithm that includes all the relevant characteristics of such processes has been implemented in the AIRES air shower simulation system. We have simulated ultra high energy showers in different conditions in order to measure the influence of these  muonic electromagnetic interactions. We have found that  during the late stages of the shower development (well beyond the shower maximum) many global observables are significantly modified in relative terms when the mentioned interactions are taken into account. This is most  evident in the case of the electromagnetic component of very inclined showers. On the other hand, our simulations indicate that the studied processes do not induce significant changes either in the position of the shower maximum or the structure of the shower front surface. ",
        "introduction": "When an ultra high energy astroparticle interacts with an atom of the Earth's atmosphere, it produces a shower of secondary particles that continues interacting and generating more secondary particles. The study of the characteristics of air showers initiated by ultra high energy cosmic rays is of central importance. This is due to the fact that in our days such primary particles cannot be detected directly; instead, they must be studied from different measurements of the  air showers they produce.  We have been studying the physics of air showers for several years. We started working on the topic of the electromagnetic processes in air showers analyzing the modifications in the shower development  due to the  reduction of the electron bremsstrahlung and electron pair production by the Landau-Pomeranchuk-Migdal effect and the dielectric suppression \\cite{PRD}; and we also studied the influence of the geomagnetic field in an air shower \\cite{GF}.  The main goal of this work is to analyze other radiative processes that take place during the development of an ultra high energy air shower. We have  studied the processes of muon bremsstrahlung,  muonic pair production (electron and  positron) and muon-nucleus interaction.  At high energies these processes become important and dominate the energy  losses of the energetic muons that are present in an air showers.  The mentioned mechanisms  are characterized by small cross sections, hard spectra, large energy fluctuations  and generation of electromagnetic sub-showers for the case of muon bremsstrahlung and muonic pair production, and hadronic sub-showers for the case of muon-nucleus interaction. As a consequence,  the treatment of such energy losses as uniform and continuous processes is for many purposes inadequate \\cite{PDG}.  We have studied the three mentioned processes concluding that the muon nucleus interaction has less probability than the other ones to  produce hard events, and therefore, to generate sub-showers and introduce  significant modifications in the air shower development. For this  reason, this work will be primarily focused on studying the consequences of the purely electromagnetic  processes, namely, muon bremsstrahlung and muonic pair production.  In order to analyze the influence  of muon bremsstrahlung and  muonic pair  production on the air shower observables, we have  developed new procedures for these mechanisms and have incorporated them in AIRES (AIRshower Extended Simulations) \\cite{Aires,Aires2}. Using  the data generated with the AIRES code, we have studied the changes introduced by those processes on the different physical quantities.  This work is organized as follows: in section II we briefly  review the  theory of muon bremsstrahlung, muonic pair production and muon-nucleus interaction. At the end of this section we compare the three effects and analyze under which conditions they can modify  the development of air showers. In section III we show the results of our simulations. Finally  we present our conclusions and comments in section IV. ",
        "conclusions": "We have studied the influence of the MBR and MPP in the development of air showers initiated by ultra high energy astroparticles. We have incorporated in the AIRES air shower simulation system the corresponding procedures to emulate these effects and  have then performed simulations in a  number of  different initial conditions.  The analysis of the evolution of a single muon indicates that such particle can eventually generate  secondary showers after undergoing hard  MBR and MPP processes. This indicates clearly that these interactions  cannot be simulated accurately as continuous energy loss processes.  For $3 \\times 10^{20}$ eV proton and iron primaries the main modifications  introduced by MBR and MPP  affect the  electromagnetic component of the showers. The number and energy of gammas and electrons increase significantly in the  late stages of the shower development  (well beyond the shower maximum), but the mentioned effects do not  generate visible changes in the position of $X_{max}$.  The changes generated by MBR and MPP for muon observables are less significant: The number of muons practically does not change and their energies diminish about 3 \\% (1.5 \\%) for the case of proton (iron) showers  at the tail of the shower.  The shower front arrival time profile does not present modifications due to the MBR and MPP processes.  For primary energies below $3\\times 10^{20}$ eV the modifications in the electromagnetic shower induced by MBR and MPP are qualitatively similar to the ones described in the preceding paragraphs. In the case of muon observables like $\\xi_{\\mu}$ we have found, in the entire range of primary energies that was considered, small variations due to MBR and MPP. Such small modifications correspond, in general, to decrements in the average muon energies when MBR and MPP are switched on. However, critical combinations of event probabilities determine that  $\\xi_{\\mu}$ can remain unchanged or be slightly increased for primary energies around $10^{15}-10^{17}$ eV.  The fact that the alterations in the electromagnetic showers are only significant in the late stages of the development of the showers, i.e., $X_s>X_{max}$, implies that in normal conditions there will be no visible changes in the electromagnetic shower at ground level, for showers with zenith angle less than $45^{\\circ}$.  For showers with zenith angles larger than $60^{\\circ}$, the MBR and MPP processes are responsible for an increment of the density of electromagnetic particles at ground, which is most important in a narrow region around the shower axis. In connection with this result, we have also found that the signal produced by  \\u{C}erenkov detectors will be also larger near the shower axis if the mentioned effects are taken into account."
    },
    "0010/astro-ph0010441_arXiv.txt": {
        "abstract": "Fluctuations of the gamma-ray background (GRB) are investigated.  It is assumed that the GRB is produced by discrete sources. Two basic classes of objects as dominant contributors to the GRB are considered: a) blazars and b) starburst galaxies. Predicted $\\log N - \\log S$ counts are calculated for these two models. In both cases strong cosmic evolution is needed to produce a substantial fraction of the observed GRB. Since on the average blazars are several orders of magnitude more luminous than starburst galaxies, their number required to saturate the GRB is much smaller than the number of starburst galaxies. In effect, fluctuations of the GRB in the blazar model are relatively high. It is shown that moderate improvement in the measurement accuracy of the GRB flux will allow us to discriminate between these models. ",
        "introduction": "Using the present detection techniques we are unable to define a true nature of the diffuse extragalactic gamma-ray background. The status of the GRB is now analogous to that of the X-ray backgrund in 1960s and 70s. Although some classes of objects are known $\\gamma$-ray emitters, it is by no means certain what fraction of the GRB is produced by any particular classes of objects and even by the discrete sources in general. Furthermore, it seems likely that some fraction of the diffuse emission considered previously to be of the extragalactic origin is produced in the halo of our Galaxy \\citep{strong99}. \\cite{dar00} postulate that most if not all of the GRB is produced in the Galactic halo.  One should expect that structural features of the GRB are related to its origin. It is likely that trully diffuse processes produce the GRB of smooth distribution without strong well defined fluctuations. On the contrary, small scale variations would indicate the discrete nature of the background. In the present paper we concentrate on a class of models which assume that the whole GRB is produced by a well defined population of discrete sources. We discuss two cases in detail. The source counts are estimated and prospects for the detection of individual sources by the present-day and the near future instruments are assessed. Next we estimate amplitude of the GRB fluctuations in various angular scales characteristic to both these models. ",
        "conclusions": "Simple, discrete source models of the GRB have been constructed to investigate effects of the background anisotropies at different angular scales. Source counts predicted in the balazar and starburst galaxy models are distinctly different. We have shown that precise measurements of the GRB using of the INTEGRAL detectors would allow to distinguish between these two cases. Isotropic GRB would indicate that the large number of faint sources contributes to the GRB making the starburst model more likely, while detectable GRB fluctuations would point to the blazar model."
    },
    "0010/astro-ph0010394_arXiv.txt": {
        "abstract": "We mapped a $13 \\times 22$ pc region in emission from 492 GHz [\\ci] and, for the first time, 551 GHz \\thcof\\ in the giant molecular cloud M17SW. The morphologies of the [\\ci] and \\thco\\ emission are strikingly similar. The extent and intensity of the [\\ci] and \\thcof\\ emission is explained as arising from photodissociation regions on the surfaces of embedded molecular clumps.  Modeling of the \\thcof\\ emission in comparison to \\thcoo\\ indicates a temperature gradient across the cloud, peaking to at least 63 K near the M17 ionization front and decreasing to at least 20 K at the western edge of the cloud.  We see no correlation between gas density and column density.  The beam-averaged column density of \\ci\\ in the core is $1\\times10^{18}$ \\cmsqr, and the mean column density ratio $N$(\\ci)/$N$(CO) is about 0.4.  The variations of $N$(\\ci)/$N$(CO) with position in M17SW indicate a similar clump size distribution throughout the cloud. ",
        "introduction": "M17SW is a giant molecular cloud at a distance of 2.2 kpc \\citep{chi80} illuminated by the M17 OB association about 1 pc to the east. Studies of molecular excitation \\citep{sne84}, molecular line profiles \\citep*{mar84}, and maps of molecular emission \\citep{stu90} from the M17SW core have all concluded that the dense gas is highly clumped. Early photochemical models of molecular cloud/\\hii\\ region interfaces predicted an overlying layer of ionized carbon and a thin underlying layer of neutral carbon within a photodissociation region (PDR) between the ionized and molecular gas components \\citep*[e.g.,][]{lan76a,tie85}. Given the edge-on geometry of M17SW and the \\hii\\ region, observations of [\\ci] and [C~{\\sc ii}] emission along cuts through the interface region unexpectedly detected these species more than a parsec into the molecular cloud \\citep{kee85,gen88,stu88}.  \\citeauthor{gen88} and \\citeauthor{stu88} concluded that far-ultraviolet (FUV) radiation penetrates the interclump medium and excites PDRs on the surfaces of illuminated clumps.  Subsequent multi-component PDR models of the cloud have accounted for the intensity of the emission from numerous far-infrared and submillimeter atomic and molecular lines \\citep*{bur90,mei92}.  \\citet{sek99} present observations of [\\ci] emission extending more than a degree southwest of the M17SW cloud core.  The dual radiometers on board the {\\it Submillimeter Wave Astronomy Satellite} (\\swas) provide the ability to simultaneously observe the emission from the $J = 5\\rightarrow4$ rotational transition of \\thco\\ at 550.926 GHz and the $^3P_1\\rightarrow{^3P_0}$ fine-structure transition of \\ci\\ at 492.161 GHz.  The roughly 4\\arcmin\\ beamsize of \\swas\\ affords us the opportunity to map large regions in the \\thcof\\ and [\\ci] lines and study the large-scale physical conditions in and structure of the dense interstellar medium.  This is especially important for the \\thcof\\ line, since few maps are available in the mid-$J$ \\thco\\ transitions beyond the small-scale strip maps of \\citet{gra93}.  Whereas the critical density ($n_{\\rm cr} \\sim 2\\times10^3$ \\cmcube) and upper-level energy ($E_u = 24$ K) of [\\ci] enable it to be easily excited wherever neutral carbon is present in the dense ISM, the \\thcof\\ line ($n_{\\rm cr} \\sim 2\\times10^5$ \\cmcube; $E_u = 79$ K) will preferentially probe warmer, denser regions.  We obtained with \\swas\\ the first large-scale map of the \\thcof\\ line in M17SW, as well as the [\\ci] line over the same region.  We combine these results with millimeter \\thco\\ observations to probe the physical conditions in the extended cloud and the spatial and density structure of the bulk of the molecular gas. ",
        "conclusions": "\\subsection{Temperature, Density, and Column Density Structure}  We used the \\thcof\\ and \\thcoo\\ integrated intensities together with an iterative algorithm incorporating the Large Velocity Gradient (LVG) radiative transfer approximation \\cite[e.g.,][]{sco74} to model the \\thco\\ column density $N$(\\thco), H$_2$ number density $n$(H$_2$), and kinetic temperature $T_K$.  In the LVG modeling we used collision rates for CO with para-H$_2$ \\citep{flo85}.  Substituting ortho-H$_2$ rates resulted in insignificant differences in the model results. For each position, we began with initial estimates of $T_K = 10$ K, $N$(\\thco) appropriate for LTE, and $n$(H$_2$) from \\citet{sne84} for the core or $n$(H$_2$) = $1 \\times 10^5$ \\cmcube\\ elsewhere \\citep{wil99}. We used these estimates as input for the LVG algorithm, together with velocity widths derived from the \\thcoo\\ data, to produce model \\thcof\\ and \\thcoo\\ integrated intensities.  We iterated the model with density as the free parameter until the model \\thcof/\\thcoo\\ integrated intensity ratio reproduced the observations.  If the $J \\leq 5$ level populations thermalized with too low a ratio (i.e. further increases in density no longer affected the line ratio), we incremented $T_K$ and reiterated until the observed ratio was reached.  We then iterated the model with $N$(\\thco) as the free parameter until the absolute intensities of the lines agreed with observations.  This value of $N$(\\thco) was then used again in the iteration with $n$(H$_2$) to check that line opacity effects did not change the line ratio, and the entire loop repeated as necessary until the model line intensities and ratios agreed with the observations. This procedure then yields the minimum temperatures necessary to explain the line ratios.  The model results show clear evidence that the molecular cloud is heated by the \\hii\\ region/ionization front at its eastern edge. In Figure~\\ref{fig2} we show the temperature distribution over the region mapped in \\thcoo\\ as well as the derived \\thco\\ column density distribution.  The peak beam-averaged column \\hbox to\\columnwidth{density is $6.1\\times10^{16}$ \\cmsqr, with a mean value of $1.2\\times10^{16}$}  \\centerline{\\includegraphics[width=0.92\\hsize,clip]{m17_ci_fig3.ps}} \\figcaption{({\\it Top}) The linear correlation of the integrated intensity ($T_{\\rm MB}$ scale) of [\\ci] with \\thcoo\\ at each position in M17SW.  The dashed line traces the least-squares linear fit to the data (slope 0.84).  The dotted line plots the equivalent fit for the Orion A cloud \\citep[slope 0.5;][]{ike99}.  ({\\it Bottom}) The derived ratio of the column densities of \\ci\\ to CO [where $N$(CO) $\\equiv 50N$(\\thco)] is plotted against the \\thco\\ column density at each position in M17SW.  Errorbars are $\\pm$ 1$\\sigma$ as described in the text.  The dashed line marks the mean value of $N$(\\ci)/$N$(CO) (0.37) for all data points.  The triangles plot the value of $N$(\\ci)/$N$(CO) in bins containing equal numbers of \\thco\\ molecules, with the abscissa at the center of each binning range. \\label{fig3}} \\centerline{ } \\centerline{ }  \\noindent \\cmsqr.  The column density map appears virtually identical to the \\thcoo\\ integrated intensity map (Fig.~\\ref{fig1}), a property attributable to the low opacity of the \\thco\\ lines ($\\tau \\lesssim 0.6$ for either line).  The highest gas temperatures are found along the eastern edge of the cloud, peaking at about 60~K near the cloud core.  The mean cloud temperature is 35~K, similar to the global cloud temperature of 30~K derived by \\citet{wil99}.  The temperature we derive for the core region is also consistent with the temperatures derived from  millimeter observations of optically thin CH$_3$CCH emission \\citep{ber94}.  Our observations apparently are insensitive to the high temperature gas ($T_K \\geq 200$ K) responsible for CO(7$\\rightarrow$6) line emission observed in the \\hii/H$_2$ interface region, however this gas is estimated to comprise only about 5 percent of the total column density \\citep{har87} and would contribute only about 10 percent of the observed peak \\thcof\\ intensity.  We derive molecular gas densities in the range 1--6 $\\times 10^5$ \\cmcube\\ over the entire region mapped, with a median value of $3\\times10^5$ \\cmcube, about half as large as densities derived from multi-transition CS observations of the core region of M17SW \\citep{sne84,wan93}. In contrast to the well-ordered column density and temperature distribution across the M17SW cloud, the density distribution appears almost random.  In particular, there is no correlation between the gas density and the column density.  Instead, high- and low-density regions are peppered randomly throughout the region.  Since the excitation of the \\thcof\\ emission depends both on the temperature and density, with higher densities requiring lower temperatures for the same $J = 5$ fractional population, we investigated whether there was any trend between $T_K$ and $n$(H$_2$) as an artifact of the modeling algorithm. We see no correlation between the gas temperature distribution and the density distribution in the model results.  Although the model distributions of temperature and density are uncorrelated, the densities we derive are sensitive to the model cloud temperature.  For example, uniform cloud temperatures of 50~K and 100~K yield median cloud densities of $4\\times10^4$ and $1\\times10^4$ \\cmcube, respectively.  The model column densities are insensitive to temperature, decreasing by only 20 percent if $T_K = 100$~K.  Since such large column densities of hot gas are difficult to produce throughout the cloud, and since our model mean cloud temperature is already somewhat higher than the \\citet{wil99} global cloud temperature, we consider $4\\times10^4$ \\cmcube\\ as a reasonable lower limit to the mean cloud density.  The modeling results for the density and column density of the molecular gas in M17SW require that the gas is highly clumped.  We derive the volume filling factor of the gas by dividing the ratio of the peak H$_2$ column density ($4\\times10^{22}$ \\cmsqr, assuming a \\thco\\ abundance of $1.5\\times10^{-6}$ relative to H$_2$) to the mean H$_2$ number density ($3\\times10^5$ \\cmcube) by the linear extent of the core (3 pc core size, deconvolved from the \\swas\\ beam).  From these quantities, we derive a volume filling factor of 0.02. \\citet{stu90} derive toward the central 13\\arcsec\\ of the core a volume filling factor of 0.13 for gas with densities greater than $10^5$ \\cmcube, reflecting a higher concentration of clumps at the core peak region.  \\subsection{The Correlation of {\\rm [C~{\\sc i}]} and \\thco\\ Emission}  The large extent of the [\\ci] emission in M17SW and its similarity to isotopic CO emission was first noted by \\citet{kee85} and \\citet{gen88} along linear cuts through the ionization front/molecular cloud interface. The large-scale [\\ci] and \\thco\\ maps in Figure~\\ref{fig1} show that this similarity extends over the entire region mapped.  We plot in the top panel of Figure~\\ref{fig3} the observed correlation of the integrated intensities of the [\\ci] and \\thcoo\\ emission.  We converted the data to the main-beam temperature scale $T_{\\rm mb}$ to remove differences in telescope beam efficiencies from the observed relation \\citep[$\\eta_{\\rm mb} = 0.52$ for the FCRAO data;][]{wil99}.  The observed trend is well fit (linear least-squares, correlation coefficient 0.95) by the relation $\\int T_{\\rm mb}dV$([\\ci])\\ $ = (0.84\\pm0.02)\\int T_{\\rm mb}dV$(\\thcoo) $-$ $(0.4\\pm0.4)$.  The slope of this relation is similar to the slope observed between [\\ci] and \\thcot\\ in the Orion Bar region by Tauber et al.\\ (1995; $\\sim 0.8$), but is somewhat higher than the relation reported by Ikeda et al.\\ (1999; $\\sim 0.5$) for the Orion A cloud as a whole, and for the ensemble of clouds W3, L1630, S140, and Cep A \\citep{plu99}.  We propose a possible explanation for this in \\S3.3.  We used the \\swas\\ [\\ci] observations to model the \\ci\\ column density $N$(\\ci) in a manner similar to the \\thco\\ LVG modeling, using the collision rates of \\citet{sch91}.  Since the spatial distribution and velocity profiles of the \\ci\\ and \\thco\\ lines are nearly identical, we used the temperatures and densities derived from the \\thco\\ observations, along with the \\thcoo\\ linewidths, as fixed input parameters to the LVG algorithm.  We then iterated the models with $N$(\\ci) as the sole free parameter until the model [\\ci] intensities matched the observations.  $N$(\\ci) ranges from $1.1\\times10^{16}$ \\cmsqr\\ to $1.1\\times10^{18}$ \\cmsqr\\ with a median value of $1.7\\times10^{17}$ \\cmsqr.  The [\\ci] line is optically thin, with a maximum opacity less than 0.6 at the core.  Our peak value of $N$(\\ci) is about half that obtained by \\citet{sek99} in a 2\\arcmin\\ beam. The spatial distribution of $N$(\\ci) is nearly identical to the distribution of $N$(\\thco) shown in Figure~\\ref{fig2}.  The derived \\ci\\ column densities show an interesting behavior when compared to the \\thco\\ column densities.  In the bottom panel of Figure~\\ref{fig3} we plot the ratio of $N$(\\ci) to the CO column density $N$(CO) (calculated assuming a CO/\\thco\\ abundance ratio of 50) as a function of \\thco\\ column density for each model position.  Errorbars reflect the random errors of the [\\ci] and \\thcoo\\ intensities, but not systematic uncertainties in the LVG models.  Whereas at the low column densities at the edges of the molecular cloud the $N$(\\ci)/$N$(CO) ratio ranges from 0.1 to 1.0, at high column densities in the core of the cloud the lower and upper bound for the range asymptotically approach the mean $N$(\\ci)/$N$(CO) ratio for all positions (0.37).  This is similar to the behavior exhibited for the Orion A cloud, except the mean column density ratio is much lower, only about 0.10 in the Orion core regions \\citep{plu00}.  The behavior of the $N$(\\ci)/$N$(CO) ratio is difficult to understand in the context of chemical models where the neutral carbon is produced {\\it in situ} in the molecular gas \\citep*[e.g.,][]{lan76b,pin92}, and would thereby predict a more or less constant column density ratio.  Likewise, simple plane parallel PDR models \\citep*[e.g.,][]{lan76a,tie85,hol91} predict a rising value of $N$(\\ci)/$N$(CO) with decreasing molecular gas column density, but not the increased scatter we observe in the ratios.  Remarkably, summing the $N$(\\thco)-ordered data of Figure~\\ref{fig3} into bins of equal total number of \\thco\\ molecules (in this case, $2\\times10^{17}$ \\cmsqr\\ $\\times$ $A_{\\rm beam}$, where $A_{\\rm beam}$ is the area subtended by the \\swas\\ beam at the distance of M17SW), and then computing $N$(\\ci)/$N$(CO) for the binned data, shows that $N$(\\ci)/$N$(CO) is independent of the mean \\thco\\ column density for each bin (see Fig.~\\ref{fig3}).  The resolution to the behavior of $N$(\\ci)/$N$(CO) lies in the structure of the M17SW molecular cloud.  \\subsection{PDR Emission and the Structure of M17SW}  Theoretical models of PDRs are quite successful at predicting the observed \\thcof\\ and [\\ci] line intensities and the column density of \\ci for the physical conditions in the region mapped by \\swas.  Based on the arguments of \\citet{stu88} and \\citet{mei92}, we expect the incident FUV field within the cloud to range from $\\sim10$ $G_0$ at the northwestern edge to $4\\times10^4$ $G_0$ at the ionization front located about 1\\farcm5 northeast of the map reference position, where $G_0$ is in units of the Habing flux.  Our LVG modeling indicates a fairly narrow range of densities, 1--$6\\times10^5$ \\cmcube.  PDR models predict a column density $N$(\\ci) $\\sim 4\\times10^{17}$ \\cmsqr\\ insensitive to variations in $G_0$ and only weakly dependent on density \\citep{tie85}.  This is similar to the mean value of $N$(\\ci) derived for the M17SW cloud.  For the peak $N$(\\ci) we derive, only a few PDRs along the line of site through the core are required.  \\citet{kau99} present updated homogeneous PDR models covering the range of densities and FUV fluxes appropriate for M17SW, and have recently incorporated \\thco\\ into the models for comparison with \\swas\\ observations.  These models successfully predict the same range of \\thcof\\ integrated intensities, for the densities and FUV fluxes we find in M17SW, as those observed by \\swas.  The PDR models of \\citet{kos94} also predict the peak \\thcof\\ line temperature we observe, for $G_0 = 10^4$ and $n$(H$_2$) $= 10^5$ \\cmcube.  The \\citet{kau99} models match the observed [\\ci] integrated intensities for regions of the cloud with $G_0 < 100$, but are a factor of about 3--4 too low for the core of the cloud.  Again, this is easily explained by requiring several PDRs along the line of sight through the core.  The extent of the [\\ci] emission and the relationship of $N$(\\ci) with the molecular gas column density require not only that the gas is clumpy, but that the distribution of clump sizes is similar throughout the cloud, from the core out to the periphery.  In a clumpy cloud penetrated by FUV radiation, PDRs arise on the surfaces of the dense clumps, with a thin shell of \\ci\\ encompassing the CO in the clump. Consider unresolved clumps of similar density, as is the case for the \\swas\\ observations \\citep[see ][]{stu90}.  The beam-averaged column density $N$(\\ci) depends on the clump surface area while $N$(CO) depends on the volume, so the ratio $N$(\\ci)/$N$(CO) varies as the inverse size of the clumps.  Thus, one might expect for a cloud with a range of clump sizes a large scatter in the column density ratio, as we see in the low column density regions where only a few clumps are probably subtended by the \\swas\\ beam.  As the beam samples larger column densities, i.e. more clumps in the beam, the clump size distribution is more completely sampled and the scatter in $N$(\\ci)/$N$(CO) decreases, approaching the value appropriate for the mean surface area to volume ratio of the clump size distribution.  By summing up the low column density regions we increase the clump sample size, and we see from Figure~\\ref{fig3} that the $N$(\\ci)/$N$(CO) ratio for equal total gas masses gives the same value of $N$(\\ci)/$N$(CO).  Thus, the size distribution of clumps must be the same throughout the cloud.  The diameter of a spherical clump with a \\ci/CO number ratio of 0.37 and H$_2$ density $3 \\times 10^5$ \\cmcube\\ is 0.04 pc, where we assume all carbon is either in \\ci\\ or in CO and calculate the thickness of the \\ci layer from \\citet{hol91}.  This diameter scales inversely with the H$_2$ density.  The lower $N$(\\ci)/$N$(CO) ratio found in Orion is probably a consequence of a larger mean clump size, since $N$(\\ci) is insensitive to changes in density and FUV flux.  Since the intensity of optically thin lines scales directly with column density, observationally a lower $N$(\\ci)/$N$(CO) ratio produces a lower slope in the linear correlation of the [\\ci] line intensity to the \\thcoo\\ intensity (e.g, Fig.~\\ref{fig3})."
    },
    "0010/astro-ph0010507_arXiv.txt": {
        "abstract": "We use SPH/N-body simulations to study the ram-pressure stripping of the gas of dwarf galaxies orbiting in a disc galaxy. The effects of the gaseous disc and halo on the satellites are analysed and the results are compared with the observations of the dwarf spheroidals (and dwarf irregulars) in the Local Group. The stripped gas is compared with the High Velocity Clouds. Finally, a model of evolution of the dwarfs is proposed. ",
        "introduction": "Some satellite dwarf galaxies in the Local Group (LG) have low or null gas content (Mateo 1998). However, nearly half of them are (dynamically) associated with large reservoirs of atomic gas, not always placed at the dwarf center (e.g., Carignan 1999). A correlation between the dwarf gas content and its distance to the center of the main galaxy (in the LG) exists (Blitz \\& Robishaw 2000). The gas reservoirs of the dwarfs and the High Velocity Clouds (HVCs) also seem to be related, as they both have similar properties (Blitz \\& Robishaw 2000; Braun \\& Burton 1999, 2000).  In this paper, we study the evolution of a dwarf irregulars (dIrrs) into dwarf spheroidal (dSph), when the ram-pressure removes the gas of the dIrrs as they orbit in a disc galaxy. This gas can be identify as a HVC. We also present a semi-analytical method to simulate the stripping process. ",
        "conclusions": "The evolution of a gaseous dwarf galaxy depends on its density:  (a) High density satellites suffer slow evolution and marginal gas losses. That can explain the existence of satellites (as the LMC) with gaseous tails.  (b) Low density satellites lose their gas after a few pericenters, whereas a large percentage of the stellar (and dark) material remains bound to the dwarf, supporting the evolution from dIrrs into dSphs when they orbit in a disc galaxy. It can also explain the apparent correlation between the dwarf gas content and its distance to the primary.  The fate of the stripped gas depends on the distribution of the gas in the primary:  (a) If the gas of the primary is in the halo, the stripped gas of the dwarf forms condensations that can be identified as HVCs.  (b) If the gas is in the disc, the satellite gas is stripped as it crosses the disc and it finishes orbiting in it. It can feed the star formation in the disc.  In a general case, both process (stripping by a halo and by a disc) coexist, increasing the gas loss rate of the dwarf."
    },
    "0010/astro-ph0010289_arXiv.txt": {
        "abstract": "We discuss the possibility of explaining the extraordinary, correlated X-ray/TeV flares observed during April 1997 from Mrk 501, by synchrotron-pair cascades injected by synchrotron radiation from ultra-high energy protons and muons. Evaluating the jet conditions required to explain the observed features of the flares, the allowed region for this model in the parameter space of jet magnetic field and Doppler factor is identified, and compared with the parameter choices of other, both hadronic and leptonic models presented in the literature. The present model requires magnetic fields similar to other hadronic models for gamma-ray blazars ($B\\gsim 30\\G$), but a lower Doppler factor ($D\\approx 3$). ",
        "introduction": "In April 1997 the Beppo-SAX team observed X-ray flares from Mrk\\,501 \\cite{Pian98}, which are extraordinary in at least two respects: (a) they extend to energies beyond $200\\keV$, and (b) they show an extremely flat spectrum, in one case with a power law flux index $\\alpha_{\\rm x}\\approx 0.5$. Simultaneously, the Whipple and HEGRA Cherenkov telescopes observed correlated flares in the TeV band \\cite{TeV-501April}. Integrating over a larger time window, both found that the high energy emission during this period has extended on average up to at least $20\\TeV$ \\cite{TeVspec}, but with a significant curvature consistent with an exponential cutoff at ${\\sim}\\,5\\TeV$. The most common way to explain this emission is the synchrotron-self Compton (SSC) model, which naturally expects correlated variability because both the X-ray and the TeV component are radiated by the same population of particles (electrons). Another appealing possibility seems, to explain the high energy emission as synchrotron radiation from ultra-high energy (UHE) protons, while the X-rays are produced by co-accelerated electrons \\cite{MP99ff,Aha00}.  However, both models share the problem to explain the hard X-ray spectral index, which requires an electron injection spectrum $dN_e/dE \\propto E^{-1}$ \\cite{Pian98} (this is because synchrotron cooling steepens the stationary electron spectrum by one power compared to the injection spectrum). However, the Fermi shock acceleration mechanism, commonly assumed to energize the particles in the jets, cannot produce such hard spectra (the limit is $dN/dE \\propto E^{-1.5}$ \\cite{Malkov}), and convincing alternatives for electron acceleration to TeV energies in jets have not been suggested jet.  It has been pointed out previously \\cite{Rac99} that correlated variability can also find a natural explanation by considering the TeV and X-ray emission as different generations of a synchrotron-pair cascade, injected by UHE protons. This is a modification and extension of the proton-induced cascade (PIC) models \\cite{Man93ff}, considering the synchrotron emission of UHE protons and muons additionally to UHE gamma-ray injection by $\\pi^0$-decay. The $p/\\mu$ synchrotron component leads here to ``narrow'' cascades with an extremely flat spectrum, which peak in the TeV and X-ray regime. It has been shown that the asymptotic spectral indices predicted by this model for the X-ray and TeV cascade generations fit quite well to the indices measured by Beppo-SAX and HEGRA/Whipple, respectively \\cite{Rac99}, and this for particle injection spectra which are canonically expected from Fermi acceleration. In contrast to pure proton-synchrotron and SSC models this model assumes that the jet is moderately optically thick at TeV energies, to allow for a reprocessing of the power into the X-ray regime.% \\footnote{Despite occasional, contrary claims in the literature this is not in conflict with observations, since in a homogeneous emitter $\\gamma\\gamma$ absorption only induces a steepening of the power-law spectrum by the target photon index $\\alpha_{\\rm t}$, rather than an exponential cutoff \\cite{Sve-MKB,Rac99}. The reader may keep this in mind when comparing the jet Doppler factors derived here with common ``lower limits'', which all assume that the jet emission is optically thin for all observed energies.} In the following we shall derive the allowed region in the jet-parameter space to explain the observed photon energies in the April 1997 flares from Mrk 501 within this scenario, and compare with the parameter choice of other models. A detailed fit to the flare spectra, employing the full set of particle and photon transport equations will be presented elsewhere (Rachen \\& Mannheim, in preparation). ",
        "conclusions": "\\Fig{par} shows the conditions discussed above as limiting lines in the parameter space of jet magnetic field $B$ and Doppler factor $D$. We see that the condition $t_{p\\gamma} < t_{p,\\rm syn}$ together with Eqs.~\\refeq{TeV} and \\refeq{opacity} define a small allowed region in the parameter space, close to the ``star-point'' of equal cooling times defined by Rachen \\& M\\'esz\\'aros \\cite{RM98}. Other conditions turn out to be less restrictive for the present case. The result of M\\\"ucke \\& Protheroe \\cite{MP99ff}, who also considered muon-synchrotron radiation and cascade reprocessing in their MC simulations and found both effects insignificant, can thus be understood by their parameter choice (marked as point $<$A$>$ in \\fig{par}), which implied $t_{p,\\rm syn}\\ll t_{p\\gamma}$ and $\\eps_{\\gamma\\gamma}>10\\TeV$. We note that the jet Doppler factor determined here ($D\\approx 3$) is signficantly lower than those assumed in both pure proton-synchrotron and SSC models.  The small allowed parameter region would also explain why flares like those seen in April 1997 are rather rare events. We emphasize that explaining TeV radiation alone by hadronic phenomena, like PIC or proton-synchrotron radiation, is possible for a much larger set of parameters. Also correlated variability may arise from less restrictive settings, for example by correlated acceleration of protons and electrons in the jet \\cite{MP99ff}. Finally, the fact that the different models are so clearly separated in the parameter space opens the chance to distinguish between them by determining the jet Doppler factor and magnetic field {\\em independent} of the assumed radiative mechanism.  \\begin{small}  \\vspace{8pt}  \\noindent {\\bf Acknowledgments.} The author is indebted to Karl Mannheim for his contributions to this research, and to the EU-TMR network ``Astro-Plasmaphysics'' (ERBFMRX-CT98-0168) for support.  \\end{small}  \\vspace{-18pt}"
    },
    "0010/astro-ph0010240_arXiv.txt": {
        "abstract": "We present spatially extended gas kinematics at parsec-scale resolution for the nuclear regions of four nearby disk galaxies, and model them as rotation of a gas disk in the joint potential of the stellar bulge and a putative central black hole. The targets were selected from a larger set of long-slit spectra obtained with the {\\sl Hubble Space Telescope} as part of the Survey of Nearby Nuclei with STIS (SUNNS). They represents the 4 galaxies (of 24 ) that display symmetric gas velocity curves consistent with a rotating disk. We derive the stellar mass distribution from the STIS acquisition images adopting the stellar mass-to-light ratio normalized so as to match ground-based velocity dispersion measurements over a large aperture. Subsequently, we constrain the mass of a putative black hole by matching the gas rotation curve, following two distinct approaches. In the most general case we explore all the possible disk orientations, alternatively we constrain the gas disk orientation from the dust-lane morphology at similar radii. In the latter case the kinematic data indicate the presence of a central black hole for three of the four objects, with masses of $10^7 - 10^8$M$_\\odot$, representing up to $\\sim 0.025$\\% of the host bulge mass. For one object (NGC~2787) the kinematic data alone provide clear evidence for the presence of a central black hole even without external constraints on the disk orientation. These results illustrate directly the need to determine black-hole masses by differing methods for a large number of objects, demonstrate that the variance in black hole/bulge mass is much larger than previously claimed, and reinforce the recent finding that the black-hole mass is tightly correlated with the bulge stellar velocity dispersion $\\sigma$. ",
        "introduction": "\\label{sec:SMBH_intro}  Over the last few years evidence has mounted that supermassive black holes (SMBH) are nearly ubiquitous in galactic centers. This finding has resulted from high-resolution observations of various kinematical tracers of the central gravitational potential, such as stars, ionized gas, or water masers (see, e.g., Ho 1999 for a review). Additional support for this picture is provided by spectroscopic surveys demonstrating that weak active galactic nuclei (AGN) activity is common in nearby galaxies, especially among early-type systems (E-Sb; Ho, Filippenko, \\& Sargent 1997a and references therein).  An important fraction of these nuclei further exhibit broad-line emission and/or compact X-ray or radio sources that are the hallmarks of quasar activity, presumably powered by accretion onto a black hole (e.g., Ho, et al. 1997b; Terashima, Ho, \\& Ptak 2000; Nagar et al. 2000).  If SMBHs are indeed an integral part of galaxies and their formation, a natural question is whether the black-hole mass $M_\\bullet$ is related to larger properties of the host galaxy.  A correlation between $M_\\bullet$ and stellar mass of the spheroidal component $M_{bulge}$ was suggested by Kormendy (1993), and later quantified by Kormendy \\& Richstone (1995), Magorrian et al. (1998), and Ho (1999). As investigated recently by Kauffmann \\& Haehnelt (2000), such a correlation might arise naturally in hierarchical scenarios for galaxy formation, and is expected if local SMBHs are the dormant relics of past quasar activity.  Evidence of a close connection between galaxy structure and SMBHs has been bolstered by recent reports of a correlation between $M_\\bullet$ and the velocity dispersion of the surrounding stellar bulge (Ferrarese \\& Merritt 2000, hereafter FM00; Gebhardt et al. 2000a, hereafter G00).  The largest compilation of SMBH mass estimates obtained with a single technique, namely axisymmetric two-integral models for the central stellar kinematics, is that of Magorrian et al. (1998).  For most of the 32 early-type galaxies of their sample, they found that {\\sl Hubble Space Telescope} ({\\sl HST}) photometric and ground-based kinematical data were consistent with the presence of a central supermassive black hole. Based on their data they claim that $M_\\bullet \\simeq 0.006 $M$_{bulge}$ with a scatter of only a factor of $\\sim 3$.  While the restrictive orbital assumptions in their modeling can lead to an overestimate of the black-hole mass (Magorrian et al. 1998), their high detection rate nonetheless underscores the apparent prevalence of SMBHs and their close connection to the surrounding stellar bulge.  Despite this recent progress, the demography of SMBHs is far from complete. Most of the reported SMBHs have been detected in bulge-dominated systems, and it is not clear if the $M_\\bullet/M_{bulge}$ relation holds also in the case of disk-dominated galaxies.  This issue is important in light of theories for bulge formation via secular evolution of disk structures (e.g., Norman, Sellwood, \\& Hasan 1996). In addition, it is still not fully understood whether the apparent lack of low-mass central black holes in very massive galaxies is due to selection effects, in which case the correlated distribution of points in the $M_\\bullet/M_{bulge}$ plane actually represents only an upper envelope. In Seyfert nuclei and quasars, measurements of $M_\\bullet$ derived from reverberation mapping experiments provide intriguing evidence that the average $M_\\bullet/M_{bulge}$ ratio in active galaxies is systematically lower than in normal ones (Ho 1999; Wandel 1999; Kaspi et al. 2000, Gebhardt et al. 2000b), even though this discrepancy may disappear in the $M_\\bullet/\\sigma$ ratio (Gebhardt et al. 2000b). Independent determinations of $M_\\bullet$ in active galaxies are desirable for confirmation of these results.  Black-hole mass estimates and the corresponding Eddington luminosity $L_{Edd}$ are also of fundamental importance for testing theories of accretion physics in AGNs, since models typically predict a strong dependence of accretion disk structure on $\\dot M/M_\\bullet$, or alternatively $L/L_{Edd}$ (e.g., Narayan, Mahadevan, \\& Quataert 1998).  In order to address these and other issues we initiated a Survey of Nearby Nuclei with STIS (SUNNS) on {\\sl HST}, with kinematic information obtained from measurements of nebular emission lines. Because of its collisional nature, gas promises a conceptually easier way to trace the gravitational potential of a galaxy nucleus than stars, because nearly circular orbits can be assumed; at the same time, however, gas is more susceptible to non-gravitational forces and may not be in equilibrium. Use of emission-line tracers to constrain a central mass concentration is thus limited to cases where strong evidence exists that the gas dynamics are dominated by gravity.  The paper is organized as follows. In \\S 1 we present the spectroscopic and photometric STIS observations, and in \\S 2 we describe our modeling of the ionized gas kinematics.  We present our results in \\S 3 and draw our conclusions in \\S 4. ",
        "conclusions": "\\label{sec:SMBH_discus}  The results of this study are important in several ways for the understanding of SMBHs in galaxies.  For all four galaxies analyzed here, we find qualitative and/or quantitative evidence for the existence of SMBHs at their centers.  For three sources we obtained well-determined values of $M_\\bullet$; for the fourth object, NGC~4203, the likely presence of a SMBH is signaled by very broad ($\\pm $3000 km s$^{-1}$) H$\\alpha$ emission (Shields et al. 2000). An important point therefore is that {\\sl all} of the sources examined here apparently harbor a SMBH, reinforcing indications that such objects are an integral part of most galaxies.  We emphasize that our four objects were selected from the SUNNS sample on the basis of the angular extent ($>0.5\\arcsec$) and symmetry of their velocity curves, not on the existence of a steep central velocity gradient or other criteria indicative of a SMBH. We note also that the majority of objects with kinematic measurements of $M_\\bullet$ to date are elliptical galaxies (FM00, G00, and references therein), so that the present results for four S0 galaxies represent a substantial increase in the census of SMBHs in bulges of disk galaxies.  The findings described here are also important for comparison with suggested trends relating $M_\\bullet$ to galaxy bulge properties.  For our four galaxies, we obtained bulge $B$-band luminosities using $V$-band disk--bulge decompositions from Baggett, Baggett, \\& Anderson (1998), and $B-V_e$ colors within the galaxy effective radius, with the results listed in Table \\ref{tab:SMBH_in_bulges_DBdecomposition}. Bulge masses were estimated using the mean $B$-band $\\Upsilon = 8.90$ ($M/L$)$_\\odot$ from van der Marel (1991), rescaled to $H_0 = 75$ km s$^{-1}$ Mpc$^{-1}$. For comparison with FM00 and G00, we also calculated stellar velocity dispersions $\\sigma_e$ and $\\sigma_{e/8}$ within the bulge effective radius $R_e$ and within $R_e/8$ respectively, using values of $\\sigma$ from Table \\ref{tab:SMBH_in_bulges_basic_parameters}, seeing-corrected measurements of $R_e$ from Baggett et al. (1998), and the algorithm of Jorgensen, Franx, \\& Kjaegaard (1995).  \\placetable{tab:SMBH_in_bulges_DBdecomposition}  Our results for $M_\\bullet$ as a function of bulge properties are shown in relation to previous work in Figure \\ref{fig:SMBH_in_bulges_fig5}. Compared with the $M_{bulge} - M_\\bullet$ correlation reported by Magorrian et al. (1998), our constrained results yield black-hole masses that are lower by a factor of $\\sim 3$ or more at a given $M_{bulge}$ while two out of four of our unconstrained 3$\\sigma$ upper limits equally fall below the $M_{bulge} - M_\\bullet$ relation, consistent with other suggestions that the two-integral estimates of $M_\\bullet$ by these authors are systematically too large (e.g., van der Marel 1998). When compared with published analyses employing more robust measures of SMBH masses, our results show better agreement, as can be seen in the diagram of $M_\\bullet$ versus $L_{bulge}$ from Ho (1999), which derived $M_\\bullet/L_{bulge}$ ratio translate in a $M_\\bullet/M_{bulge}\\simeq0.002$ quite similar to the value first inferred by Kormendy \\& Richstone (1995). The constrained fits in particular are in good agreement with the latter correlation, although the upper limit for NGC~4203 continues to suggest a non-negligible degree of scatter in the $M_\\bullet$ versus $L_{bulge}$ relation.  \\placefigure{fig:SMBH_in_bulges_fig5}  The relationship between $M_\\bullet$ and $\\sigma$ is of particular interest in light of recent reports by FM00 and G00 of a very significant correlation with little intrinsic scatter.  Estimates of stellar velocity dispersion employed by these authors correspond to apertures large enough to avoid the influence of a black hole on $\\sigma$ itself. Even if taken {\\it per se}, our data show no correlation between M$_{bulge}$ and any of the other quantities considered so far, it is worth noticing that while going from the unconstrained to the constrained procedure, the black-hole mass determinations move closer to the M$_{\\bullet}$ {\\it vs.} L$_{bulge}$, $\\sigma_{e}$, and $\\sigma_{e/8}$ relations.  Our results, and the constrained fits in particular, are in good agreement with the quantitative relations given in both papers.  These findings provide added support for the strength of an $M_\\bullet - \\sigma$ trend, and its persistence in bulges associated with disk galaxies.  As a final comment, we note that while the present work provides an illustration of the power of gas kinematics for finding and quantifying the masses of SMBHs, it also shows that a significant investment of observing time and effort is necessary in order to yield results. In particular, the SUNNS survey demonstrates that only a modest fraction ($\\la 25$\\%) of nearby galaxies displaying emission-line nuclei in ground-based observations can be expected with {\\sl HST} to show spatially resolved nebular emission with regular kinematics suitable for estimation of $M_\\bullet$.  Determination of the gas velocity field with high confidence requires 2-D mapping with multiple slit positions (e.g., Barth et al. 2000), which is expensive in terms of telescope time. The alternative of employing long-slit observations at a single PA can be employed with some effort to extract useful estimates of $M_\\bullet$, as demonstrated by the present work.  In particular the analysis reported here further illustrates the power of using dust morphology as an indicator of gas disk orientation, in the absence of full 2-D velocity data."
    },
    "0010/astro-ph0010076_arXiv.txt": {
        "abstract": "We present $V$ and $I$ CCD photometry of suspected low-surface brightness dwarf galaxies detected in a survey covering $\\sim2.4$ deg$^{2}$ around the  central region of the Dorado group of galaxies. The low-surface brightness galaxies  were chosen  based on their sizes and magnitudes at the limiting isophote of 26.0V$\\mu$. The  selected galaxies have magnitudes brighter than $V\\approx20$ ($M_{V}\\approx-11$ for  an assumed distance to the group of 17.2 Mpc), with central  surface  brightnesses $\\mu_{0}>22.5$ V mag/arcsec$^{2}$, scale lengths $h>2$\\arcsec, and diameters $\\ge14$\\arcsec at the limiting isophote. Using these criteria, we identified 69 dwarf galaxy candidates. Four of them are large very low-surface  brightness galaxies that were detected on a smoothed image, after masking high  surface brightness objects. Monte Carlo simulations performed to estimate  completeness, photometric uncertainties and to evaluate our ability to detect  extended low-surface brightness galaxies show that the completeness fraction is, on  average, $>80$\\% for dwarf galaxies with $-17<M_{V}<-10.5$ and  $22.5<\\mu_{0}<25.5$ V mag/arcsec$^{2}$, for the range of sizes considered by us ($D\\ge$ 14\\arcsec). The  $V-I$ colors of the dwarf candidates vary from -0.3 to 2.3 with a peak on $V-I=0.98$,  suggesting a range of different stellar populations in these galaxies. The projected surface density of the dwarf galaxies shows a concentration towards the group center similar in extent to that found around five X-ray groups and the elliptical  galaxy NGC1132 studied by Mulchaey \\& Zabludoff (1999), suggesting that the dwarf galaxies in Dorado are probably physically associated with the overall potential  well of the group. ",
        "introduction": "Dwarf galaxies are the most common type of galaxies in the local universe. In addition, they are thought to be the single systems with the largest dark-matter contents, with M/L ratios as high as that of groups and poor clusters (e.g. Carignan and Freeman 1988) and hence their spatial distribution and mass spectrum may give us important insights into the spatial scales over which mass is distributed. However, they are also the hardest galaxies to observe, due to their low-surface brightnesses and low luminosities. Only recently it has become possible to obtain deep wide-field CCD {\\em photometry} of significant numbers of these galaxies, mostly in clusters. {\\em Spectroscopy} of such low luminosity systems is still a challenge even with large-class telescopes, unless emission lines are present.  Dwarf galaxies in nearby clusters and groups have been studied in detail with photographic material (e.g.  Coma: Thompsom \\& Gregory 1993; Virgo: Sandage, Binggeli and Tammann 1985, Impey et al. 1988; Fornax: Ferguson 1989; Bothun et al. 1991; groups: Ferguson and Sandage 1991). From these works two general trends were uncovered: 1) clusters and rich groups usually have a luminosity function with a faint end that is steeper, i.e., has more low-luminosity galaxies, than that for poorer groups and 2) Dwarf elliptical and dwarf spheroidal galaxies (dE/dS0) seem to dominate the faint-end of the number counts in clusters while the dwarf-irregular population (dIrr) are more representative in poor groups and in the field.  Point (1) was well illustrated by Ferguson \\& Sandage (1991) who found a faint-end slope (fit to a Schecter function) of $\\alpha=-1.3$ in the composite luminosity function of seven nearby groups (the sample included the Dorado group) with a tendency for the richer groups to present a larger ratio of dwarf to giant galaxies than the poorer groups.  Point (2) has been challenged by observations of the most well known group of all, our own Local Group. Despite being a poor group with just over 30 members, the Local Group has a larger number of dE/dS0 than of dIrr (Pritchet \\& van den Bergh 1999).  The majority of the dwarf spheroidal galaxies are concentrated around M31 and the Milky Way with a slope of the faint-end of the luminosity function (fit to a Schechter function) of $\\sim -1.3$ while the dwarf irregulars are less concentrated with a flat luminosity distribution (van den Bergh 1999, Grebel 2000). This seems to contradict the results  obtained for the CfA groups and field survey where the largest contribution to the steep faint-end was from the irregular galaxies (Marzke et al. 1998). Recent results coming from CCD studies in nearby groups up to $D=10$ Mpc show an increasing number of new dwarf members (e.g. Jerjen et al. 2000) with a spatial distribution similar to that found in the Local Group (i.e the dwarf spheroidals concentrated around the brightest galaxies while the irregulars distributed throughout the group).  This is the first paper of a series dedicated to survey nearby groups with  the aim of identifying low-surface brightness dwarf galaxies (hereafter LSBD)  down to $M_{V}\\sim -11$. Our primary goal is to study the physical properties  and luminosity distributions of the LSBD and ultimately determine the  luminosity function of the groups as a function of their richness and other  group parameters.  Here we report the results of the CCD photometry of the dwarf galaxy candidates of the nearby group of galaxies  Dorado. A further study will analyze other nearby groups of galaxies.  The Dorado group is  a loose concentration of galaxies centered at $\\sim 4^{h} 15^{m}$,  $-55^{\\circ} 41{'}$ (epoch J2000).  It is also known as Shk 18 (Shakhbazian 1957),  G16 (de Vaucouleurs 1975) and HG3 (Huchra \\& Geller 1982).  Dorado was first identified by de Vaucouleurs (1975) as a ``large and complex nebulae'' in the region of Dorado. In 1982, Huchra \\& Geller (1982), measured velocities for 18 galaxies in the group. Maia et al. (1989) extended the number of known members to 46. Finally Ferguson \\& Sandage (1990, 1991), in a study of the galaxy properties in seven nearby groups identified 34 possible members of Dorado with magnitudes $B_{T} > 19$ in photographic plates. The central part of the group is dominated by two almost equally bright early-type systems: NGC1549 (an E1 galaxy with $V_{r}=1247$ km/s) and NGC1553 (an S0 galaxy with $V_{r}=1280$ km/s).  No detailed distance determination has been made for Dorado and therefore we will adopt a group distance of 17.2 h$^{-1}$ Mpc, where $h=H_{0}/75$. This was determined for the central galaxy NGC1549 based on the D$_n$-$\\sigma$ relation from Faber et al. (1989).  For the determination of the absolute magnitudes of the galaxies we chose the value of $H_{0}=75$ km s$^{-1}$ Mpc$^{-1}$. The group covers an area on the sky of $\\sim$ 10\\arcdeg $\\times$ 10\\arcdeg ($\\sim$ 3$\\times$3 h$^{-1}$ Mpc).  This paper is arranged as follows. Section 2 describes the observations, the data reduction and the methodology used to detect, classify and to obtain photometric parameters for galaxies in the frames. In that section we  also describe the Monte-Carlo simulations performed in order to estimate  the completeness for the LSBD population, and to determine the distribution  of detected central surface brightness and scale lengths to be expected if Local Group and Virgo-like LSBD were present in the Dorado group. In section 3 we present the results and analyze the central surface brightness, scale length, spatial and  color distributions of the LSBD candidates. Finally, in section 4, a general  discussion and a brief summary of the main results are presented.  The luminosity function of the group will be determined, along with those for other groups, in a later publication. ",
        "conclusions": "\\subsection{Colors and color profiles}  The colors and color profiles of the LSBD may give us information on the stellar content and other physical parameters as a function of radius (for example, metallicity). The color distribution of the LSBD candidates of the Dorado group range from $-0.3 < V-I < 2.3$ with a peak at $V - I= 0.98$ (Figure \\ref{fig9}). The color range and peak value are in a good agreement with the values obtained for the LSBD in the Virgo and Fornax clusters (Impey et al. 1988, Bothun et al. 1991) and in the Pegasus and Cancer spiral-rich clusters (O'Neil et al. 1997). About half the LSBD population of Dorado is composed of blue galaxies with $V-I<1$ and 12\\% are very blue with $V-I<0.5$. These values are comparable, in percentage, to those found by O'Neil et al. (1997) in the Pegasus and Cancer nearby groups. As pointed out by these authors, the blue LSBD are important because of the restrictions imposed on their star formation histories and also because they could be the local counterparts of the distant faint blue galaxies.  The average flat color profiles found for the LSBD in Dorado shown in section 3.3 suggest that there is no star formation or, if it exists, it is very weak.  The very blue colors in some galaxies is restricted to the central regions and they can be explained by the existence of a sporadic burst with a constant, but very low, star formation rate. The star formation decreases or disappears in the outer parts of the galaxies.  Rich clusters like Coma do not contain a large population of blue LSBD objects most probably because the evolution of these galaxies is accelerated and they could not survive a long time in such a hostile environment. The color distribution of the Coma dwarf galaxies peaks at $B-R = 1.4$ and at fainter magnitudes they become bluer, to a mean $B-R$ of $1.15$ (Secker et al. 1997).  These two peaks are much redder than the peak in the Dorado group ($V-I=0.98$, corresponds to a typical $B-R\\sim0.5$). This is expected, since the Dorado group has a very low density environment of bright galaxies.  Blue LSBD have been identified and are well studied in a number of nearby groups and in the field (e.g. McGaugh \\& Bothun 1994, de Block et al. 1995, Impey et al. 1996, Pildis et al. 1997, O'Neil et al. 1997). However, we still lack good information on red LSBD galaxies since most of the previous works were done with photographic plates. Practically all of the LSBD observed in photographic surveys are blue since photographic data are biased towards finding blue objects. About 12\\% of the LSBD sample in this work shows colors $V-I>1.5$ mag. For some galaxies the colors are very red (e.g. 779-01, 4536-01 and 3330-10 with $V-I>1.8$) with a flat color profile (see Fig. \\ref{fig11}). The red colors found for these galaxies could not be explained by photometric errors (the errors are $<0.3$ mag, see section 2.6), by interstellar absorption (the color correction is less than 0.02 mag) or by internal reddening (there is no evidence for a high dust content in the LSBD galaxies, O'Neil et al. 1997).  Detection of very red LSBD galaxies in nearby poor groups is not new. Recently, in a survey of low-surface brightness galaxies in the Pegasus and Cancer clusters taken by O'Neil et al. (1997), a significant population of red LSBD galaxies was discovered. About 20\\% of the galaxies in their sample have colors $V-I>1.5$. The percentage of red galaxies in Dorado is similar ($\\sim$ 12\\%). One probable scenario to explain the colors of these galaxies, discussed by O'Neil et al. (1997), is that these objects ``underwent starburst early in their existence, consuming most of the galaxies gas and leaving them with an old stellar population''.  Some examples of very red LSBD ($V-I>1.8$) in the Dorado group are 779-01, 4536-01 and 3330-10. Due to the poor resolution of the CTIO image, it was impossible for us to obtain a morphological classification for 779-01 and 4536-01.  However, the 3330-10 galaxy was classified as a dIrr galaxy. This galaxy is a special case because it is very red ($V-I\\sim2.2$) and shows an irregular morphology, with a large disc and strong bulge.  A number of very red low surface brightness galaxies detected by 0'Neil et al. (1997) were recently surveyed in HI by O'Neil et al. (2000). They showed that a few of them are gas-rich low-surface brightness galaxies with no massive star formation and with $M_{HI}/L_{B}>9$. We have no information about the HI content of the 3330-10 galaxy and we can only speculate that this galaxy could have properties similar to those of the galaxies discovered by O'Neil et al. (2000) in their survey. Additional HI observations are needed to confirm this point.  \\subsection{Projected number density of the LSBD galaxies}  The clustering properties of the dwarf galaxy population is an important tool for cosmology. The space distribution of the dwarf satellites around bright galaxies could indicate how luminous and dark material are distributed at different scales. Moreover it may be a tracer of the dynamical evolution of the dark halos around bright galaxies.  As determined in Section 3.4, the central galaxy concentration dies off outside a distance to the group center of $\\sim$ 250 h$^{-1}$ kpc. The radial distribution (Fig. \\ref{fig12}) is in general agreement with the results obtained by Mulchaey \\& Zabludoff (1999) for five X-ray groups, lthough they cannot be compared in detail given the different magnitudes considered in this and their studies (we consider galaxies with $-16< M_V <-11$ and they considered galaxies with $-17 < M_V <-15$ for H$_0$ $=$ 75 km s$^{-1}$ Mpc $^{-2}$ and $V-R\\sim0.6$).  It is well known that dwarf elliptical galaxies show a stronger clustering around early-type galaxies (e.g. Vader \\& Sandage 1991) than around late-type galaxies (e.g. Lorrimer et al. 1994) and are more common in large and rich environments. The Dorado group has two bright early-type galaxies (one E1 and one S0) in the center, separated by only 58 h$^{-1}$ kpc, and has no other prominent early-type object in its outskirts. In a radius of $\\sim0.5$ h$^{-1}$ Mpc from the central galaxies, only four galaxies with absolute magnitudes brighter than M$_V = -20$ are present, and they are all late-type objects. In order to analyze if the LSBD galaxies detected in Dorado group are associated to the bright central early type NGC1549 and NGC1553 galaxies of Dorado or to the potential well of the group, we plotted the distribution of all LSBD detected in our survey around three different positions: {\\it 1)} the E1 galaxy NGC1549, {\\it 2)} the centroid of the two central group galaxies and {\\it 3)} the S0 galaxy NGC1553. We found no significant difference between these three plots. The central concentration detected suggests the galaxies are true dwarf galaxies of the Dorado group but we cannot decide from the exercise above if they belong to the group or to the two central early-type galaxies.  To further test this point, we compared our result with those obtained for satellites of isolated bright elliptical galaxies.  The study of Vader \\& Sandage (1991) found that the density of satellites, if fit to a function $\\Sigma \\propto r^{-\\beta}$, with $\\beta = -1.22$. This agrees with our value of $-1.09$.  Furthermore, the distribution of satellites around the center of Dorado is similar in extent to that found around the the X-ray bright elliptical galaxy NGC 1132 studied by Mulchaey and Zabludoff (1999).  These comparisons would seem to suggest that the LSBD galaxies in the Dorado group are most probably physically associated with the two central galaxies. However, since these galaxies strongly dominate the potential well of the group, this may be equivalent to saying that the satellites we measured are associated with the overall potential of the group.  An alternative interpretation of these results would be that isolated elliptical galaxies and groups have similar distributions of satellites around them perhaps because all bright field elliptical galaxies may be merged groups (as may be the case for NGC 1132 discussed by Mulchaey and Zabludoff 1999).  \\subsection {LSBD, tidal interacting feature or Galactic cloud?}  We discovered an extended low-surface brightness ``cloud'' $\\sim$ 10\\arcmin (about 50 h$^{-1}$ kpc) to the NW of the central NGC1549 galaxy (galaxy L02 in Table \\ref{tab3}, see Section 2.7.2).  The object, shown in Fig. \\ref{fig14} , has a smooth appearance, like a dwarf elliptical galaxy. However, it is much larger than all the other satellite candidates ($\\sim$ 10 $\\times$ 5 h$^{-1}$ kpc) and it is very close to one of the central galaxies.  Bremnes et al. (1998) has detected several low-surface brightness clouds in the direction of M81 which overlap with detections in the IRAS 100 $\\micron$ map, suggesting that these may indeed be cirrus structures in our own Galaxy.  However, in the case of Dorado, the IRAS 100 $\\micron$ map of the region shows no similar features. Moreover, the ``cloud'' in Dorado is much closer to a bright galaxy than in the case for the clouds in the M81 group.  An alternative explanation would be that this object is the residual of a past galaxy-galaxy interaction or a disrupted dwarf galaxy like Sagittarius in the Local group. Higher resolution imaging of this object may reveal its nature.  \\subsection {Summary}  Summarizing, we have detected 69 low-surface brightness dwarf galaxies as candidate dwarf members of the nearby group of galaxies Dorado. Our CCD survey covered $\\sim 2.6$ square degrees of the group. The galaxies were chosen based on  their sizes and magnitudes at the average limiting isophote of $26$ V mag/arcsec$^{2}$. We showed, using Monte Carlo simulations, that we are able to detect possible dwarf members in Dorado with $-10.5>M_{V}>-17$ with $22.5<\\mu_{0}<25.5$ V mag/arcsec$^{2}$ with a completion fraction $f>80$\\%. Four new large low-surface brightness galaxies previously undetected by standard techniques were detected using a smoothing technic. One of these galaxies is a large low-surface brightness ``cloud'' with an irregular shape much more extended than the other dwarf galaxies in the survey.  The color distribution for the 65 LSBD galaxies with color information varies from very blue to very red with a peak at $V-I = 0.98$.  The color profiles of the galaxies show very small gradients (on the order of the photometric errors). In some cases we measured a small inclination in the profiles. The reddest galaxy (3330-10) shows an irregular morphology, a strong bulge, a large halo and a flat color profile. This galaxy could be a HI-rich galaxy like others of the same type found in the Pegasus and Cancer clusters by O'Neil et al. (2000). The projected surface density shows a significant excess of LSBD galaxies within $\\sim 250$ h$^{-1}$ kpc ($\\sim$ 0.8 deg) from the center of the group. Comparisons with others results obtained from X-ray groups and for the isolated elliptical galaxy NGC1132 suggest that the Dorado LSBD are probably physically associated with the overall potential well of the group."
    },
    "0010/astro-ph0010595_arXiv.txt": {
        "abstract": "We emphasize the use of high-resolution CO line observations to derive central rotation curves of galaxies. We present an example for high-resolution interferometer observations of NGC 3079, and discuss the PV diagram and derived rotation curve. The CO central rotation curves are combined with the outer curves from \\ha\\ and HI-line observations to obtain total RC. We show that well resolved RCs for massive galaxies generally start from non-zero velocities at the nucleus. ",
        "introduction": "The CO molecular lines are useful to derive accurate rotation curves in the central regions of spiral galaxies, because of the high concentration in the center as well as for negligible extinction (Sofue 1996, Sofue et al 1997, 1998, 1999). Recent high-dynamic-range CCD spectroscopy in optical lines also make it possible to obtain high accuracy rotation curves in the central regions (Rubin et al 1997; Bertola et al 1998). However, in general,  optical lines suffer from significant extinction by the central dusty disks, which is particularly signficant for highly-inclined and edge-on galaxies.  Hence, the CO lines will be the most appropriate tool to investigate the central kinematics of spiral galaxies, if the angular resolution is sufficiently high. In this paper, we present recent high-resolution interferometer observations of CO lines, and discuss the general properties of central rotation kinematics based on the high-accuracy rotation curves.  \\begin{figure} \\psfig{figure=fig1a.ps,width=10cm} \\psfig{figure=fig1b.ps,width=10cm} Fig. 1. Position-velocity diagram along the major axis of NGC 3079 in CO J=1-0 line at resolutions of $1''6 \\times 1.4''$ (top) and $1''.2 \\times 1''.1$ (bottom). Note that the rotation velocity does not decline to zero at the nucleus, but remains always at finite value. UHC stands for ultra-high-density molecular core, and NMD for nuclear molecular disk. $1''$ corresponds to 75 pc. \\end{figure} ",
        "conclusions": ""
    },
    "0010/astro-ph0010306_arXiv.txt": {
        "abstract": "Mean proper motions and parallaxes of 205 open clusters were determined from their member stars found in the Hipparcos Catalogue. 360 clusters were searched for possible members, excluding nearby clusters with distances $D < 200$ pc. Members were selected using ground based information (photometry, radial velocity, proper motion, distance from the cluster centre) and information provided by Hipparcos (proper motion, parallax). Altogether 630 certain and 100 possible members were found. A comparison of the Hipparcos parallaxes with photometric distances of open clusters shows good agreement. The Hipparcos data confirm or reject the membership of several Cepheids in the studied clusters. ",
        "introduction": "The aim of this work is to determine proper motions and parallaxes of open clusters using the Hipparcos Catalogue (ESA \\cite{esa}). We will focus here on the more distant open clusters and try to find proper motions for all clusters with distances greater then $D=200$ pc. In combination with accurate photometric distances and radial velocities, the proper motions will provide valuable information on the kinematic parameters of the galactic rotation curve and the distance to the galactic centre. They are also useful to study the formation and evolution of the open cluster system. These questions will be addressed in a forthcoming paper. The present paper provides the observational database for these studies.  Stars in open clusters were searched and proposed by Wielen \\& Dettbarn (\\cite{wd85}) and selected by a specific working group within the INCA Consortium during the creation of the Hipparcos Input Catalogue. However, the Hipparcos Catalogue may contain additional cluster stars not selected by this working group: First, there may exist bright cluster stars that have so far not been studied at all or were studied in detail only after the creation of the Hipparcos Input Catalogue. Such stars could have entered the Hipparcos Catalogue as part of the all sky survey. In addition, cluster stars may have been proposed by other research groups if they are of astrophysical interest.  It therefore seemed appropriate to perform a new search for stars in open clusters. Sections 2 and 3 describe the selection of the open clusters and the search for Hipparcos stars in the cluster fields.  Section 4 describes the way in which cluster members were identified among the candidates and the next section describes how the mean astrometric parameters of the clusters were determined. Section 6 lists the classification of the Hipparcos stars and presents the mean astrometric parameters of the open clusters. Two applications of the astrometric parameters are also described in this section: First, we compare the Hipparcos parallaxes with the photometric parallaxes of the open clusters to check for systematic errors in both methods. Second we use the proper motions of the open clusters to check the membership of several Cepheids. Section 7 summarises our results. ",
        "conclusions": "The proper motions and parallaxes of 205 open clusters were determined from 630 certain and 100 possible members found in the Hipparcos Catalogue. A comparison of the parallaxes with photometric distances from Loktin et al.\\ (\\cite{lok97}) argues for a decrease of the photometric distance scale by $12\\% \\pm 6\\%$, while the distance scale of Dambis (\\cite{dam99}) is in good agreement with Hipparcos. No evidence for unaccounted small-scale correlated errors in the Hipparcos Catalogue is found by these comparisons. It therefore seems unlikely that such errors can explain the discrepancy between the Hipparcos parallax and photometric distance estimates of the Pleiades as proposed by Narayanan \\& Gould (\\cite{nago}).  With the help of the Hipparcos proper motions, we can confirm the membership of the Cepheids U Sgr in IC 4725, S Nor in NGC 6087, DL Cas in NGC 129 and V Cen in NGC 5662 and can reject the membership of SZ Tau in NGC 1647. The improved membership information may lead to better estimates for the absolute magnitudes of Cepheids."
    },
    "0010/astro-ph0010656_arXiv.txt": {
        "abstract": "Hydrodynamic simulations of \\lya clouds based on {\\it ab inito} cosmological models have produced results that are in broad agreement with observations.  However, further analyses have revealed that, with progressively higher numerical resolution, the median or cutoff line width of the simulated \\lya clouds (i.e. the Doppler parameter) appears to converge to a value significantly below what is observed at $z \\sim 3$ (by about a factor of $1.5$).  These convergence test simulations do not include feedback from star formation. Given the observed metallicity in the Lyman alpha clouds we suggest that supernovae, which presumably polluted the IGM with metals, may have deposited a sufficient amount of energy in the IGM to reconcile the theory with observations.  Simple arguments immediately narrow the redshift range of pollution down to $4<z_{dep}<9$. It seems quite certain that dwarf and sub-dwarf galaxies with total masses in the range $10^{6.5-9.0}\\msun$ have to be largely responsible for the pollution. Furthermore, it is implied that either star formation is very efficient or metal yield is very high for these early dwarf galaxies, if the mean metallicity in the universe at $z=3$ is as high as $<Z>=10^{-2}\\zdot$. Finally, assuming the specific supernova heating energy is proportional to the metallicity of a gas, we note that the picture proposed here would be consistent with supernovae being the apparently needed heating source for intra-cluster gas if the required heating of the intra-cluster gas is no greater than $1$keV per particle. ",
        "introduction": "Recent cosmological hydrodynamic simulations have shifted the paradigm of \\lya cloud formation.  Several independent groups have consistently shown that \\lya clouds are an integral part of the cosmic structure, resulting from the gravitational growth of density fluctuations on small-to-intermediate scales at $z\\sim 2-4$ ($\\sim 100$ kpc to a few comoving megaparsecs) with most of the clouds in the form of filamentary structures of moderate overdensity ($\\delta_\\rho \\sim 1-10$) (Cen \\etal 1994; Zhang \\etal 1995; Hernquist \\etal 1996; Miralda-Escud\\'e \\etal 1996; Wadsley \\& Bond 1996; Cen \\& Simcoe 1997; Theuns \\etal 1998; Bryan \\etal 1999; Dav\\'e \\etal 1999).  While the best agreement between model computations and observations is seen in the column density distribution, a close examination reveals that the Doppler width distribution shows a disturbingly large discrepancy.  Specifically, Bryan \\etal (1999) and Theuns \\etal (1998) have done convergence tests and shown that higher resolution simulations predict a median Doppler width $b_{med,comp}\\sim 20$km/sec versus the observed $b_{med,obs}\\sim 30$km/sec.  A comparable difference is found between the low cutoff width of simulated and observed clouds.  If this is interpreted as being due to a difference in temperature, it means that the computed clouds are colder than the real ones by approximately a factor of two, translating to a temperature deficit of approximately $10^4~$K in the computed low column density clouds.  There are several ways to raise the temperature in the simulations to bridge the gap.  First, increasing the baryonic density $\\Omega_b$ allows more photoelectric heating and thus a higher IGM temperature, with the scaling being $T\\propto (\\Omega_b h)^{1/1.7}$ (Hui \\& Gnedin 1997), where $h$ is the Hubble constant in units of $100$ km/sec/Mpc. Second, reionizing helium at a lower redshift will help somewhat (Hui \\& Gnedin 1997; Haehnelt \\& Steinmetz 1998).  Third, Compton heating by the high energy background may be important, if most of the observed background at $z=0$ is produced at high redshift ($z>4$) (Madau \\& Efstathiou 1999). Oh (2000) has suggested that X-rays from the first star clusters could provide this high energy X-ray background. Fourth, radiative transfer effects, which are neglected in the simulations, may provide additional heating (Abel \\& Haehnelt 1999). Finally, photoelectric heating by dust grains (Nath, Sethi, \\& Shchekinov 1999) provides yet another mechanism and is more closely related to what is proposed in this paper.  In this {\\it Letter} we examine the possibility that feedback due to early star formation in dwarf galaxies with masses $M\\le 10^9\\msun$ forming in the redshift range $7\\le z_{f} \\le 15$ may have played a significant role in changing the thermodynamic state of the intergalactic gas at moderate density. The idea of supernova (SN) feedback into IGM can be traced back at least two and a half decades (Schwarz, Ostriker, \\& Yahil 1975) and many authors have subsequently addressed this issue in various contexts (e.g., Ikeuchi \\& Ostriker 1986; Couchman \\& Rees 1986; Silk, Wyse, \\& Shields 1987; Tegmark, Silk, \\& Evrard 1993; Shapiro, Giroux, \\& Babul 1994). While the basic physical process is by no means new, we address its effect in the new context of the temperature of the \\lya clouds. In particular, we adopt the standard picture of re-ionization due to radiation from quasars or massive stars, but argue that a subsequent (or slightly overlapping) phase of energy ejection from galaxies provides an additional source of heat for the IGM. ",
        "conclusions": "We suggest that energy feedback from supernovae in dwarf or sub-dwarf galaxies with masses in the range $10^{6.5-9.0}\\msun$ forming at $z_{f}=7-15$ could raise the temperature of the \\lya clouds at $z\\sim 3$ by $\\sim 10^4$K, given the observed metals in the \\lya clouds at $z\\sim 3$.  This provides an alternative mechanism to bring the results of recent hydrodynamic simulations into agreement with the observed Doppler width distribution of \\lya clouds.   This scenario may be tested in a number of ways.  First, since metals are a sign of heating, we would expect a correlation between metallicity and temperature (i.e. Doppler width).  A search for this effect should be restricted to low column density lines ($N_{HI} < 10^{14}$ cm$^{-2}$ at $z \\sim 3$) where hydrogen line cooling is unimportant.  Since it is difficult to detect metal lines in such systems, a statistical search may be necessary (e.g. Cowie \\& Songalia 1998).  Another test would be to examine the evolution of the temperature of Ly$\\alpha$ clouds at high redshift $z \\ge 4$.  In particular, the mean temperature should not increase with redshift as rapidly as in the case with only photoionization heating and adiabatic cooling.  This is because heating (and subsequent cooling) due to contamination is likely to be continuous over the entire redshift range.  A third prediction would be a larger spread in the temperature of the gas at a given density, since heating would no longer depend just on density (as it does for radiative heating) but instead would occur suddenly.  This might show up as a larger scatter in the joint $N_{HI}-b$ distribution for the Ly$\\alpha$ forest.  Since we require strong outflows from dwarf galaxies, a fourth suggestion would be to attempt to directly observe these outflows.  This would be difficult at the relatively high redshifts we predict, but may be possible if the outflows are driven by strong starbursts. Finally, at low redshift we predict that the metallicity in the low column density Ly$\\alpha$ forest clouds should remain approximately constant at $Z \\sim 10^{-2} - 10^{-3} Z_{\\odot}$, since later, larger galaxies are unlikely to be able to signficantly contaminate the regions further. An interesting aside is that collisional ionization at earlier times could be significant if energy feedback from SNe raised the IGM temperature, and may significantly alter the reionization picture of the universe.  This issue will be examined in detail subsequently.  We note that the SN heating scenario for the high redshift IGM may be related to the apparently wanted additional heating of intra-cluster gas, {\\it if the latter is also due to SNe}. If we take the apparently required specific energy per baryonic particle of $0.3$keV for the intra-cluster gas (e.g., Lloyd-Davies, Ponman, \\& Cannon 2000) and simply assume that specific heating energy is linearly proportional to metallicity, we obtain the specific energy per baryonic particle for the high redshift IGM to be $Z_{IGM}/Z_{ICM}\\times 0.3$keV$=3\\times 10^{-3}/0.3\\times 0.3$keV$=3$eV, not inconsistent with the results presented in this paper. However, a specific heating energy per baryonic particle for the intra-cluster gas greater than $1$keV is unattainable within our picture."
    },
    "0010/astro-ph0010460_arXiv.txt": {
        "abstract": "The far infrared background is a sink for the hidden aspects of galaxy formation. At optical wavelengths, ellipticals and spheroids are old, even at $z \\sim 1.$ Neither the luminous formation phase nor  their early evolution is seen in the visible. We infer that ellipticals and, more generally, most spheroids must have formed in dust-shrouded starbursts.  In this article, we show how separate tracking of disk and spheroid star formation  enables us to infer that disks dominate near the peak in the cosmic star formation rate at  $z \\lapproxeq 2$  and in the diffuse ultraviolet/optical/infrared background, whereas spheroid formation dominates the submillimetre background. ",
        "introduction": "The theory of galaxy formation became firmly established once temperature anisotropies in the cosmic microwave background were discovered.  Only then did we have some indication of the amplitude and power spectrum of the primordial density fluctuations that seeded structure formation via gravitational instability once the universe  became matter-dominated.  Adiabatic fluctuations in cold dark matter, with the nucleosynthesis baryon component of $\\Omega_bh^2 \\approx 0.02$ and the cold dark matter component of $\\Omega_0 \\approx 0.3,$ form the basic paradigm for structure formation, supplemented by the recent realisation that  $\\Omega_0 + \\Omega_{\\Lambda} \\approx 1$ and $\\Omega_{\\Lambda} - \\Omega_0 \\approx 0.4$ from the CMB $\\Delta$T/T degree-scale and SNIa cosmological probes, respectively.  Inflation suggests that the fluctuations are nearly gaussian distributed, and the problem of galaxy formation, as defined by the assembly of dark halos, is sufficiently well-posed that it has been systematically probed via computer simulations, with varying degrees of success. Highlights of the N-body simulation approach to structure formation include studies of the evolution of the galaxy correlation function, of the mass function and structure of galaxy halos and galaxy clusters, and of the structure of the intergalactic medium.  Much of this work has successfully accounted for the observed characteristics of the observed universe. However, not all is in good shape.  Galaxy formation to observers is the assembly of the structural components of a galaxy.  Predictions of the properties of the luminous components of galaxies have not met with much success. Disk sizes are far too small.  The ratio of luminosity to mass is too low. Even the detailed structure of the dark galaxy halos, as probed by high resolution N-body simulations, is producing results that are discrepant with observations: a density profile with a central cusp $\\rho (r) \\propto r^{-1.5}$ is predicted that is seen neither in low surface brightness dwarf spirals, barred galaxies, the Milky Way, nor in giant ellipticals.  Moreover, excessive substructure is found, leading to apparent conflict with the observed galaxy luminosity function.  Cosmologists have rushed in to resurrect cold dark matter with various alternatives. These range from particle astrophysics solutions that involve for example, dabbling with the weakly interacting nature of standard CDM, or astrophysical solutions, that invoke recourse to, for example, dynamical feedback of the dissipating baryon matter. While any of these solutions might eventually turn out to have some relevance, a more useful exercise may be  to assess the protogalaxy phase of galaxy formation. Improving our understanding of early galaxy evolution is not only closely coupled to observation, but may also help clarify the nature of the protogalactic environment, the role of such astrophysical processes as feedback, and the possible need for more exotic physics.  Ideally, one would like to solve the galaxy formation problem numerically. While with a large N-body simulation, one can identify the sites of galaxy formation, the resolution required to adequately model dynamics alone is at the limit of current computing capabilities.   Incorporation of the gas hydrodynamics, required to study star formation, at adequate resolution to cover an entire galaxy, is not feasible.  The current limit of simulations only allows study of the formation of the first stellar mass clouds in a protogalaxy, with no possibility of studying either  efficiency of star formation or continuing fragmentation. Yet the first generation of stars must have involved millions of massive stars in order to produce the chemical elements seen in the most metal-poor halo stars.  Cosmologists have therefore developed a semi-analytical approach to galaxy formation. This consists of constructing Monte-Carlo realisations of the merging histories of dark matter halos which have collapsed under their own self-gravity. Galaxy formation is then followed by imposing certain simple rules, for example for the rate of star formation.  The size of a galaxy is determined by ensuring that angular momentum is conserved in the collapsing cloud.  The cloud forms with the small value of initial dimensionless angular momentum $\\lambda \\approx 0.05$ that is generated by tidal torques between neighbouring fluctuations, and attains virial equilibrium at a fraction $\\lambda$ of the radius at maximum extent. A simple Schmidt law is used for the star formation rate in the resulting disk, and cold gas infall continues to feed the disk.  Spiral galaxies are relatively isolated galaxies that have not undergone a merger (that is, of nearly equal mass) within the past 10 Gyr or so.  Elliptical galaxies are assumed to form in major mergers on a dynamical time scale: stars are assumed to form in a starburst.  Semi-analytical galaxy formation is able to make realistic maps of the galaxy redshift distribution, and once the feedback parameter is carefully tweaked, the galaxy luminosity function and cosmic star formation history can be derived. However, the model is not very robust, since exploration of the large parameter space for the gas and star formation physics is not really feasible.  It is therefore useful to consider fully analytic models of galaxy formation. We briefly describe such a model in section 2, detail the key procedure which allows us to distinguish between spheroid/disk formation in section 3, and present our results in section 4. ",
        "conclusions": "We have been able to develop an analytical model that enables us to separate the contributions of disk galaxies and spheroids to the panchromatic galaxy counts, the cosmic star formation history and the diffuse background radiation. The quiescent star formation mode characteristic of disks self-consistently accounts for all optical, NIR and FIR data.  Only in the submillimetre range does the starburst mode, associated with spheroid formation, dominate. It will be particularly interesting to exploit  observations near 400 $\\mu$m, where a flux-limited survey should find comparable numbers of forming disks and protospheroids/ellipticals."
    },
    "0010/astro-ph0010526_arXiv.txt": {
        "abstract": "We have determined the distance to a second eclipsing binary system (EB) in the Large Magellanic Cloud, HV~982 ($\\sim$B1 IV-V + $\\sim$B1 IV-V).   The measurement of the distance --- among other properties of the system --- is based on optical photometry and spectroscopy and space-based UV/optical spectrophotometry.  The analysis combines the ``classical'' EB study of light and radial velocity curves, which yields the stellar masses and radii, with a new analysis of the observed energy distribution, which yields the effective temperature, metallicity, and reddening of the system plus the distance ``attenuation factor'', essentially $({\\rm radius/distance})^2$. Combining the results gives the distance to HV~982, which is $50.2 \\pm 1.2$ kpc.  This distance determination consists of a detailed study of well-understood objects (B stars) in a well-understood evolutionary phase (core H burning).   The results are entirely consistent with --- but do not depend on --- stellar evolution calculations.  There are no ``zeropoint'' uncertainties as, for example, with the use of Cepheid variables.  Neither is the result subject to sampling biases, as may affect techniques which utilize whole stellar populations, such as red giant branch stars.  Moreover, the analysis is insensitive to stellar metallicity  (although the metallicity of the stars is explicitly determined) and the effects of interstellar extinction are determined for each object studied.   After correcting for the location of HV~982, we find an implied distance to the optical center of the LMC's bar of $d_{\\rm LMC} = 50.7\\pm1.2$ kpc.  This result differs by nearly 5 kpc from our earlier result for the EB HV~2274, which implies a bar distance of 45.9 kpc. These results may reflect either marginally compatible measures of a unique LMC distance or, alternatively, suggest a significant depth to the stellar distribution in the LMC.  Some evidence for this latter hypothesis is discussed. ",
        "introduction": "The distance to the Large Magellanic Cloud (LMC) is a key factor in determining the size scale of the Universe.  Indeed, the uncertainty in the length of this ``cosmic meter stick'' is responsible for much of the current uncertainty in the value of the Hubble constant, as noted by Mould et al. (2000).  The LMC distance is as controversial as it is important.  Existing determinations span a wide range (see Fig. 1 of Mould et al.) and are often grouped into the ``long'' distance scale results ($d > 50$ kpc and ($V_0-M_V) \\simeq 18.7$ mag) and the ``short'' distance scale results ($d < 50$ kpc and ($V_0-M_V) \\simeq 18.3$ mag).  In some cases, the same technique (e.g., Cepheids) can support both the long and short scales, depending on the assumptions and details of the analysis.  Reviews of LMC distance determinations can be found in Westerlund (1997) and Cole (1998), and numerous new papers have appeared in the last several years, indicating the high activity level and interest in the field.  Recently, we showed that well-detached main sequence B-type eclipsing binary (EB) systems are ideal standard candles and have the potential to resolve the LMC distance controversy (Guinan et al. 1998a; hereafter Paper I).  The advantages of using EBs are numerous.  First, an accurate distance can be determined for each individual system --- and there are many systems.  This is in contrast to techniques that utilize, for example, Cepheids or red giant stars, where entire populations are used to derive a single distance estimate.  It also contrasts with analyses of SN 1987A, which have the potential to yield a precise distance but for which there is only one object to study.  The EBs can provide not only the mean LMC distance, but also can be used to probe the structure of the LMC.  Second, the analysis involves well-understood objects in a well-understood phase of stellar evolution (core hydrogen burning) and the results for each object can be verified independently by --- but do not depend on --- stellar evolution theory.  Third, the analysis is robust and not subject to any zeropoint uncertainties, nor are there any adjustable parameters.   For example, the results are extremely insensitive to stellar metallicity, although the metallicity of the individual EBs is explicitly determined and incorporated in the analysis.  Likewise, the determination of interstellar extinction is an integral part of the analysis of each object.  The study of the EB system HV~2274 in Paper I yielded a distance of $46.8\\pm1.6$ kpc and stellar properties consistent with stellar evolution theory (Ribas et al. 2000a).  Correcting for the position of HV~2274 relative to the LMC center yielded a LMC distance of $45.7\\pm1.6$ kpc corresponding to $(V_0-M_V) = 18.30\\pm0.07$ mag.  This result argues strongly in favor of the ``short distance'' to the LMC. Since Paper I, several partial-reanalyses of the HV~2274 system have appeared in the literature (Udalski et al. 1998; Nelson et al. 2000; Groenewegen \\& Salaris 2001).  These authors advocate various adjustments in the results from Paper I which yield LMC distance moduli in the range 18.22--18.42 mag.  These adjustments all stem from complications arising from the incorporation of optical photometry in the analysis.  We will return to this issue later in this paper.  The main goal of this paper is to apply our analysis to a second LMC EB system, HV~982, and derive its stellar properties and distance.  This EB, with $V \\simeq 14.6$, consists of two mildly evolved main sequence B stars each corresponding to spectral class $\\sim$B1 IV--V.  A key difference between this analysis and that presented in Paper I for HV~2274 is that we have here obtained space-based spectrophotometric measurements extending from 1150 \\AA\\/ to 7500 \\AA, eliminating the reliance on optical photometry.  These new data preclude the ambiguities which plague the HV~2274 result and allow us to realize the full potential of our analysis technique.  In \\S 2, we describe the data included in this study.  In \\S 3, the analysis --- which incorporates the light curve, radial velocity curve, and spectral energy distribution of the system  --- is discussed.   Some aspects of our results relating to the interstellar medium towards HV~982 are described in \\S4, including an indication of the relative location of HV~982 within the LMC.  The general stellar properties of the HV~982 system and their consistency with stellar evolution theory are described in \\S 5.  In \\S 6, we show how the distance to the system is derived from our analysis, and compare this result with a reanalysis of the HV~2274 data.  We discuss the distance to the LMC and summarize our conclusions from study of two binary systems in \\S 7. ",
        "conclusions": ""
    },
    "0010/gr-qc0010082_arXiv.txt": {
        "abstract": "The formation of energetic rings of matter in a Kerr spacetime with an outward pointing acceleration field does not appear to have previously been noted as a relativistic effect. In this paper we show that such rings are a gravimagneto effect with no Newtonian analog, and that they do not occur in the static limit. The energy efficiency of these rings can, depending of the strength of the acceleration field, be much greater than that of Keplerian disks. The rings rotate in a direction opposite to that of compact star about which they form. The size and energy efficiency of the rings depend on the fundamental parameters of the spacetime as well as the strength the acceleration field. ",
        "introduction": "\\label{sec:i}  A compact celestial body may accrete matter from a nearby companion. If the accreting matter has large enough angular momentum, a potential barrier will form stopping the in-fall.  Matter bouncing back from the angular momentum barrier, will collide with the in-falling one and eventually an equilibrium condition is reached where by most of matter moves on circular orbits and is  confined to a plane.  Once equilibrium is established, if matter is subjected to no forces other than gravity, it moves in geodesic orbits.  Matter confined to a plane and moving in geodesic circular motion is said to form a Keplerian disk.  Keplerian disks adequately describe galactic motion since stars are sufficiently far apart to be considered  non-interacting point sources. On the other hand, matter surrounding a hot compact source of radiation, is likely to be affected not only by the radiation pressure of the source but also by its own internal forces. For these reasons the accreted matter will not necessarily follow Keplerian orbits.  Despite this, not much work has been done on modeling non-Keplerian disks, see~\\cite{Bur98, Kin98} for some examples.  In this paper we analyze the dynamics underlining the formation of rotating structures in the presence of an acceleration field within the full theory of general relativity. Although we limit our considerations to a point particle approach, the fundamental results should also be manifest also in a fluidodynamical treatment.  The first problem one faces in examining motion exterior to a rotating mass is the lack of an exact solution to Einstein's equations. To describe the dynamics in general relativity we must know the space-time geometry.  Birkhoff's theorem~\\cite{Wei72} tells us that the space-time exterior to a non-rotating, spherically symmetric, electrically neutral configuration, is the Schwarzschild solution. Unfortunately though, there is no generalization of  Birkhoff's theorem for rotating stars. The space-time of a stationary, uncharged and rotating black hole is uniquely described by Kerr solution \\cite{Cha83}. This has led a number of authors~\\cite{Kra78, McM91, Mag93, Mag95, Dra97}. to suggest that the exterior of a rotating star may be described to sufficient accuracy by Kerr geometry.  Thus far however, nobody has been able to match a ``physically sensible'' interior solution to the Kerr metric.  Although one may think that Kerr metric still describes the basic properties of a space-time exterior to a rotating star, mainly stemming from stationarity and axisymmetry.  In this paper, we investigate the astrophysical importance of a general relativistic effect arising in Kerr geometry which has no Newtonian or Schwarzschild analogue. In Kerr space-time, one finds that non-geodesic (spatially) circular orbits may have, at each value of their coordinate radius, an extreme acceleration for non-zero  orbital angular velocities (with respect to infinity). As we shall see, this effect is responsible for the existence of narrow and stable rings of matter, populated by highly energetic particles. In Newtonian theory and Schwarzschild geometry acceleration extrema occur only for zero angular velocity.  The existence of an extremal acceleration implies that for a range of angular velocities, an increase in the {\\it modulus} of the angular velocity, requires a larger outward pointing acceleration to maintain a circular orbit circular see.  This is contrary to the Newtonian case were an increase in the modulus of angular velocity, requires a smaller outward acceleration to maintain a circular orbit, in all regions.  This effect was first noticed in the Schwarzschild space-time by~\\cite{Abr74} and then in the Kerr metric by~\\cite{Fel91}. In the latter case the effect exists at all values of the coordinate distance from a rotating source so one can even hope to measure it in a weak field regime~\\cite{Fel95}.  In Section~\\ref{sec:cmitkm}, we shall outline the main properties of accelerated circular orbits in the Kerr metric. In Section~\\ref{sec:nkr}, we show how ring structures form due to the existence of an acceleration field and how energy considerations allow us to decide about the stability of the rings. In Section~\\ref{sec:ai} we discuss their possible astrophysical importance.   Finally in the last Section we summarize the results, draw our conclusions, and discuss possible further work.  In what follows we shall use geometrized units such that $G = c= 1$, $G$ being the gravitational constant and  $ c $ the vacuum speed of light; Greek indices run from 0 to 3 and signature is chosen as $+2$. ",
        "conclusions": "\\label{sec:c}  In this paper, we have shown that highly energetic rings of matter can occur around the exterior of compact stars. We have shown that these rings can achieve energy efficiencies much greater than those of Keplerian disks. We have also shown that the size and energy efficiency of these rings depends on, the specific angular momentum of the star, its mass,  the strength of the acceleration field it produces, and the properties of the matter with the ring itself.  This paper forms the corner stone of work to come. Having established the general theory behind the formation of non-Keplerian rings there is still a lot of theoretical and modelling work to be done. For example, the results of this work make it possible to determine the energy efficiency as a function of  angular velocity for a given acceleration field. Once this has been done it is then possible to deduce the spin rate of a star from the amount of energy produced by its rings.  The acceleration field we have studied in this paper is only one of many possibilities. While the general property of the formation of rings is not likely to change for different models of the acceleration the specific nature of the rings, such as their size and efficiency will.  The structure of matter within the rings will determine the acceleration field there. It would be interesting to examine the fields produced by matter with a polytropic equation of state.  Similarly, the Hamiltonian describing the particle dynamics was chosen to be a minimally coupled one, but there are many other options. Even the choice of vector potential is not unique and more theoretical work is needed to understand the form it should take.  The possibilities for extensions of this work, while not endless are certainly large. The consequences of such work are that it will make it possible to determine some of the fundamental properties of the stars by observing the rings around them.  \\subsection*"
    },
    "0010/astro-ph0010532_arXiv.txt": {
        "abstract": "Spiral arm modulation in NGC 4062 (an optical flocculent) is detected for the first time; the Fourier spectra of NGC 4062 and NGC 5248 (a classic grand design in optical images) are almost identical. ",
        "introduction": " ",
        "conclusions": ""
    },
    "0010/astro-ph0010018_arXiv.txt": {
        "abstract": "Long-term X-ray variability of the low-luminosity active galactic nucleus of M81 was studied, using 16 {\\it ASCA} observations spanning 5.5 years. The object exhibits a factor three variation over the 5.5 years. The source intensity was relatively constant within each observation which lasted typically for one day, but intra-day variability by 30\\% was detected on the 15th observation. The power-spectral density (PSD) was estimated in ``forward'' manner, over a frequency range of $10^{-8.2}$--$10^{-4.3}$ Hz (period range of 1/4 day--5.5 years), by utilizing structure function and extensive Monte-Carlo simulations in order to overcome the very sparse and uneven data samplings. When the PSD is assumed to be white below a ``break frequency'' $f_{\\rm b}$ and falls off as $\\propto f^{-\\alpha}$ above $f_{\\rm b}$, where $f$ is frequency and $\\alpha$ is a positive parameter, the M81 light curve is well described with 1/$f_{\\rm b} \\ge 800$ days and $\\alpha = 1.4 \\pm 0.2$. ",
        "introduction": "A number of low-luminosity AGNs (LLAGNs), with X-ray luminosities in the range $10^{40-42}$ erg s$^{-1}$, have been discovered in nearby normal galaxies with {\\it ASCA} (e.g. Makishima et al. 1994; Iyomoto et al. 1996; Ishisaki et al. 1996; Iyomoto et al. 1998; Terashima et al. 1998; Ptak et al. 1999). LLAGNs are now considered to be ubiquitous at least in our neighbourhood.  The low luminosities of LLAGNs imply lower black-hole masses and/or lower mass accretion rates as compared to more luminous AGNs. These alternatives may not be distinguished by X-ray spectra, since AGNs exhibit similar X-ray spectra over wide range of luminosities, from LLAGNs through quasars. On the other hand, X-ray variability is expected to provide useful clues to the issue, as a zero-th order approximation, because characteristic time scales of the X-ray emission is considered to correlate positively with the system size and hence the system mass.  During individual observations with {\\it ASCA}, which typically last for one day, LLAGNs tend to be much less variable than the lower-luminosity end of the Seyfert galaxies (e.g. Ptak et al. 1998). We are therefore tempted to speculate that LLAGNs have longer characteristic time scales of the X-ray variation and hence larger black-hole masses compared to Seyferts. In this case, LLAGNs might be ``dead quasars'' in our neighbourhood. To examine this possibility, it is important to determine the power spectral density (PSD) of intensity variations of LLAGNs on time scales such as dozens of days or longer. So far, no such attempts have been made, except for a rough estimation by Ishisaki et al. (1996), because only few observations with uneven intervals are usually available for each LLAGN in X-rays.  Using {\\it ASCA}, we investigate long-term variability of an LLAGN at the nucleus of M81, which is an Sab galaxy at a distance of 3.6 Mpc (Freedman et al. 1994). Its nucleus exhibits weak AGN properties in various wave bands. Optically, it is classified as a LINER (Ho et al. 1997). In radio, the spectrum from the nuclear region is interpreted as a jet emission (Boettcher et al. 1997). The nucleus is also an X-ray source with a 2--10 keV luminosity of $\\sim 2 \\times 10^{40}$ erg s$^{-1}$, as confirmed repeatedly by {\\it Einstein Observatory} (Fabbiano et al. 1988), {\\it EXOSAT} (Barr et al. 1995), {\\it Ginga} (Tsuru 1992), {\\it ROSAT} (Boller et al. 1992), {\\it BBXRT} (Petre et al. 1993), {\\it ASCA} (Ishisaki et al. 1996), and {\\it BeppoSAX} (Pellegrini et al. 2000). These multi-mission results indicate a mild X-ray intensity variation. Ishisaki et al. (1996), using multiple observations with {\\it ASCA} spanning $\\sim$2.5 years, detected long-term variation as well as a short-term ($\\sim 1$ day) variation on one occasion. Pellegrini et al. (2000) detected a periodic oscillation with a sub-day period. In addition, the spectra of this LLAGN, except that measured with {\\it Einstein}, are well fitted with a power-law model having a photon index of 1.5--2.0. Therefore, M81 has been considered to possess an LLAGN. ",
        "conclusions": "\\subsection{Comparison with Other Objects} Let us compare the estimated PSD of the M81 nucleus with those of other objects. We utilize the plot invented by Hayashida et al. (1998): namely, a (frequency) $\\times$ (NPSD) plot, or $fP(f)$ plot. As Hayashida et al. (1998) pointed out, the $fP(f)$ plot displays relative contribution of the power at each frequency decade, like the $\\nu f_\\nu$ plot of energy spectrum. As illustrated in Figure~\\ref{fig:npsd}, the power concentrates around $f_{\\rm b}$, so that $f_{\\rm b}$ is an important parameter characterized each system.  So far, Ptak et al. (1998) and Awaki et al. (1999) compared the high-frequency information of LLAGNs with those of the other objects and suggested that LLAGNs may have long time scale and/or large black-hole mass. However, there was an implicit or explicit assumption in their analysis, that the individual AGNs have the same PSD shape and same total area of NPSD on the $fP(f)$ plot. There was no concrete evidence of the assumption, and it should be examined by measuring the low frequency portion of the PSD, as we have performed. Moreover, the slope index of PSD differs object to object ($1< \\alpha< 2$). Consequently, their result strongly depend on the frequency where the NPSD amplitude was measured. Therefore, it is important to consider the PSD over wide frequency band.  In Figure~\\ref{fig:npsd}, we show the M81 NPSD, derived from Figure~\\ref{fig:cont}b, together with those of Seyfert galaxies and a Galactic black-hole candidate taken from Hayashida et al. (1998) and Pounds et al. (1988). The figure reveals two facts. First, the power peak of M81 locates at a very low frequency for its low luminosity. Second, the total area of the M81 NPSD is similar to those of the other objects. Consequently, the observed lack of significant short-term variability in M81 is not because the total variation power is small, but because the power concentrates on low frequencies. This suggests that the other LLAGNs with small NPSD amplitude at high frequency also have large power on low frequencies.  \\subsection{Black-Hole Mass} Finally, we estimate the black-hole mass, $M_{\\rm BH}$, of the M81 nucleus. Conventionally, characteristic variation time scales of AGNs have been translated into $M_{\\rm BH}$ assuming a linear proportionality between the time scale and mass, and often utilizing Cygnus X-1 as a standard, although this estimation is only appropriate if the physics of accretion are similar between Cygnus X-1 and M81. This allows us to estimate $M_{\\rm BH}$ in the M81 nucleus to be $> 4.0 \\times 10^7~M_\\odot$.  So far, $M_{\\rm BH}$ of the M81 nucleus has been estimated in various ways. From measurements of the velocity of the broad-line region or narrow-line region, several authors estimated the mass within these regions of the M81 nucleus, although their results have large scatter: $1\\times 10^6~M_\\odot$--$1.8\\times 10^7~M_\\odot$ (Peimbert \\& Torres-Peimbert 1981), $(3-8)\\times 10^5~M_\\odot$ (Filippenko \\& Sargent 1988), and $7\\times 10^5~M_\\odot$--$3\\times 10^6~M_\\odot$ (Ho et al. 1996). From stellar velocity dispersion measurements, Keel (1989) obtained an upper limit of $2\\times10^7 M_\\odot$, and Bower et al. (1996) obtained a lower limit of $3\\times10^6 M_\\odot$. Our result falls on the relatively higher side of these previous estimates, though still consistent considering large errors associated with various estimates. Our estimate reinforces the view that LLAGNs have low X-ray luminosities because they have very low accretion rates, rather than because they have less massive black holes than the more luminous AGNs. This further encourages us to interpret LLAGNs as a relic of the past luminous AGNs."
    },
    "0010/astro-ph0010197_arXiv.txt": {
        "abstract": "The results of new spectroscopic analyses of 20 recently reported extrasolar planet parent stars are presented.  The companion of one of these stars, HD\\,10697, has recently been shown to have a mass in the brown dwarf regime; we find [Fe/H] $= +0.16$ for it.  For the remaining sample, we derive [Fe/H] estimates ranging from $-0.41$ to $+0.37$, with an average value of $+0.18 \\pm 0.19$.  If we add the 13 stars included in the previous papers of this series and 6 other stars with companions below the 11 M$_{\\rm Jup}$ limit from the recent studies of Santos et al., we derive $\\langle$[Fe/H]$\\rangle = +0.17 \\pm 0.20$.  Among the youngest stars with planets with F or G0 spectral types, [Fe/H] is systematically larger than young field stars of the same Galactocentric distance by 0.15 to 0.20 dex.  This confirms the recent finding of Laughlin that the most massive stars with planets are systematically more metal rich than field stars of the same mass.  We interpret these trends as supporting a scenario in which these stars accreted high-Z material after their convective envelopes shrunk to near their present masses.  Correcting these young star metallicities by 0.15 dex still does not fully account for the difference in mean metallicity between the field stars and the full parent stars sample.  The stars with planets appear to have smaller [Na/Fe], [Mg/Fe], and [Al/Fe] values than field dwarfs of the same [Fe/H].  They do not appear to have significantly different values of [O/Fe], [Si/Fe], [Ca/Fe], or [Ti/Fe], though.  The claim made in Paper V that stars with planets have low [C/Fe/] is found to be spurious, due to unrecognized systematic differences among published studies.  When corrected for these differences, they instead display slightly enhanced [C/Fe] (but not significantly so).  If these abundance anomalies are due to the accretion of high-Z matter, it must have a composition different from that of the Earth. ",
        "introduction": "\\label{intro}  In our continuing series on stars-with-planets (hereafter, SWPs), we have reported on the results of our spectroscopic analyses of these stars (Gonzalez 1997, Paper I; Gonzalez 1998, Paper II; Gonzalez \\& Vanture 1998, Paper III; and Gonzalez et al. 1999, Paper IV; Gonzalez \\& Laws 2000, Paper V).  Other similar studies include Fuhrmann et al. (1997, 1998) and Santos et al. (2000b,c).  The most significant finding so far has been the high mean metallicity of SWPs, as a group, compared to the metallicity distribution of nearby solar-type stars (Gonzalez 2000; Santos et al. 2000b,c).  Additional extrasolar planet candidates continue to be announced by planet hunting groups using the Doppler method.  We follow-up these annoucements with high resolution spectroscopic observations as time and resources permit.  Herein, we report on the results of our abundance analyses of 20 new candidate SWPs.  We compare our findings with those of other recent similar studies, look for trends in the data suggested in previous studies, and evaluate proposed mechanisms in light of the new dataset. ",
        "conclusions": "Employing a sample of 38 SWPs with high quality spectroscopic abundance analyses, we find the following anomalies:  \\begin{itemize}  \\item The present results confirm the high average [Fe/H] of SWPs compared to nearby field stars.  The average [Fe/H] of the 38 SWPs with high quality spectroscopic [Fe/H] estimates is $+0.17 \\pm 0.20$.  \\item Among the youngest SWPs, which presumably have suffered the least amount of migration in the Milky Way's disk, most have [Fe/H] values about 0.15 dex larger than young field stars at the same mean Galactocentric distance.  Young SWPs more massive than $\\sim$1 M$_{\\odot}$, in particular, have the largest positive ``excess [Fe/H]'' relative to the field star trend.  \\item There does not appear to be significant differences in the [C/Fe] and [O/Fe] values between SWPs and field stars, though [C/Fe] may be somewhat high among the SWPs.  \\item We found evidence for smaller values of [Na/Fe], [Mg/Fe], and [Al/Fe] among SWP's compared to field stars of the same [Fe/H].  They do not appear to differ in [Si/Fe], [Ca/Fe], or [Ti/Fe].  \\end{itemize}  We have suggested the ``self-pollution'' scenario as an explanation for the anomalous trends among the F type SWPs.  However, it is not likely that this mechanism can account for the extremely high [Fe/H] values of 14\\,Her and $\\rho^{1}$ Cnc or of the highly evolved subgiant, HD\\,177830.  Given the trends uncovered in the present work and by Laughlin, one is virtually guaranteed of discovering a giant planet orbiting a young F dwarf with a [Fe/H] value $\\sim$0.25 dex greater than that of field stars at the same Galactocentric distance.  In this regard, we expect that a planet will be found orbiting HD\\,75332.\\footnote{Unfortunately, this star displays Doppler variations with an amplitude near 50 m~s$^{\\rm -1}$, which appear to be due to chromospheric activity (Geoff Marcy, private communication). A high level of chromospheric activity for this star is consistent with our age estimate for it.}  The following four super metal rich F or G0 dwarfs from Feltzing \\& Gustafsson (1998) also should be searched for planets: HD\\,71479, HD\\,87646, HD\\,110010, and HD\\,130087.  Given the trends uncovered in the present work, we believe the chances are high that these stars harbor giant planets.  It would be even more interesting if any of these stars do not have planets!"
    },
    "0010/astro-ph0010474_arXiv.txt": {
        "abstract": "Cosmic rays may provide opportunities for probing fundamental physics. For example, ultra-high-energy cosmic rays might originate from the decays of metastable heavy particles, and astrophysical $\\gamma$ rays can be used to test models of quantum gravity. Both scenarios offer ways to avoid the GZK cut-off. \\\\ \\begin{center} CERN-TH/2000-314 ~~~~~~ - ~~~~~~ astro-ph/y0010474 \\end{center} ",
        "introduction": " ",
        "conclusions": ""
    },
    "0010/astro-ph0010642_arXiv.txt": {
        "abstract": "We model the spin pulse intensity and hardness ratio profiles of the isolated neutron star RXJ0720.4--3125 using {\\sl XMM-Newton} data. The observed variation is approximately sinusoidal with a peak-to-peak amplitude of 15\\%, and the hardness ratio is softest slightly before flux maximum. By using polar cap models we are able to derive maximum polar cap sizes and acceptable viewing geometries. The inferred sizes of the caps turn out to be more compatible with a scenario in which the neutron star is heated by accretion, and place limits on the magnetic field strength. The hardness ratio modulation can then be explained in terms of energy-dependent beaming effects, and this constrains the acceptable models of the emerging radiation to cases in which softer photons are more strongly beamed than harder photons. An alternative explanation in terms of spatially variable absorption co-rotating in the magnetosphere is also discussed. ",
        "introduction": "\\label{intr}  From its high ratio of X-ray to optical flux, soft X-ray spectrum, 8.4~s X-ray period and location in the Galactic plane, Haberl et al.~(\\cite{hmb97:1997}) proposed that the ROSAT source RXJ0720.4--3125 is an isolated neutron star (NS). A probable optical counterpart has been identified (Motch \\& Haberl \\cite{mh98:1998}; Kulkarni \\& van Kerkwijk \\cite{kvk98:1998}). Paerels et al.~(\\cite{p00:2001}, Paper~1) present spectra of RXJ0720.4--3125 using {\\it XMM-Newton}, and find that there is no evidence for absorption lines and edges in the X-ray spectrum. They are able to refine the ROSAT spectral fits using black-body spectra of 86.2$\\pm$0.3 eV absorbed by a cold absorber of $6.0 \\pm 0.4 \\times 10^{19}$~cm$^{-2}$.  Paper~1 mainly addressed the spectral characteristics and the behaviour of RXJ0720.4--3125 averaged over the 8.4~s spin period. In this paper we examine the implications of the phase-resolved data. ",
        "conclusions": "By exploiting the superior quality of the data obtained from {\\it XMM-Newton \\/}, we have been able to present a detailed analysis of the spin pulse profile of RXJ0720.4--3125. We have modelled these data by using polar cap models based on the Pechenick et al.~(\\cite{pfc83:1983}) formulation. The smooth nature of the data does not allow a unique conclusion about the best fit values to be reached, but nevertheless we are able to exclude a large volume of the parameter space and to derive an upper limit for the size of the emitting region. The derived values are at best marginally compatible with an interpretation in terms of surface temperature variation during the cooling phase of a middle-age NS. An accretor model is more satisfactory in accounting for the intensity plus hardness-ratio variations. Accreting NS appear to be relatively rare objects, as recently found by Popov et al.~\\cite{pct00:2000} from a statistical evolutionary analysis. RXJ0720.4--3125 may therefore be more important than perhaps initially appreciated.  We have investigated a number of possible explanations for the intensity and spectral variations over the spin period. The long period of RXJ0720.4--3125 is somewhat unusual and hints at a possible evolutionary link between dim NS, soft gamma-ray repeaters (SGRs) and anomalous X-ray pulsars (AXPs) (Haberl et al.~\\cite{hmb97:1997}, Alpar~\\cite{a00:2000}), whereby isolated NS in the propellor phase are the progenitors of AXPs and SGRs. If the hardness ratio variation is indeed caused by absorption, then the large amount of material close by to the neutron star may be explained in terms of matter being propelled outwards. However the luminosity in this phase is caused by energy dissipation in the neutron star, and we have already found (Section 3.3) this requires polar cap sizes which are uncomfortably large. Moreover, such an evolutionary link would produce significantly larger numbers of AXPs and SGRs (Mereghetti, private communication).  The most plausible interpretation is that the observed modulation is produced by beaming effects in the emerging radiation. The angular distribution of the emerging radiation is still a matter for further investigation, and is extremely sensitive to details in the radiative computations and to a number of assumptions about chemical composition and magnetic field (see e.g. Zavlin et al.~\\cite{zps96:1996}), Rajagopal et al.~\\cite{rrm97:1997}). Our modelling provides observational inputs for these investigations by suggesting that the softer X-rays from the heated cap are more strongly beamed than the harder X-rays."
    },
    "0010/astro-ph0010368_arXiv.txt": {
        "abstract": "At this moment, the use of SNIa for cosmology lies on the assumption that the SNe at high redshifts are equal to the local ones. However, some observations indicate a correlation between light curve (LC)  properties and the morphological type of the host galaxy. This could indicate a dependence with the age (mass/composition) of the underlying population. In this work we have chosen the delayed detonation explosion model in CO Chandrasekhar mass WDs to explore the dependence of the SNIa LC and nucleosynthesis with the initial mass and composition of the WD progenitor. The progenitor influences the final SNIa via the mass of the CO core formed and the C/O ratio within it (1D explosion models). We have followed the evolution of stars with masses between 1.5 and 8 M$_\\odot$ and metallicity, Z=0, 10$^{-5}$, 0.001 and 0.02, from the pre-main sequence to the TP-AGB phase. The differences obtained in the final C/O ratio within the explosive WD are smaller than 22$\\%$. This results in a difference at maximum of 0.03 mag and of 0.1 mag when the  brightness-decline relation is applied. ",
        "introduction": "The increasing quality and quantity of Type Ia observations, indicate that they are not such an homogeneous class of events as it was previously thought. Moreover, several interesting correlations between observed properties are obtained (see Leibundgut this volume).  These correlations could give hints about the underlying explosion mechanisms and progenitors. To understand the event and the reasons of the differences is a pre-requisite to use SNIa for cosmology (at least, for us). The observations of SNIa at high redshifts have lead to the conclusion that the expansion of the Universe is accelerating \\cite{Ri98,Pe99}. Together with the Cosmic Microwave Background  observations, the mass density parameter and the cosmological constant have been estimated, $\\Omega_{m}\\sim$0.3 and $\\Omega_{\\lambda}\\sim$0.7.  There are some observational evidences that  indicate a relation of SNIa with the morphological type of the host galaxy \\cite{Br98,Wa97,Ha00}. In ellipticals, as compare with spirals, less events are found, they are more homogeneous (in maximum brightness and decline rate of the light curve) and in general, they are dimmer. This could mean a correlation with the progenitor population and consequently  evolution in time.  Different evolutionary paths could lead to an explosive degenerate WD  and several explosion mechanisms could blow up this WD (see Woosley this volume). Probably each of them contribute {\\em up to some degree} to the SNIa population.  In this work we have decided to fix both, the scenario and explosion mechanism and just change the initial mass and composition of the WD progenitor.  We have chosen the delayed detonation (1D) explosion of a Chandrasekhar mass WD because this scenario accounts for most of the observational constraints. The key parameter is the transition density, $\\rho_{tr}$, the density at which the deflagration turns into a detonation. Varying $\\rho_{tr}$ the optical and IR light curves and spectra evolution of normal and subluminous SNIa are reproduced. Subluminous SNIa are found to be redder,  as observed. Moreover, the brightness-decline relation (LCs of brighter events decline slower) used in all the cosmological applications, is obtained \\cite{Ho95,Ho96}. Other correlations, like the CaII H+K  $\\&$ Si minimum velocities and mean Ni velocity with brightness are as well reproduced. ",
        "conclusions": ""
    },
    "0010/astro-ph0010312_arXiv.txt": {
        "abstract": "The bright stellar content near the center of the Local Group elliptical galaxy M32 is investigated with 0.12 arcsec FWHM $H$ and $K$ images obtained with the Gemini Mauna Kea telescope. Stars with $K = 15.5$, which are likely evolving near the tip of the asymptotic giant branch (AGB), are resolved to within 2 arcsec of the nucleus, and it is concluded that the peak stellar brightness near the center of M32 is similar to that in the outer regions of the galaxy. Moreover, the projected density of bright AGB stars follows the visible light profile to within 2 arcsec of the nucleus, indicating that the brightest stars are well mixed throughout the galaxy. Thus, there is no evidence for an age gradient, and the radial variations in spectroscopic indices and ultraviolet colors that have been detected previously must be due to metallicity and/or some other parameter. We suggest that either the bright AGB stars formed as part of a highly uniform and coherent galaxy-wide episode of star formation, or they originated in a separate system that merged with M32. ",
        "introduction": "As the closest galaxy with characteristics reminiscent of massive classical ellipticals (e.g. Wirth \\& Gallagher 1984, Kormendy 1985), M32 provides an unprecedented laboratory for probing the stellar contents of early-type galaxies, and testing predictions drawn from integrated spectra. The star-forming history of M32 is a matter of on-going debate, although the bulk of evidence seems to indicate that an intermediate-age population is present. The spectrum of the central few arcsec of M32 can not be represented by that of an old star cluster population (O'Connell 1980; Rose 1984; Davidge 1990; Bica, Alloin, \\& Schmidt 1990) and, while Cole et al. (1998) argue that the integrated UV -- visible color of M32 is consistent with an old population, they also stress the considerable uncertainties present in the models. The spectrum of M32 shows deep H$\\beta$ absorption (Burstein et al. 1984), and the origin of this feature is of critical importance for establishing if an intermediate-age component is present. Brown et al. (2000) resolved hot horizontal branch (HB) stars near the center of M32 and, after finding that these objects can not on their own explain the strength of H$\\beta$ absorption, conclude that it will be difficult to explain this feature without resorting to a young or a large blue straggler population. Finally, Guarnieri, Renzini, \\& Ortolani (1997) investigated the stellar content of M32 using the metal-rich globular cluster NGC 6553 as a benchmark, and found that the brightest infrared source in this cluster is a long-period variable that, if viewed at the distance of M32, is more than 0.5 mag fainter in $K$ than the brightest stars in M32 (Davidge 2000).  With a few exceptions (Brown et al. 2000; Davidge 2000), crowding has restricted studies of resolved stars to the outer regions of M32 (Freedman 1989, 1992; Elston \\& Silva 1992; Davidge \\& Jones 1992; Grillmair et al. 1996), and population gradients (Cohen 1979; Davidge 1991; O'Connell et al. 1992; Hardy et al. 1994; Brown et al. 1998; Ohl et al. 1998) complicate efforts to tie together predictions made from resolved stars and integrated spectra. In fact, the ultraviolet-visible color of M32 changes by 1.5 mag in the central 10 arcsec (Ohl et al. 1998), thus demonstrating the importance of pushing studies of the resolved stellar content to smaller and smaller radii. In the current paper, images with 0.12 arcsec FWHM are used to resolve bright AGB stars to within 2 arcsec of the galaxy center; for comparison, the majority of AGB-tip stars resolved by Davidge (2000) are at distances in excess of 10 arcsec from the nucleus. ",
        "conclusions": "High angular resolution near-infrared images have been used to study the bright stellar content near the center of the Local Group elliptical galaxy M32. Individual bright stars are resolved to within 2 arcsec of the nucleus, and it is concluded that, with the possible exception of the central 2 arcsec of the galaxy, where stars have not yet been resolved, the AGB-tip remains constant at $K = 15.5$ mag throughout the galaxy. This result, coupled with the spatial distribution of the AGB component (see below), suggests that the peculiar ultraviolet-visible color profile of M32 (O'Connell et al. 1992, Ohl et al. 1998) is not due to an age gradient. The brightest stars in the infrared evidently do not trace the population that dominates in the ultraviolet.  AGB-tip stars are resolved close to the nucleus of M32 because they are 2.5 mag brighter in $K$ than the RGB-tip, where the number density of stars increases dramatically. Moreover, AGB stars are part of a rapid evolutionary phase, and the density of stars within one mag of the AGB-tip, which is the portion of the LF where stars are most likely to combine to produce sources with brightnesses comparable to, or brighter than, the AGB-tip, is relatively low. The low density of bright AGB stars can be demonstrated using star counts from the outer region of the galaxy. Davidge (2000) found 5 stars within one mag of the AGB-tip in a 0.25 arcmin$^2$ field 2.3 arcmin from the nucleus and, after scaling according to the Kent (1987) r-band surface brightness profile, the predicted densities of bright AGB stars near the nucleus are 0.008 (7.3 -- 13.4 arcsec), 0.016 (4 -- 7.3 arcsec), and 0.038 (2 -- 4 arcsec) stars per 0.06 arcsec radius resolution element. The number of blends expected in each radial interval is then 0.01, 0.05, and 0.2. The incidence of blending climbs very quickly for fainter objects.  At a fixed age, the brightness of the AGB-tip is expected to vary with metallicity. However, metallicity is likely not a major consideration when comparing AGB-tip brightnesses throughout M32. Long-slit spectroscopic observations indicate that $\\Delta$Mg$_2 = -0.06$ between the Davidge (2000) outer field and 2 arcsec if $\\Delta$Mg$_2$/$\\Delta$r$=-0.03$ (Hardy et al. 1994), and this corresponds to $\\Delta$[Fe/H] = 0.3 -- 0.4 dex using Worthey's (1994) solar metallicity models. Bertelli et al. (1994) modelled the evolution of stars up to and including the AGB. These models predict that M$_{K}^{AGB-tip}$ changes by only 0.1 -- 0.2 mag between z = 0.008 and z = 0.020, in the sense of becoming brighter towards higher metallicities.  The spatial distribution of bright AGB stars follows the r-band surface brightness profile measured by Kent (1987) to within 2 arcsec of the nucleus, indicating that these stars are uniformly mixed throughout the galaxy. If the brightest stars are younger than the underlying body of the galaxy, then this result could indicate that these objects formed throughout the galaxy during a remarkably coherent episode of star formation, which would require that the star-forming material cooled in such a way as to follow the mass profile of the underlying galaxy. This is evidently not the case in most early-type galaxies, which typically harbour an age gradient of size $\\Delta$log$t_{Gyr}$/$\\Delta$logR = 0.1 (Henry \\& Worthey 1999). Alternatively, a uniformly mixed population of bright stars is consistent with a coeval system, which is a model for M32 suggested by del Burgo et al. (2000).  Many nearby elliptical galaxies have experienced mergers, and so it is worth investigating if the bright stars, which presumably formed during intermediate epochs, were part of another system that was subsequently accreted by M32. If the intermediate-age component accounts for 10\\% of M32 by mass (Davidge 2000), then the accreted galaxy would have had an integrated brightness M$_B \\leq -13.5$ mag, which overlaps with those of Local Group dwarf galaxies. If merger-induced mixing did occur in M32 then (1) the age of the brightest AGB stars predicted by models indicates that any redistribution of stellar content must have occured on a timescale of only a few Gyr (Davidge 2000), and (2) any mixing would have to act preferentially on the intermediate-age population, and not annihilate the metallicity gradient. Could an interaction between two galaxies of different masses redistribute the stellar content of the smaller object, while not completely removing population gradients in the larger system? Models of non-dissipative mergers between equal-mass systems indicate that the progenitor populations are not fully mixed by such interactions (White 1980). However, the extent of mixing may be very different if the mass ratio differs significantly from unity. Indeed, models in which a low-mass galaxy interacts with a disk indicate that the stars in the smaller system can be redistributed in much less than a Hubble time (Helmi \\& White 1999), while not destroying the metallicity gradient in the progenitor disk (Quinn, Hernquist, \\& Fullagar 1993). Hernquist \\& Quinn (1988) modelled the interaction between a low mass system and larger spherical galaxy, and found that shells may be produced as the smaller object is consumed by the larger system. M32 does not contain shells, and structures of this nature may have been removed by interactions with M31.  A problem with the merging model is that the time scale for dynamical friction depends on the inverse of system mass (e.g. Binney \\& Tremaine 1987), and it is not clear that low-mass systems can merge in only a few Gyr. A further complication is that if the AGB stars formed in a low-mass system, then they would likely be more metal-poor than the majority of stars in M32. If this is the case then the intermediate-age component would be even younger than estimated by Davidge (2000), who assumed a solar metallicity, thereby producing even tighter requirements on the mixing time scale.  \\vspace{0.3cm} The development of Hokupa'a was supported by the NSF. Sincere thanks are extended to Sidney van den Berg, Jim Hesser, and Liese van Zee for commenting on an earlier draft of this paper. An anonymous referee also provided suggestions that improved the paper.  \\clearpage"
    },
    "0010/astro-ph0010548_arXiv.txt": {
        "abstract": "We have numerically investigated the dynamics of how a close stellar fly-by encounter of a symmetrical circumstellar planetesimal disc can give rise to the many kinds of asymmetries and substructures attributed to the dusty disc of $\\beta$ Pic. We find three recognizable groupings of test particles that can be related to the morphology of the $\\beta$ Pic disc. These are: highly eccentric and inclined orbit particles that reach apocentre in the southwest, moderately eccentric and inclined orbit particles that reach apocentre in the northeast, and a relatively unperturbed region inside $\\sim$200\\,AU radius. ",
        "introduction": "The starlight of $\\beta$ Pic scatters off dust grains in its edge-on circumstellar disc (Smith \\& Terrile 1984). Imaging reveals that the disc exhibits asymmetrical structure with respect to an ideal disc on projected radial scales from $\\sim$50\\,AU (Burrows et al. 1995; Mouillet et al. 1997) to $\\sim$1000\\,AU (Smith \\& Terrile 1987). Sensitive $R$-band images (Larwood \\& Kalas 2000) show the northeast (NE) extension can be detected out to $1835$\\,AU, whereas the southwest (SW) extension is detectable out to just $1450$\\,AU, corresponding to a $\\sim$25 per cent length asymmetry. More recently, the longer NE midplane extension has been discovered to carry several brightness enhancements from $\\sim$500 to $800$\\,AU (Kalas et al. 2000), that have been attributed to a series of rings within the disc's midplane. Similar features in the SW extension were not detected and so eccentric rings were hypothesised. Kalas et al. (2000) presented a dynamical model in which the stellar fly-by scenario (Kalas \\& Jewitt 1995) for generating the disc asymmetries was applied to the $\\beta$ Pic system. Their finding was that the asymmetry types present in the $\\beta$ Pic disc have analogues in the dynamical response of a particle disc to a perturbing stellar fly-by encounter. Furthermore, they demonstrated that these features could occur simultaneously with the formation of eccentric circumstellar rings.  As in the other works mentioned above we assume that the simulation particles' distribution represents the distribution of an underlying disc of planetesimals, which are parent bodies whose infrequent and therefore dynamically inconsequential collisions supply the dust scattering surface that is observed in the real system. A more complete model should include consideration of grain removal and generation processes in determining the appearence of the perturbed disc. However, we defer treatment of those issues to future work, and focus here on the first-stage problem of the dynamics. This stage is important in deducing the mass and orbital parameters of the postulated stellar perturber, for which it is possible to perform star catalogue searches (Kalas, Deltorn \\& Larwood 2000), and we proceed to do this by examining the length asymmetry, and other measures of the disc response, as a function of stellar encounter parameters. ",
        "conclusions": "The dynamical response of the models suggests that the $\\beta$ Pic disc might be separable into different components corresponding to groupings of perturbed particle orbits. For example, we find that outwards of $\\sim$250\\,AU the SW extension may not have a distinct midplane owing to pumping of particle inclinations. The positions close to pericentre of the moderately eccentric NE midplane particles could account for the measurement of a SW midplane out to $\\sim$650\\,AU, in superposition with the more diffuse vertically and radially extended component. When the optical image is vertically compressed (Larwood \\& Kalas 2000) we are able to measure the SW extension to much larger radii ($1450$\\,AU). In our model the corresponding (very eccentric and inclined) particle orbits also intersect the disc near to their pericentres (down to $\\sim$50\\,AU) possibly accounting for the inferred inner warp of the disc, which is aligned with the outer flared envelope in the SW. In summary, there are three recognizable particle groupings that can be related to the morphology of the $\\beta$ Pic disc. These are: the highly eccentric and inclined particles that reach apocentre in the SW, the moderately eccentric and low inclination orbiting particles that reach apocentre in the NE, and the relatively unperturbed particles inside $\\sim$200\\,AU radius."
    },
    "0010/astro-ph0010581_arXiv.txt": {
        "abstract": "We review the various techniques through which wind properties of massive stars -- O stars, AB supergiants, Luminous Blue Variables (LBVs), Wolf-Rayet (WR) stars and cool supergiants -- are derived. The wind momentum-luminosity relation (e.g. Kudritzki et al. 1999) provides a method of predicting mass-loss rates of O stars and blue supergiants which is superior to previous parameterizations. Assuming the theoretical $Z^{0.5}$ metallicity dependence, Magellanic Cloud O star mass-loss rates are typically matched to within a factor of two for various calibrations. Stellar winds from LBVs are typically denser and slower than equivalent B supergiants, with exceptional mass-loss rates during giant eruptions $\\dot{M}=10^{-3} \\ldots 10^{-1} M_{\\odot}$ yr$^{-1}$ (Drissen et al. 2001). Recent mass-loss rates for Galactic WR stars indicate a downward revision of 2--4 relative to previous calibrations due to clumping (e.g. Schmutz 1997), although evidence for a metallicity dependence remains inconclusive (Crowther 2000). Mass-loss properties of luminous ($\\ge 10^{5} L_{\\odot}$) yellow and red supergiants from alternative techniques remain highly contradictory. Recent Galactic and LMC results for RSG reveal a large scatter such that typical mass-loss rates lie in the range $10^{-6} \\ldots 10^{-4} M_{\\odot}$ yr$^{-1}$, with a few cases exhibiting $\\sim 10^{-3} M_{\\odot}$ yr$^{-1}$. ",
        "introduction": "Winds are ubiquitous in massive stars, although the physical processes involved  depend upon the location of the star within the H-R diagram. Mass-loss crucially affects the evolution and fate of a massive star (Meynet et al. 1994), while the momentum and energy expelled contribute to the dynamics and energetics of the ISM. The interested reader is referred to Lamers \\& Cassinelli (1999) for the topic of stellar winds in general, or Kudritzki \\& Puls (2000) for a detailed discussion of mass-loss from OB stars. This review will cover the broad topics of mass-loss from early-type (OBA, Wolf-Rayet, Luminous Blue Variable) and late-type (yellow/red supergiants) stars. Theoretical predictions for hot star winds follow line-driven radiation pressure (Castor et al. 1975; Pauldrach et al. 1986). Pulsations are thought to initiate the winds of cool supergiants, together with other mechanisms (e.g. Alfven waves). Radiation pressure on dust grains helps to maintain outflows in their cool, outer envelopes (Wickramasinghe et al. 1966). ",
        "conclusions": ""
    },
    "0010/astro-ph0010254_arXiv.txt": {
        "abstract": "We critically examine the use of the Washington photometric system (with the 51 filter) for identifying distant halo giants.  While this is the most powerful photometric technique for isolating G and K giant stars, spectroscopic follow-up of giant candidates is vital.  There are two situations in which interlopers outnumber genuine giants in the diagnostic \\m51/\\mt2\\ plot, and are indistinguishable photometrically from the giants. (1) In deep surveys covering tens of square degrees, very metal-poor halo dwarfs are a significant contaminant.  An example is our survey of the outer halo \\citep{spag1,spag2}, where these metal-poor dwarfs dominate the number of photometric giant candidates at magnitudes fainter than $V$ = 18 and cannot be isolated photometrically. (2) In deep surveys of smaller areas with low photometric precision, most objects in the giant region of the color-color plot are dwarfs whose photometric errors have moved them there. Color errors in \\m51\\ and \\mt2\\ need to be smaller than 0.03 mag to avoid this problem.  An example of a survey whose photometric errors place the giant identifications under question is the survey for extra-tidal giants around the Carina dwarf spheroidal of \\citet{srmcarina}.  Accurate photometry and spectroscopic follow-up of giant candidates are essential when using the Washington system to identify the rare outer-halo giants. ",
        "introduction": "Photometric searches over wide areas for rare, distant giants in the Galactic halo are now a reality, thanks to the advent of wide-field CCD systems. One particularly useful photometric system is the Washington system \\citep{canterna,harriscan}. Since \\citet{doug84} added the DDO ``51'' filter to allow discrimination between late-type dwarfs and giants and re-calibrated its \\fe\\ indicator, this system has steadily gained in popularity. Its broad-band filters and strong abundance sensitivity for late-type giants \\citep{doug91,pb94} mean that it was well-suited to studies of objects of known distance such as open and globular cluster giants. In these cases there is little need for luminosity discrimination.  However, when seeking rare objects such as halo giants, which are seen through a thick veil of foreground disk stars, luminosity discrimination is vital. The \\mgb/MgH feature near 5170\\AA\\ is strong in late-type dwarfs and weak in giants.  The ``51'' filter (with $\\lambda_c$ = 5130\\AA\\ and $\\Delta\\lambda$ = 154\\AA) is centered on this feature.  It works well for discriminating between late G/K dwarfs of solar or near-solar metallicity and G/K giants.  The addition of this filter has allowed the use of the Washington system in surveys for late G and K giants where foreground contamination by disk dwarfs is a problem.  This is superior to other photometric methods of identifying giants such as DDO or Str\\\"omgren photometry \\citep{ddo,bond80} because only one filter has a narrow passband, and it it not necessary to observe in the ultraviolet.  Surveys well-suited to the Washington system include:  \\begin{itemize} \\item studies of giant stars in clusters at low latitude, where it is now possible to remove foreground dwarf non-members \\citep[eg][]{doug97}  \\item surveys for rare field giants of the outer halo (Morrison et al.\\ 2000a, Paper I hereafter; Majewski et al.\\ 2000b)  \\item surveys for extra-tidal giants from Galactic dwarf  satellites \\citep{srm3,srmcarina}  \\end{itemize}  The last two types of survey make contributions to our understanding of the formation and subsequent evolution of the Galaxy. Distant halo field giants, such as the one recently discovered by our survey at 90 kpc \\citep{edo800} provide in-situ probes of the outer halo, and are also important probes of the Galactic potential \\citep{zaritsky}. But their identification as giants must be solid in order to make them useful -- the spatial distribution and kinematics of local halo dwarfs will contribute little to our understanding of the outer Galaxy. Indeed, misidentification will lead to serious errors.  We began work on this paper in order to understand our low success rate at identifying halo giants for $V$ magnitudes fainter than 18. For brighter magnitudes our success rate, confirmed by spectroscopic follow-up, is better than 90\\%, but this falls precipitously to 24\\% for $V$ magnitudes between 18 and 19 \\citep{spag2}, despite comparably good mean photometric uncertainties for these fainter stars.  A detailed investigation into the limitations of the Washington luminosity classification may also cast light on some surprising claims from the surveys around dwarf satellites of the Milky Way. For example, \\citet{srmcarina} have claimed  detection of  large numbers of extra-tidal stars in the Carina dwarf spheroidal galaxy. Such a large extra-tidal population, if confirmed, has implications not only for galaxy evolution but also for our understanding of the dark matter content of the dSph satellites. However, these surveys extend to much fainter magnitudes than previously possible, and it is important to understand possible contamination by foreground and background objects, and the role of photometric errors, at such faint levels.  Our essential result is that there are two major problems with the Washington luminosity classification: first,  very metal-poor foreground subdwarfs are indistinguishable photometrically from halo giants; second, photometric errors can scatter disk dwarfs into the halo giant region of the \\mt2/\\m51 diagnostic plot.  In Section 2 we discuss possible contaminants of halo giant candidates in the \\mt2/\\m51 diagram: disk and halo dwarfs, galaxies and QSOs. With accurate \\m51 photometry, the only serious contaminants are the very metal-poor halo dwarfs whose weak lines make them indistinguishable photometrically from halo giants. We derive a luminosity function specifically for such metal-poor halo dwarfs and use it, with a simple halo model, to predict the number of subdwarfs per \\degsq\\ in different directions in the Galaxy.  Although foreground disk and thick disk dwarfs are neglible contaminants if accurate photometry is obtained, the high density of these stars, especially at low latitudes, makes the number scattered into the giant region of the \\mt2/\\m51\\ diagram a sensitive function of photometric error. In Section 3 we investigate this in detail, and describe the double effect of large errors on giant searches: not only are dwarfs scattered into the giant region, but genuine giants are scattered out.  In Section 4 we use what we have learned about giant selection to examine our own pencil-beam survey of at high galactic latitudes \\citep{spag2}. The BTC subset (obtained using the Big Throughput Camera on the CTIO 4m in April 1999) covers 12.7 square degrees, and has spectroscopic follow-up of approximately half of the giant candidates \\citep{spag6}. We find that foreground subdwarfs are the principal cause of our low detection efficiency for giants below $V=18$, and describe our plans to optimize future searches for the faint, red giants of the extreme outer halo so that fewer subdwarfs are misidentified as giants.  In Section 5 we discuss the survey of \\citet{srmcarina} (MOPKJG hereafter) of approximately two square degrees centered on the Carina dwarf spheroidal galaxy. Their study found a large number ($\\sim$100) of giant candidates outside the tidal radius of \\citet{ih95}, but they were only able to obtain follow-up spectra of three of the brightest extra-tidal candidates. The photometric errors of the MPOKJG data are more typical of photographic than of CCD photometry, and we discuss the possible effect of these errors on the giant selection. ",
        "conclusions": "Washington photometry, with the addition of the 51 filter, is a powerful method of finding distant halo giants. We find that at the faintest magnitudes probed by our survey, our giant selection efficiency falls, and we make a careful examination of possible problems with the Washington giant classification to improve future surveys. We show that accurate photometry (color errors less than 0.03 mag) is essential to cut down on the scattering of the numerous foreground dwarfs into the giant region in the \\m51/\\mt2\\ color-color plot. We have made estimates of the contamination of giant candidates by very metal-poor halo dwarfs by deriving a new luminosity function specifically for the subdwarfs which mimic halo giants in the \\m51/\\mt2\\ diagram, and using it in a simple halo model.  We used our results on luminosity discrimination with the Washington system to examine two recent surveys: our own survey of 12.7 square degrees at high galactic latitudes \\citep{spag1,spag2}, and the study of \\citet{srmcarina} of fields centered on the Carina dwarf spheroidal galaxy, which finds over 100 Carina-associated giant candidates outside its tidal radius.  For surveys such as our own of the halo field, very metal-poor halo dwarfs are the major source of contamination of candidate giant samples, and we propose using a smaller region of the \\m51/\\mt2\\ diagram to optimise spectroscopic follow-up at the faint magnitudes where the subdwarfs dominate.  The survey of \\citet{srmcarina} has spectroscopic follow-up only for three of the brightest ``extra-tidal'' giant candidates, and we show that their photometric errors, which will increase as stars become fainter, can have a very serious effect on their giant classification.  In summary, spectroscopic follow-up of giant stars identified using the Washington system is vital. If one wishes to make statistical corrections for the number of bogus giants caused by photometric errors, it is important to quantify accurately the photometric errors in the data, both in terms of the size of the average error and the shape of the distribution."
    },
    "0010/astro-ph0010458_arXiv.txt": {
        "abstract": "We present a three-component mixing model for the evolution of O abundance relative to Fe, taking into account the contributions of the first very massive stars (with masses of $\\gtrsim 100\\,M_\\odot$) formed from Big Bang debris. We show that the observations of O and Fe abundances in metal-poor stars in the Galaxy by Israelian et al. (1998) and Boesgaard et al. (1999) can be well represented both qualitatively and quantitatively by this model. We use the representation of the number ratios (O/Fe) versus 1/(Fe/H). In this representation, if there is only a single source with a fixed production ratio of O to Fe beyond a certain point, the subsequent evolution of (O/Fe) is along a straight line segment. Under the assumption of an initial Fe ([Fe/H]~$\\sim -3$) and O inventory due to the prompt production by the first very massive stars, the data of Israelian et al. (1998) and Boesgaard et al. (1999) at $-3\\lesssim$~[Fe/H]~$\\lesssim -1$ are interpreted to result from the addition of O and Fe only from type II supernovae (SNII) to the prompt inventory. At [Fe/H]~$\\gtrsim -1$, SNII still contribute O while both SNII and type Ia supernovae contribute Fe. During this later stage, (O/Fe) sharply drops off to an asymptotic value of $\\sim 0.8$~(O/Fe)$_{\\odot}$. The value of (O/Fe) for the prompt inventory at [Fe/H]~$\\sim -3$ is found to be (O/Fe)~$\\sim 20$~(O/Fe)$_\\odot$.  This result suggests that protogalaxies with low ``metallicities'' should exhibit high values of (O/Fe). The C/O ratio produced by the first very massive stars is expected to be $\\ll 1$ so that all the C should be tied up as CO and that C dust and hydrocarbon compounds should be quite rare at epochs corresponding to [Fe/H]~$\\lesssim -3$. ",
        "introduction": "In a recent theoretical study of Fe and $r$-process abundances in the early Galaxy by Wasserburg \\& Qian (2000a), it was proposed that the first stars formed after the Big Bang were very massive ($\\gtrsim 100\\,M_\\odot$) and promptly enriched the interstellar medium (ISM) to [Fe/H]~$\\sim -3$, at which metallicity formation of regular stars (with masses of $\\sim 1$--$60\\,M_\\odot$) took over. Subsequent Fe enrichment was provided initially by a subset of Type II supernovae (SNII). The interpretations of Wasserburg \\& Qian (2000a) were based on observations of metal-poor stars in the Galaxy by Gratton \\& Sneden (1994), McWilliam et al. (1995), McWilliam (1998), and Sneden et al. (1996, 1998). The critical metallicity [Fe/H]~$\\sim -3$ deduced by Wasserburg \\& Qian (2000a) for transition from formation of very massive stars to regular stars is in remarkable coincidence with the lower bound of [Fe/H]~$\\sim -2.7$ observed in damped Ly$\\alpha$ systems over a wide range in redshift (e.g., Prochaska \\& Wolfe 2000). Furthermore, Wasserburg \\& Qian (2000b) showed that the [Fe/H] of damped Ly$\\alpha$ systems can be explained in a straightforward manner by addition of Fe from SNII to the prompt inventory provided by the first very massive stars and that the evolution of [Fe/H] with redshift reflects the chronology of turn-on of protogalaxies.  The first very massive stars would certainly produce elements other than Fe. One of the issues to be resolved is the general problem of the nucleosynthetic signatures of these stars. This problem was discussed in the early work by Ezer \\& Cameron (1971). More recently, studies on condensation of Big Bang debris (e.g., Bromm, Coppi, \\& Larson 1999) indicated that protostellar aggregates of $\\sim 100\\,M_\\odot$ may be formed. Evolution of zero-metallicity stars of $\\gtrsim 100\\,M_\\odot$ was discussed by Fryer, Woosley, \\& Heger (2000). Calculations by Heger, Woosley, \\& Waters (2000) suggest that C, N, O, Mg, Si, and Fe are produced in large amounts by these stars. The question is how the theoretical models of nucleosynthesis in the first very massive stars formed from Big Bang debris are related to the observations. Another issue concerns the Fe enrichment by Type Ia supernovae (SNIa). It was assumed by Wasserburg \\& Qian (2000a,b) that SNIa started to contribute Fe at [Fe/H]~$\\sim -1$ and were responsible for $\\sim 2/3$ of the solar Fe inventory. In a self-consistent Fe enrichment model, the history of SNIa Fe contribution needs to be examined more carefully based on observations.  In this paper we present a phenomenological model for the evolution of O abundance relative to Fe. Our approach differs from previous studies (e.g., Timmes, Woosley, \\& Weaver 1995) in the following respects. Previous studies rely on many inputs such as the nucleosynthetic yields predicted by various models of SNII and SNIa and some description of star formation. In the approach here we attempt to identify a chemical evolution model that can account for the observed evolution of O abundance relative to Fe in terms of several key parameters. We hope that this approach will give a simple yet complete interpretation of the data. The usual representation of the O and Fe data is to display [O/Fe] against [Fe/H]. In the approach here we use (O/Fe) versus 1/(Fe/H), with the round brackets indicating the number ratios. If there is only a single source with a fixed production ratio of O to Fe beyond a certain point, the subsequent evolution of (O/Fe) as a function of 1/(Fe/H) is along a straight line segment. As will be shown, this representation also facilitates a simple geometric means to describe the evolution of O and Fe abundances in the general case. As the first very massive stars are thought to be a source of O and Fe (Heger et al. 2000), the value of (O/Fe) at [Fe/H]~$\\lesssim -3$ should serve as a nucleosynthetic signature of these stars, thus addressing the first issue raised above. Subsequent to the prompt enrichment by the very massive stars, SNII are essentially the only source of O, while Fe enrichment was provided by SNII initially and augmented by the contribution from SNIa only at a later time. Thus the evolution of O abundance relative to Fe at [Fe/H]~$>-3$ provides a means of addressing questions concerning the onset and nature of SNIa Fe contribution.  Based on the observations of metal-poor stars in the Galaxy by Israelian, Garcia Lopez, \\& Rebolo (1998) and Boesgaard et al. (1999), the value of (O/Fe) resulting from the prompt production by the first very massive stars is $\\sim 20$~(O/Fe)$_\\odot$. Comparison of our model with the same observations gives a history of SNIa Fe contribution that is consistent with the results of previous studies (e.g., Timmes et al. 1995) and with the assumptions of Wasserburg \\& Qian (2000a,b). The average production of O relative to Fe for SNII is found to be (O/Fe)$_{\\rm II}\\sim 3$~(O/Fe)$_\\odot$, which is $\\sim 7$ times lower than that for the very massive stars. Our model is presented in \\S2 and compared with observations in \\S3. Discussion and conclusions are given in \\S4. ",
        "conclusions": "We have presented a phenomenological model for the evolution of O abundance relative to Fe. In this model, the first very massive stars formed from Big Bang debris provided a prompt inventory of Fe ([Fe/H]$_{\\rm P}\\sim -3$) and  O, thus defining an initial state of the ISM. Subsequent O enrichment was provided by SNII that are related to regular massive stars, while subsequent Fe enrichment was provided by SNII followed by additional contribution from SNIa. Prior to the onset of SNIa Fe contribution, the evolution of (O/Fe) as a function of 1/(Fe/H) is along a straight line segment (Fig. 1). Subsequent evolution is governed by the rate per unit mass of gas for Fe production by SNIa relative to SNII (Fig. 1). Based on the observations of Israelian et al. (1998) and Boesgaard et al. (1999), the value (O/Fe)$_{\\rm P}$ resulting from the prompt production by the first very massive stars is $\\sim 20$~(O/Fe)$_\\odot$. Comparison of our model with the same observations indicate that SNIa began to contribute Fe at [Fe/H]~$\\sim -1$ and were responsible for $\\sim 2/3$ of the solar Fe inventory, consistent with the results of previous studies (e.g., Timmes et al. 1995) and with the assumptions of Wasserburg \\& Qian (2000a,b). The average production of O relative to Fe for SNII is found to be (O/Fe)$_{\\rm II}\\sim 3$~(O/Fe)$_\\odot$, which is $\\sim 7$ times lower than that for the first very massive stars.  The model developed here neither depends on nor gives numerical values of the rates per unit mass of gas for O and Fe production by SNII, $d\\,{\\rm (O/H)}_{\\rm II}/d\\,t$ and $d\\,{\\rm (Fe/H)}_{\\rm II}/d\\,t$. Likewise, the model only involves $\\gamma(t)$, the rate per unit mass of gas for Fe production by SNIa relative to SNII. Taking the Galaxy for example, we can obtain the numerical values of $d\\,{\\rm (O/H)}_{\\rm II}/d\\,t$ and $d\\,{\\rm (Fe/H)}_{\\rm II}/d\\,t$ by introducing the time for solar system formation $t_{\\rm SSF}\\approx 10^{10}$~yr (eqs. [\\ref{odot}] and [\\ref{rfe2}]). The absolute yields of O and Fe from each SNII can also be obtained by further introducing the SNII rate per unit mass of gas and the total mass of gas at $t_{\\rm SSF}$. The progenitor stars of SNII have very short lifetimes so that an equilibrium between the birth and death of these stars is established on a correspondingly short timescale. If the rate for formation of these stars is proportional to the mass of gas, then the SNII rate per unit mass of gas can be considered as constant over Galactic history. This is the justification for our earlier assumption of a constant rate per unit mass of gas for Fe production by SNII (eq. [\\ref{rfe2}]). At the present time, the Galactic SNII rate is $\\sim (30\\ {\\rm yr})^{-1}$, corresponding to a total gas mass of $\\sim 10^{10}\\,M_\\odot$. Assuming that all SNII have the same O yield, we find that each SNII must produce $\\sim 0.3\\,M_\\odot$ of O in order to enrich the gas with a solar O mass fraction ($\\approx 9.6\\times 10^{-3}$) at $t_{\\rm SSF}\\approx 10^{10}$~yr (eq. [\\ref{xe}]).  In the model presented here, we have assigned the O and Fe abundances at [Fe/H]~$\\sim -3$ to the prompt production by the first very massive stars formed from Big Bang Debris and have rejected the possibility that these could be produced by SNII. Indeed, although the value of (O/Fe)$_{\\rm P}$ found here is $\\sim 7$ times larger than the average relative production of O to Fe that we associate with SNII, it is still possible to attribute the large (O/Fe) value at [Fe/H]~$\\sim -3$ to contributions from a subset of SNII that produce O but little Fe. The key argument that we have used in rejecting this possibility comes from the results on the production of $r$-process elements and Fe by SNII that were obtained from meteoritic data and observations of metal-poor stars. We give a short summary of these results below as they will be used to place bounds on the O and Fe enrichment to be expected from the early occurrence of SNII in comparison with the prompt inventory (O/H)$_{\\rm P}$ and (Fe/H)$_{\\rm P}$.  Meteoritic data on the inventory of radioactive $^{129}$I and $^{182}$Hf in the early solar system require at least two distinct kinds of SNII as sources for the $r$-process (Wasserburg, Busso, \\& Gallino 1996; Qian, Vogel, \\& Wasserburg 1998; Qian \\& Wasserburg 2000). These are the high-frequency SNIIH events responsible for heavy $r$-process elements with mass numbers $A>130$ (e.g., Ba and Eu; ``H'' for ``high-frequency'' and ``heavy $r$-process elements'') and the low-frequency SNIIL events dominantly producing light $r$-process elements with $A\\leq 130$ (e.g., Ag; ``L'' for ``low-frequency'' and ``light $r$-process elements''). The timescales required for replenishment of the appropriate radioactive nuclei in an average ISM by the SNIIH and SNIIL events are $\\sim 10^7$~yr and $\\sim 10^8$~yr, respectively. These timescales can be explained by considering the mixing of the ejecta from an individual SNII with the ISM. The amount of ISM to mix the SNII ejecta is the mass swept up by a supernova remnant (SNR). The expansion of an SNR can be roughly described by an energy-conserving phase followed by a momentum-conserving one (e.g., Thornton et al. 1998). This expansion results in a total swept-up mass of $\\sim 2E_{\\rm SN}/(v_{\\rm tr}v_{\\rm f}) \\sim 3\\times 10^4\\,M_\\odot$, where $E_{\\rm SN}\\sim 10^{51}$~erg is the supernova explosion energy, $v_{\\rm tr}\\sim 200$~km~s$^{-1}$ is the velocity at the transition between the two phases of expansion, and $v_{\\rm f}\\sim 15$~km~s$^{-1}$ is the final velocity typical of the random motion in the ISM. For Galactic rates of $\\sim (30\\ {\\rm yr})^{-1}$ for H events and $\\sim (300\\ {\\rm yr})^{-1}$ for L events in a total gas mass of $\\sim 10^{10}\\,M_\\odot$, we obtain that the rates of SNIIH and SNIIL events in an ISM of $\\sim 3\\times 10^4\\,M_\\odot$ are $\\sim (10^7\\ {\\rm yr})^{-1}$ and $\\sim (10^8\\ {\\rm yr})^{-1}$, respectively, as required by replenishment of the appropriate radioactive nuclei in this ISM.  Observations of metal-poor stars in the Galactic halo (McWilliam et al. 1995; McWilliam 1998; Sneden et al. 1996, 1998) show that there is wide dispersion in the Ba abundance $\\log\\epsilon({\\rm Ba})\\equiv\\log({\\rm Ba/H})+12$ at [Fe/H]~$\\sim -3$ (region b in Fig. 3). This dispersion in the Ba abundance ($\\sim 2$ dex) at a nearly constant [Fe/H] led Wasserburg \\& Qian (2000a) to conclude that the SNIIH events cannot produce a significant amount of Fe. The Ba abundance becomes extremely low for [Fe/H]~$<-3$ and there is no correlation with Fe in this region (region a in Fig. 3), indicating that there was some Fe production prior to the occurrence of SNII. Observations (Gratton \\& Sneden 1994; see also McWilliam et al. 1995) also show that there is a correlation between the abundances of Ba and Fe at [Fe/H]~$\\gtrsim -2.5$ (region c in Fig. 3).  Estimates of the Ba abundance resulting from a single SNIIH event (Qian \\& Wasserburg 2000; Wasserburg \\& Qian 2000a) are shown as the band labeled ``1 SNIIH'' in Figure 3. The low Ba abundances in stars with [Fe/H]~$\\sim -4$ to $-3$ (region a in Fig. 3) compared with this band and the sharp increase in the Ba abundance at [Fe/H]~$\\sim -3$ led Wasserburg \\& Qian (2000a) to conclude that a source other than SNII must exist to produce Fe (along with other elements such as C, N, O, Mg, and Si) at [Fe/H]~$\\lesssim -3$. They attributed this source to the first very massive stars (with masses of $\\gtrsim 100\\,M_\\odot$) formed from Big Bang debris. They further argued that formation of regular stars (with masses of $\\sim 1$--$60\\,M_\\odot$) could not occur until later when [Fe/H]~$\\sim -3$ was reached. In this later regime, regular stars with masses of $\\gtrsim 10\\,M_\\odot$ explode as SNII. The more frequent occurrence of SNIIH events with no Fe production over a period of $\\lesssim 10^8$~yr gives rise to the sharp increase in the Ba abundance at [Fe/H]~$\\sim -3$. This is followed by the occurrence of low-frequency Fe-producing SNIIL events. Contribution to Fe from SNIa only takes place at [Fe/H]~$\\gtrsim -1$. Enrichment in Fe of metal-poor stars with $-1\\gtrsim$~[Fe/H]~$\\gtrsim -2.5$ must be provided by the SNIIL events. For a Galactic rate of $\\sim (300\\ {\\rm yr})^{-1}$ corresponding to a total gas mass of $\\sim 10^{10}\\,M_\\odot$, each SNIIL must produce $\\sim 0.1\\,M_\\odot$ of Fe in order to enrich the gas with $\\sim 1/3$ of a solar Fe mass fraction ($\\approx 1.2\\times 10^{-3}$) at $t_{\\rm SSF}\\approx 10^{10}$~yr (eq. [\\ref{xe}]). Diluting this amount of Fe in an ISM of $\\sim 3\\times 10^4\\,M_\\odot$ that mixes with the ejecta of a single SNII event results in [Fe/H]~$\\sim -2.5$. Thus the correlation between the abundances of Ba and Fe at [Fe/H]~$\\gtrsim -2.5$ shows that the variation of Fe yields between SNIIH and SNIIL events is smoothed out over a timescale of $\\gtrsim 10^8$~yr.  While the SNIIH events would produce oxygen, they will not be sufficient to provide the large O inventory at [Fe/H]~$\\sim -3$. Diluting the O yield of $\\sim 0.3\\,M_\\odot$ from an SNIIH event into an ISM of $\\sim 3\\times 10^4\\,M_\\odot$ would result in (O/H)~$\\sim 10^{-3}$~(O/H)$_\\odot$ in this mixture. We note that two stars with Ba abundances close to our estimate for a single SNIIH event and with essentially the same values of [Fe/H]~$\\sim -3$ are both observed to have (O/H)~$\\approx 10^{-2}$~(O/H)$_\\odot$, which is a factor of $\\sim 10$ greater than our estimate for a single SNIIH event (HD122563 has $\\log\\epsilon({\\rm Ba})=-1.54$ and [Fe/H]~$=-2.74$, and BD $-18^\\circ$5550 has $\\log\\epsilon({\\rm Ba})=-1.72$ and [Fe/H]~$=-2.91$; Westin et al. 2000; McWilliam 1998; Cavallo, Pilachowski, \\& Rebolo 1997).  We also note that the prompt value (O/Fe)$_{\\rm P}\\sim 20$~(O/Fe)$_\\odot$ at [Fe/H]~$\\sim -3$ assigned by us is based on the observations of BD +23$^\\circ$3130 with [Fe/H]~$=-2.9$ (Israelian et al. 1998), BD +03$^\\circ$740 with [Fe/H]~$=-3$, and BD $-13^\\circ$3442 with [Fe/H]~$=-3.13$ (Boesgaard et al. 1999; Israelian et al. 2000) but there are no relevant Ba data. This value of (O/Fe)$_{\\rm P}$ corresponds to (O/H)$_{\\rm P}\\sim 2\\times 10^{-2}$~(O/H)$_\\odot$, which could be explained by the O contribution from $\\sim 20$ SNIIH events at $\\sim 0.3\\,M_\\odot$ of O per event. If this were the case, then the relevant stars should have high $\\log\\epsilon({\\rm Ba})$ values. Future measurements of Ba abundances in these stars are needed to identify possible SNIIH contribution to the observed O abundances and to determine the prompt O inventory more accurately. However, from the above arguments based on the observations of HD 122563 and BD $-18^\\circ$5550, we conclude that the origin of the O inventory at [Fe/H]~$\\lesssim -3$ must be assigned to the prompt source.  Based on the assignment of (O/Fe)$_{\\rm P}\\sim 20$~(O/Fe)$_\\odot$, the average production ratio of O to Fe for the first very massive stars is $\\sim 560$ by number [(O/Fe)$_\\odot\\approx 28$]. As the amount of O production is much less than the mass of the star, this means that the typical Fe yield from a zero-metallicity star of mass $M$ must be $\\ll 0.6\\,M_\\odot\\times(M/100\\,M_\\odot)$. In considering the prompt enrichment of the ISM by the very massive stars, it is necessary to take several effects into account. The universe at the epoch for formation of these stars was probably dominated by gas. We do not know how many very massive stars contributed to the prompt inventory. Explosions of these stars are likely to be much more energetic than supernova explosions. So the mass of ISM for diluting the ejecta from the explosion of a very massive star is expected to be $\\gg 3\\times 10^4\\,M_\\odot$. With an O yield of $\\sim 0.1\\,M$ for a very massive star of mass $M$ and with the number $N_{\\rm VMS}$ of such stars contributing to the prompt O inventory of (O/H)$_{\\rm P}\\sim 2\\times 10^{-2}$~(O/H)$_\\odot$, this dilution mass can be estimated as $\\sim 5\\times 10^4\\,M_\\odot(M/100\\,M_\\odot)\\times N_{\\rm VMS}$. The expectation that this mass should be $\\gg 3\\times 10^4\\,M_\\odot$ leads us to conclude that the prompt inventory must have been provided by $N_{\\rm VMS}\\gg 1$ very massive stars. That is, many very massive stars are required to provide (O/H)$_{\\rm P}$. Clearly, further studies are needed to understand the formation, evolution, nucleosynthesis, and explosion of these stars.  To determine the relationship between (O/Fe) and (Fe/H) in the Galaxy, Israelian et al. (1998) and Boesgaard et al. (1999) used the OH lines of metal-poor stars with a wide range of (Fe/H). Their results represent the most complete and self-consistent data set available. There is some disagreement between different workers on the evolution of O abundance relative to Fe, in part due to the choice of spectral lines used to determine the O abundance (e.g., Barbuy 1988). While the general model developed here must stand on its own, the numerical value of (O/Fe)$_{\\rm P}$ for the prompt inventory is dependent on the observational results of Israelian et al. (1998) and Boesgaard et al. (1999).  Up to this point, we have assumed that the ISM consists of gas only and that the abundances in an individual star are representative of the average gas at the birth of the star. In fact, the Galactic ISM consists of both gas and dust, with dust making up $\\sim 1\\%$ of the total mass at present. Certain elements such as Fe in the ISM are greatly depleted in the gas phase and are considered to be condensed into dust. So long as stars sample the composition of the bulk ISM including both gas and dust, we can still interpret stellar observations in terms of Galactic chemical evolution as done here because dust is converted into gas in stars. In contrast, when applying the model developed here to damped Ly$\\alpha$ systems, it must be further assumed that Fe and O are not substantially fractionated relative to each other in the protogalactic ISM of such systems. If for example, Fe were dominantly in dust and O in gas in the ISM of damped Ly$\\alpha$ systems (as is the case in many galaxies of about the present epoch), then a straightforward application of the model to such systems would give completely misleading interpretation of the data, which only sample elements in the gas phase.  The most accurate measurement of O abundance in a damped Ly$\\alpha$ system so far was reported by Molaro et al. (2000). They gave [O/H]~$=-1.85\\pm 0.1$ corresponding to [O/Fe]~$=0.19\\pm 0.14$ for a damped Ly$\\alpha$ system with [Fe/H]~$=-2.04\\pm 0.1$ at a redshift $z=3.39$. These data do not fit the trend exhibited by the Galactic data shown in Figure 2d. The data on other elements (e.g., Cr and Zn) in the above damped Ly$\\alpha$ system also suggest that there is no effect of dust depletion (Molaro et al. 2000). We note that the observed value of [O/H] in this system is close to the prompt O inventory assigned here based on the Galactic data while the value of [Fe/H] is $\\sim 10$ times higher than the prompt Fe inventory. We cannot offer a simple explanation for these O and Fe abundances.  Substantial efforts have been made to establish the dust-to-gas ratio for damped Ly$\\alpha$ systems at redshifts $z\\gtrsim 1$. Estimates have been made for the possible dust depletion of condensible elements (e.g., Si, Cr, and Fe) relative to more volatile ones (e.g., Zn) that are considered to remain in the gas phase. Pettini et al. (1997) have argued that there is some significant dust depletion of condensible elements (by a factor of $\\sim 2$) in the damped Ly$\\alpha$ systems at very early epochs. However, there is strong disagreement between various workers (e.g., Lu et al. 1996; Prochaska \\& Wolfe 2000) regarding the degree of dust depletion in these systems. Wasserburg \\& Qian (2000b) assumed that dust depletion is not a dominant effect on the chemical evolution of damped Ly$\\alpha$ systems at $z\\gtrsim 1$. There is clear evidence for large amounts of dust in a variety of galaxies at high $z$. We cannot reconcile the assumption of low dust-to-gas ratios for damped Ly$\\alpha$ systems with these observations. The cause of the transition to the high dust content for galaxies of about the present epoch is not understood unless the later dust contribution of Asymptotic Giant Branch (AGB) stars plays a major role.  It is expected that the first very massive stars formed from Big Bang debris produce C/O ratios of $\\ll 1$ (Heger et al. 2000). The dominance of O over C should cause all the C to be converted into CO, leaving an excess of O in the ISM. This would prevent the formation of C grains and also the formation of polyaromatic hydrocarbon (PAH) as well as other hydrocarbon molecules. A strict upper limit to the dust content of the ISM at [Fe/H]~$\\sim -3$ can be estimated by taking the total abundance of the major condensible elements Mg, Si, and Fe to be $\\sim 10$ times the Fe abundance (see below). As no C dust should be formed, dust would constitute at most $\\sim 10^{-5}$ of the total mass of the ISM at [Fe/H]~$\\sim -3$. For a typical dust size $d$, the obscuration due to this presence of dust is at most at the level of $\\sim 10^{-3} (0.1\\ \\mu{\\rm m}/d)[N({\\rm H\\ I})/10^{21}\\ {\\rm cm}^{-2}]$ in a system of neutral H column density $N({\\rm H\\ I})$.  By attributing the O inventory at [Fe/H]~$\\sim -3$ to the prompt production by the first very massive stars formed from Big Bang debris, the model presented here also predicts that the Mg and Si abundances in stars with (O/Fe)~$\\sim$~(O/Fe)$_{\\rm P}$ and [Fe/H]~$\\sim -3$ should to a large extent reflect the prompt inventory of Mg and Si. Based on the observations of McWilliam et al. (1995), we obtain (Mg/H)$_{\\rm P}\\sim 3\\times 10^{-3}$~(Mg/H)$_\\odot$ and (Si/H)$_{\\rm P}\\sim 10^{-2}$~(Si/H)$_\\odot$, corresponding to (Mg/Fe)~$\\sim 3$~(Mg/Fe)$_\\odot$ and (Si/Fe)~$\\sim 10$~(Si/Fe)$_\\odot$ for stars with [Fe/H]~$\\sim -3$. We also consider that the abundances in the prompt inventory represent the critical condition for transition from formation of very massive stars to regular ones. Taking into account the radiative properties of H, O, Mg, Si, and Fe, we predict that the abundances (O/H)~$\\sim 2\\times 10^{-2}$~(O/H)$_\\odot$, (Mg/H)~$\\sim 3\\times 10^{-3}$~(Mg/H)$_\\odot$, (Si/H)~$\\sim 10^{-2}$~(Si/H)$_\\odot$, and (Fe/H)~$\\sim 10^{-3}$~(Fe/H)$_\\odot$ in the prompt inventory should give sufficient ``metals'' to permit cooling and fragmentation of large condensing gas clouds to form stars with masses of $\\sim 1$--$60\\,M_\\odot$."
    },
    "0010/astro-ph0010172_arXiv.txt": {
        "abstract": "Thermal emission from a rotating, supermassive star will cause the configuration to contract slowly and spin up. If internal viscosity and magnetic fields are sufficiently weak, the contracting star will rotate differentially. For each of the six initial angular momentum distributions considered, a cooling supermassive star will likely encounter the dynamical bar mode instability, which may trigger the growth of nonaxisymmetric bars. This scenario will lead to the generation of long wavelength gravitational waves, which could be detectable by future space-based laser interferometers like LISA. \\bigskip ",
        "introduction": "The existence of supermassive black holes (SMBHs) and the theory of their role as fuel sources for active galactic nuclei and quasars are supported by a growing body of solid observational evidence (see, e.g., the reviews of Rees 1998 and Macchetto 1999,  and references therein).  For example, VLBI observations of the Keplerian motion of masing knots orbiting the nucleus of NGC 4258 indicate the presence of a central object of mass $\\sim 3.6 \\times 10^7 M_{\\sun}$ and radius $< 13$ pc \\citep{miyo95}.  The most reasonable explanation is that this extremely compact object is a black hole.  High resolution HST observations of matter orbiting the nuclei of M87 and NGC 4151 also show nearly Keplerian motion around SMBHs \\citep{macc97, wing99}.  In addition, reverberation mapping and photoionization methods indicate the presence of SMBHs in the centers of 17 Seyfert quasars \\citep{wand99}.  However, the nature of the progenitors of SMBHs is rather uncertain (see Rees 1984, 1999 for overviews).  SMBH formation scenarios that involve the stellar dynamics of dense clusters and the hydrodynamics of supermassive stars have both been proposed.  The suggested stellar dynamical routes focus on the evolution of dense star clusters.  These clusters are formed via a conglomeration of stars, produced by fragmentation of the primordial gas. In one such cluster scenario, massive stars form via stellar collisions and mergers in the cluster and then evolve into stellar-mass black holes.  The merger of these holes, as they grow and settle to the cluster center, leads to the build up of one or more SMBHs (see, e.g., Quinlan and Shapiro 1990, Lee 1995, and references therein).  Alternatively, dense clusters composed of compact stars may become dynamically unstable to relativistic core collapse and thus may give birth to SMBHs \\citep{zepo65,shte85a,shte85b,qush87}.  The proposed hydrodynamical routes to SMBH formation center on the evolution of supermassive stars (SMSs).  SMSs may contract directly out of the primordial gas or they may develop from a supermassive gas cloud that has been built up from the fragments of stellar collisions in clusters (Sanders 1970; Begelman and Rees 1978; Haehnelt, Natarajan, and Rees 1998; however, see also Loeb and Rasio 1994; Abel, Bryan, and Norman 2000; Bromm, Coppi, and Larson 1999). The evolution of SMSs will ultimately lead to relativistic dynamical collapse \\citep{iben63,chan64a,chan64b} and thus possibly to the formation of SMBHs.  In addition to their relevance to AGNs and quasars, SMBHs and their formation events are likely candidates for detection by proposed space-based gravitational wave detectors, like the Laser Interferometer Space Antenna (LISA) \\citep{folk98}.  LISA would be very sensitive to the long wavelength, low frequency gravitational radiation that supermassive objects are expected to emit \\citep{thbr76,thor95}.  For example, LISA may be able to detect the collapse of a SMS to a SMBH or the coalescence of two SMBHs \\citep{lisa95}.  The rates of such events are largely unknown due to the uncertainty in the formation mechanism of SMBHs.  Identifying the scenario by which SMBHs form is hence of fundamental importance to a number of areas of astrophysics. This work examines one of the possible progenitors of SMBHs discussed above, SMSs. The structure and evolution of SMSs have been explored by many authors (see, e.g., Zel'dovich and Novikov 1971 and Shapiro and Teukolsky 1983, and references therein). In this paper, we focus specifically on the quasistatic evolution of SMSs as they contract and cool due to thermal emission.  Two related factors that affect the cooling phase of a SMS are the strengths of its viscosity and magnetic fields and the nature of its angular momentum distribution.  If strong enough, viscosity will enforce uniform rotation throughout the star as it contracts. However, the strength of molecular viscosity is unlikely to be large enough to accomplish this, so turbulent viscosity or magnetic fields must provide the necessary dissipation \\citep{zeno71, spru99a, spru99b, bash99}. However, it is not known whether these agents are sufficiently strong to maintain uniform rotation in SMSs as they evolve.  If a SMS is rotating prior to its contraction phase, conservation of angular momentum requires the star to spin up and therefore increase its ratio of kinetic energy $T$ to gravitational potential energy $W$, $\\beta=T/|W|$.  If viscosity or magnetic fields are able to maintain uniform rotation, the star will be spun up to its mass--shedding limit $\\beta_{shed} = 8.99 \\times 10^{-3}$. The value is small because a SMS is dominated by radiation pressure and thus is well represented by an $n$=3 polytropic equation of state (see \\S 2.1.1).  As a result, it is quite centrally and cannot support much rotation when spinning uniformly before mass at the equator is driven into Keplerian orbit. Thereafter, it will evolve along a mass--shedding sequence, losing both mass and angular momentum, and will contract to the point of relativistic instability and collapse. \\citet{bash99} have recently investigated this evolutionary scenario in detail and have described how it may lead to SMBH formation. Specifically, they identified the critical configuration at the onset of the relativistic radial instability.  They also demonstrated that the nondimensional (in units where $G$=$c$=1) ratios $T/|W|$; $R/M$, where $R$ is the radius and $M$ is the mass; and $J/M^2$, where $J$ is the angular momentum, are universal numbers (i.e., they do not depend on $M$, $R$, or $J$ for the crititcal configuration or on the history of the star).  This work investigates the extreme opposite regime of SMS cooling evolution, the low-viscosity  and low-magnetic field limits. In this case, as the star contracts and spins up, neither viscosity nor magnetic fields will be able to enforce uniform rotation (even if the star initially rotates uniformly).  The outcome of the evolution will then depend on the star's initial angular momentum distribution (or rotation law), as the star will conserve its angular momentum on cylindrical shells as it contracts. One possible outcome is that the star will again be spun-up to the mass-shedding limit (at a value of $\\beta_{shed}$ higher than that for uniform rotation). The subsequent evolution is not clear. It is likely that mass and angular momentum will be ejected from the equator as the star undergoes further cooling. It is possible that it will continue to evolve in quasistationary equilibrium until reaching the onset of dynamical collapse. If it implodes to a SMBH, the matter is likely to emit an appreciable burst of gravitational radiation just prior to entering the event horizon.   An alternative outcome of low-viscosity, low-magnetic field contraction of a differentially rotating SMS is that it will encounter the {\\it dynamical} ``bar mode'' instability at a critical value $\\beta_{bar}$ prior to reaching $\\beta_{shed}$. Such global rotational instabilities in fluids arise from nonaxisymmetric modes, which vary like $e^{\\pm im\\phi}$, where $m$=2  is the bar mode.  As the bar mode develops, angular momentum and mass will be transported outwards \\citep{tdm85,pick96,idp00,nct00,brow00}.  If enough angular momentum is removed, collapse to a SMBH could ensue. A SMS undergoing a bar mode distortion would be a strong source of quasi-periodic gravitational radiation because of its highly nonaxisymmetric structure. We note that a bar might form during catastrophic collapse in any scenario, if $\\beta (\\propto 1/R)$ grows above the critical value for the dynamical bar mode instability.  The outcome of the cooling and the resultant spin-up of a low-viscosity SMS depends critically on its initial angular momentum distribution.  However, the nature of the angular momentum distribution of SMSs is largely unknown. The purpose of this paper is to examine the evolution of SMS models by assuming several different initial rotation laws to determine, for example, if any of them might cause the star to encounter the dynamical bar mode, prior to mass-shedding and collapse (i.e., to determine if $\\beta_{bar}<\\beta_{shed}$).  To make this determination, we analyze several Newtonian equilibrium sequences of SMS models.  The models along each sequence are constrained to have the identical mass $M$, angular momentum $J$, and angular momentum distribution but decreasing entropy, due to cooling. Such a sequence mimicks the quasistatic evolutionary track of a SMS as it slowly cools and contracts. We find that along the sequence, the stars flatten as measured by the ratio of polar radius $R_p$ to equatorial radius $R_e$. We examine each sequence to determine whether the limits  $\\beta_{shed}$ and/or $\\beta_{bar}$ are reached.  Note that 3D hydrodynamical simulations are necessary to determine precisely $\\beta_{bar}$ for each sequence. However, previous linear and nonlinear stability analyses indicate that $0.14\\lesssim\\beta_{bar} \\lesssim 0.27 $ \\citep{chan69,shte83,duto85,imam85,mana85, hach88,toha90,pick96,toma98,idp00}. See \\S 4 for further discussion of these stability limits.  We note that {\\it secular} version of the bar instability, which sets in at a lower value of $\\beta$, is not relevant for the scenario described here.  The reason for this is that the viscosity that could drive the mode is assumed to be vanishingly small and, hence, the secular timescale is too long \\citep{bash99}. The timescale for gravitational radiation to drive the secular bar mode is also much longer than the evolution (cooling) timescale \\citep{chan70}.  In the low-viscosity and low-magnetic field limits, the angular momentum of mass shells on concentric cylinders in a contracting star will be preserved \\citep{boos73,tass78}. Thus we confine our discussion to equilibrium sequences constructed with rotation laws that conserve the angular momentum distribution in this manner. The sequences examined here include one we construct below with the $n'$=3 rotation law (see \\S 2.2). This sequence represents the cooling evolution of a SMS that is nearly spherical and rotating uniformly at formation prior to its contraction phase. It is the low-viscosity, low-field  analog of the uniformly rotating sequence considered in Baumgarte \\& Shapiro (1999); in the present case, differential rotation will ensue once cooling gets underway. Because the initial rotation profiles of SMSs are unknown, we also examine alternative sequences constructed by Hachisu, Tohline, and Eriguchi (1988; hereafter HTE) that describe the evolution of SMSs with nonuniform initial rotation laws.  The remainder of this paper is organized as follows.  In \\S 2, we describe the numerical methods used to construct our $n'$=3 equilibrium sequence and the sequences constructed by HTE. We present our $n'$=3 sequence and summarize the sequences constructed by HTE in \\S 3.  In \\S 4, we outline possible evolutionary scenarios for SMSs, based on the sequences described in \\S 3.  We discuss these results in \\S 5. ",
        "conclusions": "The thermal emission of a rotating supermassive star causes it to contract and spin up.  If the viscosity and internal magnetic fields are weak, the SMS will rotate differentially during its cooling and contraction phase. We have examined models of the quasistatic cooling evolution of SMSs, represented by equilibrium sequences with six differential rotation laws.  The evolution of a cooling SMS that is initially nearly spherical, with very low, uniform rotation can be represented by the $n'$=3 sequence we constructed. This sequence terminates due to mass--shedding at $\\beta_{shed}\\gtrsim 0.30$. This $\\beta_{shed}$ exceeds the value $0.27$, which is a likely upper limit for the onset of the rotationally induced, dynamical bar instability.  Thus it appears likely that a star with this rotation law would become unstable to the bar mode prior to reaching the mass shedding  limit.  We have also considered sequences constructed by HTE, in order to follow the evolution of SMSs with five different initial differential rotation laws.  No mass--shedding limits exist on these sequences.  Thus, a cooling SMS characterized by one of these rotation profiles will also encounter the dynamical bar mode.  This mode will transport mass and angular momentum outward and thus may hasten the onset of collapse and possible SMBH formation \\citep{tdm85,pick96,idp00,nct00,brow00}. Because of their nonaxisymmetric structure, SMSs undergoing the bar mode could be strong sources of quasi-periodic gravitational radiation for detectors like LISA, even before collapse.  The frequency of these quasiperiodic gravitational waves can be estimated from the expected bar rotation rate $\\Omega_{bar}$. The model on our $n'$=$3$ sequence with $\\beta\\sim\\beta_{bar}$=$0.27$ has a central rotation rate $\\Omega_{c}$=$1.02\\times 10^{-1} {\\rm Hz}\\, [M_6 \\beta_{-5}^{-3} R_{17}^{-3}]^{1/2}$.  In previous hydrodynamics simulations of the bar mode instability \\citep{nct00}, $\\Omega_{bar}$ was $\\sim 0.4 \\Omega_{c}$.  With this relation between $\\Omega_{bar}$ and $\\Omega_{c}$, the gravitational wave frequency $f_{GW}$ can be estimated to be \\begin{eqnarray} f_{GW} &=& 2 f_{bar} = 2 \\frac{\\Omega_{bar}}{2\\pi} \\sim \\frac{0.4}{\\pi}\\Omega_{c} \\nonumber \\\\ &\\sim& 1\\times 10^{-2} {\\rm Hz}\\, [M_6 \\beta_{-5}^{-3} R_{17}^{-3}]^{\\frac{1}{2}}. \\end{eqnarray} For a SMS of $10^6 M_{\\sun}$, which begins as a slowly rotating star of radius $10^{17} {\\rm cm}$ with $\\beta_0$=$10^{-5}$, this yields a frequency of $1\\times 10^{-2} {\\rm Hz}$. This frequency is in the range in which LISA is expected to be most sensitive, $10^{-4}$-$1 {\\rm Hz}$ (see, e.g., Thorne 1995, Folkner 1998). The choice $\\beta_0$=$2\\times 10^{-4}/R_{17}$ is the maximum value of $\\beta_0$ for which $f_{GW}$ (=$10^{-4} {\\rm Hz}$) is still in LISA's range of sensitivity.\\footnote{We note that \\citet{peeb69} has discussed the spin-up of primordial masses by tidal torquing and finds that $J \\sim M R^2$. Naively scaling his results (derived for $M \\sim 10^{11} M_{\\sun}$ configurations) to our regime, we find that $\\beta_0 \\sim 10^{-4}$, so that the resulting frequency of gravitational radiation would lie near the lower limit of LISA's range.}  The strength of the gravitational wave signal can be estimated roughly to be \\begin{eqnarray} h &\\sim& \\frac{G}{c^4} \\frac{\\ddot{Q}}{d} \\sim \\frac{G}{c^4} \\frac{M R_{bar}^2 f_{bar}^2}{d} \\nonumber \\\\ &\\sim& 4 \\times 10^{-15} \\biggl(\\frac{d}{1\\, {\\rm Gpc}}\\biggr)^{-1} M_6^2 \\beta_{-5}^{-1} R_{17}^{-1}, \\end{eqnarray} where $\\ddot{Q}$ is the second time derivative of the star's quadrupole moment and $d$ is the distance, which we scale to 1 Gpc (the Hubble distance is $\\sim$ 3 Gpc).  Here we have used $R_{bar} \\sim R_e$=$4.14\\times 10^{13} \\beta_{-5} R_{17}\\,{\\rm{cm}}$ for the $n'$=$3$ model that reaches the point $\\beta\\sim\\beta_{bar}$.  As mentioned previously, recent simulations of the bar instability indicate that the outcome is a long-lived bar-like structure \\citep {nct00, brow00, saij00}.  The bar will decay on a secular timescale due to dissipative effects.  For differentially rotating SMSs, the largest source of dissipation will be gravitational radiation. The gravitational radiation damping timescale $\\tau_{GW}$ is approximately \\begin{equation} \\tau_{GW} \\sim \\frac{T}{\\bigl(\\frac{dE}{dt}\\bigr)_{GW}}, \\end{equation} where $T$ is the rotational kinetic energy and $(dE/dt)_{GW}$ is the rate at which gravitational radiation carries energy away from the system.  Recall that $T=\\beta |W|$. The gravitational potential energy $|W| \\sim GM^2/R_{bar}$.  Thus, \\begin{equation} T=\\beta \\frac{G M^2}{R}. \\end{equation} The radiation rate can be estimated as  \\citep{shte83} \\begin{equation} \\bigl(\\frac{dE}{dt}\\bigr)_{GW} \\sim \\frac{G}{c^5} \\frac{M}{R^2} v^6. \\end{equation} In this case the characteristic velocity of the system is $v=(\\beta G M/R)^{1/2}.$  Substitution of equations (18) and (19) into equation (17) yields \\begin{eqnarray} \\tau_{GW} &\\sim& \\frac{c^5}{G^3} \\frac{R_{bar}^{4}}{\\beta^2 M^3} \\nonumber \\\\ &\\sim& 1 \\times 10^4 yrs  M_6^{-3} [\\beta_{-5} R_{17}]^4. \\end{eqnarray}  The number of cycles $\\cal{N}$ for which the signal will persist is \\begin{eqnarray} \\cal{N}&=&\\tau_{GW} f_{GW} \\nonumber \\\\ &=& 4 \\times 10^9 [M_6 \\beta_{-5} R_{17}]^{-5/2}. \\end{eqnarray} The quasiperiodicity of such a signal will assist in its detection \\citep{schutz97}.  The fraction of the mass $(dM/M)_{GW}$ radiated via gravitational radiation over the interval $\\tau_{GW}$ can be estimated as \\begin{eqnarray} \\biggl(\\frac{dM}{M}\\biggr)_{GW} &\\sim& \\tau_{GW} \\bigl(\\frac{dE}{dt}\\bigr)_{GW} \\frac{1}{Mc^2} \\nonumber \\\\ &\\sim& \\frac{G}{c^2} \\frac{\\beta M}{R} \\nonumber \\\\ &\\sim& 1 \\times 10^{-3} M_6 [\\beta_{-5} R_{17}]^{-1}. \\end{eqnarray}  It appears that SMSs with a wide range of differential rotation laws may become unstable to the development of the dynamical bar mode. However, linear and nonlinear stability analyses are needed to precisely determine the onset of the $m$=2 dynamical instability for the equilibrium sequences considered here.  Such analyses are also needed to determine the relative importance of various unstable nonaxisymmetric modes in SMS models (as a dominant $m$=1 mode would change the characteristics of the gravitational radiation emission). In addition, three-dimensional hydrodynamical simulations are necessary to study the mass and angular momentum redistribution induced by these nonaxisymmetric instabilities, to compute the gravitational waveforms emitted, and to determine the final fate of the star.  Future investigations involving hydrodynamical simulations of SMS models would benefit from improved knowledge of initial conditions, such as an appropriate value for $\\beta_0$.  These appropriate initial conditions could be determined from studies of large-scale structure and cosmology.  Most importantly, studies are required to assess the roles of viscosity and magnetic fields in rotating SMSs to judge whether they are sufficient to drive these configurations to uniform rotation prior to bar instability."
    },
    "0010/astro-ph0010491_arXiv.txt": {
        "abstract": "Recent analyses of the helioseismic observations indicate that the previously observed surface torsional oscillations with periods of about 11 years extend significantly  downwards into the solar convective zone. Furthermore, there are indications that the dynamical regimes at the base of the convection zone are different from those observed at the top, having either significantly shorter periods or non--periodic behaviour.  We propose that this behaviour can be explained by the occurrence of {\\it spatiotemporal fragmentation}, a crucial feature of which is that such behaviour can be explained solely through nonlinear spatiotemporal dynamics, without requiring separate mechanisms with different time scales at different depths.  We find evidence for this mechanism in the context of a two dimensional axisymmetric mean field dynamo model operating in a spherical shell, with a semi--open outer boundary condition, in which the only nonlinearity  is the action of the azimuthal component of the Lorentz force of the dynamo generated magnetic field on the solar angular velocity. ",
        "introduction": "Recent analyses of the helioseismic data, from both the Michelson Doppler Imager (MDI) instrument on board the SOHO spacecraft (Howe et al.\\ 2000a) and the Global Oscillation Network Group (GONG) project (Antia \\& Basu 2000) have provided strong evidence which indicates that the earlier observed time variations of the differential rotation on the solar surface, the so called `torsional oscillations' with periods of about 11 years (e.g.\\ Howard \\& LaBonte 1980; Snodgrass, Howard \\& Webster 1985; Kosovichev \\& Schou 1997; Schou et al.\\ 1998), penetrate into the convection zone, to depths of at least 8 percent in radius.  Furthermore, these data have provided some evidence to suggest that variations in the differential rotation are also present around the tachocline at the bottom of the convection zone (Howe et al.\\ 2000b). An important feature that distinguishes these variations from those observed at the upper parts of the convection zone is that they possess markedly different modes of behaviour: either possessing distinctly lower periods (of $\\sim$$1.3$ years), or being non--periodic (Antia \\& Basu 2000). Clearly, to firmly establish the precise nature of these variations, future observations are required. Whatever the outcome of such observations, however, both these sets of results point to the very interesting possibility that the variations in the differential rotation can have different periodicities/behaviours at different depths in the solar convection zone. It is therefore important to ask whether such different variations can in principle occur at different parts of the convection zone and, if so, what could be the possible mechanism(s) for their production.  The aim of this letter is to suggest that a natural mechanism for the production of such different dynamical modes of behaviour in the convection zone is through what we call {\\it spatiotemporal fragmentation}, i.e.\\ the occurrence of dynamical regimes at (given) values of the control parameters of the system, which possess different temporal behaviours at different spatial locations.  This is to be contrasted with the usual temporal bifurcations, with identical temporal behaviour at each spatial point, which require changes in parameters to occur.  We find evidence for this mechanism in the context of a two dimensional axisymmetric mean field dynamo model in a spherical shell, with a semi--open outer boundary condition, in which the only nonlinearity  is the action of the azimuthal component of the Lorentz force of the dynamo generated magnetic field on the solar angular velocity. The underlying angular velocity is chosen to be consistent with the most recent helioseismic data.  In addition to producing different dynamical variations in the differential rotation, including different periods, at the top and the bottom of the convection zone, this model is also capable of producing butterfly diagrams which are in qualitative agreement with the observations as well as displaying torsional oscillations that penetrate into the convection zone, as recently  observed by Howe et al.\\ 2000a and Antia \\& Basu 2000 and studied by Covas et al.\\ (2000). ",
        "conclusions": "We have proposed spatiotemporal fragmentation as a natural mechanism for producing different types of variation in the differential rotation at the top and the bottom of the convection zone. To demonstrate the occurrence of such behaviour, we have studied a solar dynamo model, with a semi--open outer boundary condition, calibrated to have the correct cycle period, with a mean rotation law given by recent helioseismic observations. We note in passing that in a few simulations performed with the boundary condition $B=0$ at the surface, we have not so far found this phenomenon, although we cannot yet make a definitive statement on this point. In addition to producing butterfly diagrams in qualitative agreement with those that are observed, as well as torsional oscillations that penetrate into the convection zone, we have shown that this model can also produce spatiotemporal fragmentation, resulting in different oscillatory modes of behaviour near the top and the bottom of the convection zone.  We emphasize that the main aim of this letter is to propose a mechanism that can be expected to operate in general nonlinear dynamo settings, and which is capable of producing multiple periods and/or non-periodic oscillations in parts of the convective zone. The specific results given here, such as the single period halving, are based on a particular dynamo model which inevitably includes many simplifying assumptions, not least of which is that the density is uniform. (It is unclear how the inclusion of a radial dependence $\\rho(r)$ would affect our results --- we note that current solar dynamo models commonly take a uniform density.) We expect that the mechanism is of quite general applicability, and so it is plausible that a more sophisticated model might exhibit further bifurcations, thus producing different reduced periods and oscillatory regimes. It may also be useful to bear in mind in this connection that three period halvings would result in $11$ years$/2^3 \\sim 1.3$ years! We shall return to a more detailed study of the underlying dynamics as well as a quantitative study of different dynamo models elsewhere. We have chosen $P_r=1$ in order to obtain larger amplitude torsional oscillations near the surface. We have checked that fragmentation still occurs at smaller values of $P_r$.  Inevitably the uncertainties associated with the inversion of the helioseismic data so deep in the convection zone are quite large. Thus we believe that the mechanism discussed here may, by demonstrating what modes of dynamical behaviour are theoretically possible, act as a conceptual aid in interpreting current and further observations."
    },
    "0010/astro-ph0010344_arXiv.txt": {
        "abstract": "The effects of noise and sampling on the ``Spectral Correlation Function'' (SCF) introduced by Rosolowsky et al. 1999 are studied using observational data, numerical simulations of magneto--hydrodynamic turbulence, and simple models of Gaussian spectral line profiles. The most significant innovations of this paper are: i) the normalization of the SCF based on an analytic model for the effect of noise; ii) the computation of the SCF as a function of the spatial lag between spectra within a map.  A new definition of the ``quality'' of a spectrum, $Q$, is introduced, which is correlated with the usual definition of signal--to--noise. The pre--normalization value of the SCF is a function of $Q$. We derive analytically the effect of noise on the SCF, and then {\\it normalize} the SCF to its analytic approximation.  By computing of the dependence of the SCF on the spatial lag, $S_0(\\dr)$, we have been able to conclude that:  \\begin{itemize} \\item $S_0(\\dr)$ is a power law, with slope $\\alpha$, in the range of scales $\\ell_i<\\ell<\\ell_o$. \\item The correlation outer scale, $\\ell_o$, is determined by the size of the map, and no evidence for a true departure from self--similarity on large scales has been found. \\item The correlation inner scale, $\\ell_i$, is a true estimate of the smallest self--similar scale in a map. \\item The spectral slope, $\\alpha$, in a given region, is independent of velocity resolution (above a minimum resolution threshold), spatial resolution, and average spectrum quality. \\item Molecular transitions which trace higher gas density yield larger values of $\\alpha$ (steeper slopes) than transitions tracing lower gas density. \\item Nyquist sampling, bad pixels in detector arrays, and reference sharing data acquisition need to be taken into account for a correct determination of the SCF at $\\dr =1$. The value of $\\alpha$, however, can be computed correctly without a detailed knowledge of observational procedures. \\end{itemize} ",
        "introduction": "Rosolowsky et al. (1999) (RGWW) have recently introduced a new method to analyze large maps of molecular spectral lines. They propose to use the ``Spectral Correlation Function'' (SCF) as a way to test theoretical models against observational data. The SCF at a given position in a map is defined as the quadratic sum of the difference between the spectral line profile at that position and the profile at neighboring positions: \\begin{equation} S_0(T_1,T_0)=1- \\sqrt{\\frac{\\Sigma_v(T_1(v)-T_0(v))^2}{\\Sigma_v T_1(v)^2+\\Sigma_v T_0(v)^2}} \\label{for1} \\end{equation} where $T_i(v)$ is the antenna temperature at the velocity channel $v$ at the position $i=0,1$ in the map. The definition of the SCF in RGWW is more general than this, since it allows for translation along the velocity axes and rescaling of antenna temperature that minimize the difference between neighboring spectra. In the present work we only discuss the SCF as defined in equation (\\ref{for1}), corresponding to ``$S^0$'' in RGWW.  The SCF is similar in spirit to some of the analysis tools used to extract clumps from 3-D spectral line data cubes (Stutzki \\& Gusten 1990; Williams, De~Geus \\& Blitz 1994), in that it makes direct use (no transform involved) of both spatial and velocity information. However, the SCF is different from these methods, because it makes no attempt to extract the properties of clumps from data cubes; it simply compares neighboring spectra with each other, utilizing both the spatial and velocity dimensions.  Some previous statistical analyses do not explicitly use the velocity dimension in analyzing spectral line cubes (for example the wavelet analysis by Gill \\& Henriksen 1990 and Langer, Wilson, \\& Anderson 1993; the structure-tree statistics by Houlahan \\& Scalo 1992; the column density distribution by Blitz \\& Williams 1997; the $\\Delta$--variance method by Stutzki et al. 1998 and Mac Low \\& Ossenkopf 1999; the fractal analysis by Beech 1987, Bazell \\& D\\'{e}sert 1988, Scalo 1990, Dickman, Horvath \\& Margulis 1990, Falgarone, Phillips \\& Walker 1991, Zimmermann, Stutzki \\& Winnewisser 1992, Henriksen 1991, Hetem \\& Lepine 1993, Vogelaar \\& Wakker 1994, Elmegreen \\& Falgarone 1996).  Other statistical analysis make use of the velocity dimension, but only to estimate the centroid velocity of the spectra, and compute the structure (or autocorrelation) function of velocity fluctuations (Scalo 1984; Kleiner \\& Dickman 1985, 1987; Hobson 1992; Miesch \\& Bally 1994), or the distribution of line centroids (Miesch \\& Scalo 1995; Lis et al. 1996; Miesch, Scalo \\& Bally 1999). The moment analysis computes statistical moments (velocity centroids, line width, skewness and kurtosis) of single spectra in a map, and derive their distribution over the whole map (Falgarone et al. 1994; Padoan et al. 1999), independent of their position in a map.  A method that exploits both velocity and spatial information is the Principal Components Analysis by Heyer \\& Schloerb (1997). This method describes clouds as a sum of orthogonal functions in a manner mathematically similar to wavelet analysis, and is the most promising way to extract the power spectrum of molecular cloud turbulence from observational data.  Therefore, prior to the SCF, no statistical analysis had tackled the problem of quantifying spatial correlations in spectral maps, taking into account also the velocity information, with the exception of the Principal Component Analysis. The SCF quantifies the correlation between spectra at a given distance from each other (spatial lag), using the full velocity profile information, because the comparison between spectra is made channel to channel.  \\nocite{Gill+Henriksen90} \\nocite{Langer+93} \\nocite{Houlahan+Scalo92} \\nocite{Blitz+Williams97} \\nocite{Stutzki+98} \\nocite{MacLow+Ossenkopf99} \\nocite{Beech87}  \\nocite{Bazell+Desert88}  \\nocite{Scalo90} \\nocite{Dickman+90}  \\nocite{Falgarone+91} \\nocite{Zimmermann+92} \\nocite{Henriksen91} \\nocite{Hetem+Lepine93} \\nocite{Vogelaar+Wakker94} \\nocite{Elmegreen+Falgarone96} \\nocite{Scalo84} \\nocite{Kleiner+Dickman85} \\nocite{Kleiner+Dickman87} \\nocite{Hobson92} \\nocite{Miesch+Bally94} \\nocite{Heyer+Schloerb97} \\nocite{Falgarone+94} \\nocite{Padoan+98per}   RGWW concluded that the SCF can find differences between observational data and synthetic spectra from numerical simulations of turbulent flows that are not found by other statistical analyses of line profiles (e.g. the moment analysis by Falgarone et al. 1994). That preliminary result motivates the present work, where we try to study in more detail the effect of noise and sampling on the SCF, and to improve on the first implementation of the method. The main issue is that the value of the SCF as defined in RGWW depends on the signal to noise ($S/N$), as illustrated in Figure~1 of both that paper and Figure~\\ref{fig00} of this paper. A simple way to eliminate this signal-to-noise dependence, proposed in RGWW, is to make the $S/N$ uniform over the map, by adding noise. In RGWW, spectra with $S/N<5$ were discarded, and noise was added to higher quality spectra to force $S/N=5$. In the present work, instead, we have analytically estimated the main effect of noise on the SCF, using a novel definition of the $S/N$. We can therefore compute a noise--corrected SCF that is hardly dependent on the $S/N$, without adding any extra noise to the data.  In the new-and-improved implementation of the SCF offered in this paper, when observational data are compared with synthetic maps, noise must be added to the synthetic data, under the assumption that it is uniform over the observed map. If noise is not uniform over the observed map, or if it is correlated over a few map positions, extra noise must be added to the observational spectra until the noise is both spatially uniform and uncorrelated. However, we do not need to adjust the $S/N$ to be uniform as in RGWW.  In the following two sections, the effect of noise on the SCF is discussed, and an analytic model is computed that allows the values of the SCF to be corrected for the effect of noise. In \\S 4 we study the effects of both velocity sampling and spatial lags between spectra. We show that a correlation inner scale, $l_i$, can be defined. On scales smaller than $l_i$ the spectral map is not self-similar. Results are tested in \\S 5, by computing the SCF using two maps of the Rosette Molecular Cloud with different resolution. A discussion is presented in \\S 6, and in \\S 7 we summarize our conclusions. ",
        "conclusions": "In this work we have studied the simplest form of the SCF proposed by RGWW ($S^0$ in RGWW), and we have discussed its dependence on noise, velocity sampling, and spatial sampling, using observational data, numerical simulations of magneto--hydrodynamic turbulence, and simple models of Gaussian spectral line profiles. We have computed analytically the effect of noise on the SCF, and have proposed to renormalize the SCF by its analytical maximum. As a result, the SCF now has only a weak dependence on signal--to--noise, because most of the effect of noise has been corrected for with the analytic formula. A new definition of ``spectrum quality'' has been used for this purpose, which we call $Q$. In the computation of the SCF we have never used Gaussian fits to the spectra, and we have used a velocity window of constant width, equal to 10 times the value of the standard deviation of the spectral profile averaged over the whole map. We have also computed the SCF as a function of the spatial lag between spectra, $S_0(\\dr)$, which has allowed us to describe spectral maps in terms of their spectral slope $\\alpha$, and correlation inner scale $\\ell_i$.   The main conclusions of this work are the following:  \\begin{itemize} \\item $S_0(\\dr)$ is a power law in the range of scales $\\ell_i<\\ell<\\ell_o$. \\item The correlation outer scale, $\\ell_o$, is determined by the size of the map; we have found no evidence for a true departure from self--similarity on large scales, in the map analyzed so far. \\item The correlation inner scale, $\\ell_i$, is a true estimate of the smallest self--similar scale in a map. \\item The spectral slope, $\\alpha$, is roughly independent of velocity resolution, spatial resolution, and average spectrum quality, $\\aq$; it is a robust statistical property of spectral maps of molecular clouds. \\item The correlation scales, $\\ell_i$ and $\\ell_o$, are also roughly independent of velocity and spatial resolutions, and $\\aq$, if $S_0(\\dr)$ is properly rescaled to physical units of length. \\item Molecular transitions which trace higher gas density yield steeper $S_0(\\dr)$ power--laws than transitions tracing lower gas density. \\item Nyquist sampling, bad pixels in detector arrays, and reference sharing data acquisition need to be taken into account for a correct determination of the SCF at $\\dr =1$. The value of $\\alpha$, however, can be computed correctly without a detailed knowledge of observational procedures. \\end{itemize}   We expect that the exact value of the spectral slope and of the correlation inner scale depend on the physical conditions in the clouds, such as the turbulent velocity dispersion relative the the speed of sound, the magnetic field strength, the effects of gravity, stellar outflows, and other influences. In our future work, the SCF will be applied to different molecular cloud maps and to different models, in order to study the sensitivity of the SCF method to physical conditions in the clouds."
    },
    "0010/astro-ph0010034_arXiv.txt": {
        "abstract": "We present a new detection of a Massive Dark Object in the S0 Galaxy NGC 4350, obtained applying a new dynamical model on ground-based photometric and kinematic data already present in literature. ",
        "introduction": " ",
        "conclusions": ""
    },
    "0010/astro-ph0010202_arXiv.txt": {
        "abstract": "We present analysis of HST Planetary Camera images of twenty L dwarfs identified in the course of the Two Micron All-Sky Survey. Four of the targets, 2MASSW J0746425+200032, 2MASSs J0850359+105716, 2MASSW J0920122+351742 and 2MASSW J1146345+223053, have faint, red companions at separations between 0.07 and 0.29 arcseconds (1.6 to 7.6 AU). Ground-based infrared imaging confirms the last as a common proper-motion companion. The surface density of background sources with comparable colours is extremely low, and we identify all four as physical binaries. In three cases, the bolometric magnitudes of the components differ by less than 0.3 magnitudes. Since the cooling rate for brown dwarfs is a strong function of mass, similarity in luminosities implies comparable masses. The faint component in the 2M0850 system, however, is over 1.3 magnitudes fainter than the primary in the I-band, and $\\sim0.8$ magnitudes fainter in M$_{bol}$. Indeed, 2M0850B is $\\sim0.8$ magnitudes fainter in I than the lowest luminosity L dwarf currently known, while the absolute magnitude we deduce at J is almost identical with M$_J$ for Gl 229B. We discuss the implications of these results for the temperature scale in the L/T transition region. 2M0850 is known to exhibit $\\lambda 6708$\\AA\\ Li I absorption, indicating that the primary has a mass less than 0.06M$_\\odot$. Theoretical models predict that the magnitude difference implies a mass ratio of $\\approx$0.75.  The apparent binary fraction of the current sample, 20\\%, is comparable with the results of previous surveys of late-type M dwarfs in the field and in the Hyades cluster. However, the mean separation of the L dwarf binaries in the current sample is smaller by a factor of two than the M dwarf value, and only one system would be detected at the distance of the Hyades. We discuss the likely binary frequency amongst L dwarfs in light of these new data. ",
        "introduction": " ",
        "conclusions": ""
    },
    "0010/astro-ph0010564_arXiv.txt": {
        "abstract": "Statistical arguments show that the volume- and time-averaged kinetic power of GRBs and fireball transients (FTs) into an L$^*$ galaxy like the Milky Way is at the level of $10^{40}$ ergs s$^{-1}$. This number, though with wide uncertainties related to the internal or external shock efficiency, is sufficient to power hadronic cosmic rays observed locally.  The release of energy by the high-mass progenitor stars of GRBs and FTs is sufficient to power the ultrahigh energy cosmic rays, as already shown by Waxman and Vietri in 1995.  It is sufficient to power the cosmic rays above the knee of the cosmic-ray spectrum. Indeed, all hadronic cosmic rays could originate from the high-mass ($\\gtrsim 100$ M$_\\odot$) stars that collapse to black holes, in the process forming GRBs and FTs. This source class represents a new solution to the problem of cosmic-ray origin.  The $\\sim 10^4$-$10^7$ black holes made by these stars could make their presence known by radiating as they accrete from the ISM, by microlensing background radiations, and by forming luminous binary systems. Some unidentified EGRET sources could be isolated black holes that accrete from the ISM. Better imaging and sensitivity with GLAST and TeV observatories will test this model for the unidentified $\\gamma$-ray sources, and this theory for cosmic-ray origin. ",
        "introduction": "Fireball transients are those explosive events that propel a significant fraction of their ejecta kinetic energy in the form of relativistic baryonic outflows. FTs include dirty ($\\Gamma_0 \\lesssim 300$), loaded ($\\Gamma_0\\sim 300$), and clean fireball ($\\Gamma_0 \\gtrsim 300$) subclasses \\cite{dcb99}, of which BATSE is most sensitive to those explosions with initial bulk Lorentz factors $\\Gamma_0 \\sim 300$. The dirty and clean fireball classes of exploding stars are difficult to discover because of design and sensitivity limitations of detectors flown to date\\cite{dcb99,dbc99}, though the X-ray flashes detected with Beppo-SAX \\cite{hei00} could be representatives of the dirty fireball class. HETE-II, Swift, and GLAST can all be expected to make progress on this front, though a lobster-eye X-ray telescope seems most promising for discovering dirty fireballs.  In a recent paper \\cite{der00}, I deduced the time- and volume-averaged kinetic power of GRBs and fireball transients (FTs) in the Milky Way on the basis of a statistical treatment employing the external shock model of GRBs \\cite{rm92,mr93} for both the prompt $\\gamma$-ray luminous and afterglow phase, performed by B\\\"ottcher and myself \\cite{bd00}. There I found that the FT power into the galaxy was within an order of magnitude of the inferred average Galactic cosmic-ray power, and I considered the possibility that the high-mass stellar progenitors of FTs were the sources of the cosmic rays (CRs) above and below the knee of the CR spectrum. I also discussed observational difficulties in the scenario advanced by Ginzburg and Syrovatskii \\cite{gs64} that hadronic CRs are powered when supernovae collapse to neutron stars.  Although $\\gamma$-ray astronomy was predicted to solve the CR origin problem (see \\cite{wee99} for a recent assessment), evidence for CR source origin still rests primarily upon arguments about available sources of power, which in turn hinge upon statistical analyses. The numerical simulation of GRB statistics \\cite{bd00} that was used to derive the FT power in the Milky Way modeled detector response and generic GRB spectral and temporal behavior to fit jointly large samples of GRB properties measured with BATSE, namely the peak photon energies $\\e_p^0$ of the prompt $\\nu F_\\nu$ spectra, the peak count rates, and the $\\gtrsim 1$ s $t_{50}$ durations. The predicted redshift distribution of GRBs can be used to test the model, but will require several score GRB redshifts from HETE-II and/or Swift. The redshift sample should preferably come from a single mission in order to minimize triggering differences and calibration uncertainties between detectors.  Rather than review the numerical results for a third time (see \\cite{dc99,der00a}), it seems more fruitful to devote this space to an analytic statistical model that confirms in all respects the numerical results.  We gain clarity at the expense of the accuracy found in \\cite{bd00}.  In section II, the standard argument that only supernova explosions are frequent and energetic enough to power the galactic cosmic rays is recited, while adumbrating certain weaknesses in the underlying assumptions.  The statistics of cosmological GRBs are treated in Section III. Depending on shock efficiency during the prompt phase, the power of the high-mass progenitors into an L$^*$ galaxy is evaluated. GRBs and FTs are shown to be capable of injecting sufficient energy into the ISM to power the hadronic CRs, as discussed in Section IV. Potentially observable consequences of the $10^4$-$10^7$ black holes that are formed from FTs in the Galaxy are considered in Section V, including a possible solution to the origin of some of the unidentified EGRET sources. The lineage of stars that explode as different novae types is sketched in Section VI. Section VII provides a summary and a brief discussion of the acceleration and adiabatic loss problem, that will be treated in more detail elsewhere \\cite{dh00}. ",
        "conclusions": ""
    },
    "0010/astro-ph0010087_arXiv.txt": {
        "abstract": " ",
        "introduction": "Many numerical simulations have been performed up until the present to study the distribution and dynamics of the interstellar medium in barred spiral galaxies. The behavior of the interstellar medium in a bar potential and the influence of a bar on the evolution of galaxies have been evaluated in those simulations (e.g., Athanassoula 1992; Combes, Gerin 1985). However, observational data are still insufficient to compare with the numerical simulation results. Since it takes so much time to map a galaxy with sufficient high angular resolution to resolve the bar and spiral structures, only a few barred spiral galaxies have been fully mapped in molecular lines so far (e.g., Sempere, Garcia-Burillo 1997).  We conducted CO mapping observations of barred spiral galaxies with the 45-m telescope at the Nobeyama Radio Observatory (NRO). The combination of the telescope and the $2\\times2$-beam SIS focal plane array receiver (Sunada et al.\\ 1995) results in a high angular resolution and high mapping performance, which allows highly efficient mapping observations of external galaxies. The purpose of our project is to investigate the distribution and kinematics of molecular gas in barred spiral galaxies. As a sample, we selected six galaxies (M 83, NGC 4321, NGC 253, NGC 3504, NGC 4303, NGC 6951) which are nearby and bright enough in CO to map and resolve the large-scale structures, such as the bar and spiral arms with the NRO 45-m telescope.  In this paper we present the results of one of our targets, NGC 3504. The basic parameters of NGC 3504 are listed in table 1. The central region of this galaxy has been previously mapped in CO by both an interferometer (OVRO: Kenney et al. 1993) and a single-dish telescope (IRAM 30m: Planesas et al.\\ 1997). While these maps show only a concentration of molecular gas in the center of the galaxy, we could obtain the distribution of molecular gas in the whole of the bar region because our observations were more sensitive and covered a larger area. From these data we were able to derive an upper limit to the pattern speed of the bar using our original method and the locations of the resonances. It is very important to know the pattern speed and locations of resonances in barred spiral galaxies, since it has been suggested that the resonances are related to large-scale structures in galaxies, such as inner rings, outer rings, and bar lengths (e.g., Combes 1996). Moreover, since the behavior of interstellar gas in a bar potential is different for fast and slow bars, the difference in the pattern speed of the bar affects the evolution of galaxies (Combes, Elmegreen 1993). In previous studies, the pattern speed of a bar was derived with an assumption that some feature is located at a resonance except for a few cases (Canzian 1993; Tremaine, Weinberg 1984; Sempere et al. 1995). In this framework, however, discussion about the relations between resonances and galactic structures is nonsense, since some results are settled in the assumption. Without such an assumption, we obtained a strict upper limit of the pattern speed in NGC 3504. It thus becomes possible to address the relation between the pattern speed and galactic structures.   \\begin{table*}[t] \\begin{center} Table~1.\\hspace{4pt}Parameters of NGC 3504. \\end{center} \\vspace{6pt} \\begin{tabular*}{\\textwidth}{@{\\hspace{\\tabcolsep} \\extracolsep{\\fill}}p{20pc}c} \\hline\\hline\\\\[-6pt] Center position (1950.0)$^\\ast$ & R.A.\\ = $11^{\\rm h}00^{\\rm m}28^{\\rm s}\\hspace{-5pt}.\\hspace{2pt}53$\\\\ & Decl.\\ = $28^{\\circ}14'31''\\hspace{-5pt}.\\hspace{2pt}2$\\\\ Morphological type$^\\dagger$ & ($R_{1}$)SAB($rs$)ab\\\\ Distance$^\\ddagger$ & 20 Mpc\\\\ Systemic velocity (LSR)$^\\ddagger$ & 1535 km s$^{-1}$\\\\ Position angle of major axis$^\\ddagger$ & $149^{\\circ}$\\\\ Position angle of bar$^\\ddagger$ & $143^{\\circ}$\\\\ Inclination angle$^{\\S}$ & $22^{\\circ}$\\\\ \\hline \\end{tabular*}  \\vspace{6pt}\\par\\noindent $*$ Condon et al.\\ (1990) \\par\\noindent $\\dagger$ de Vaucouleurs et al.\\ (1991) \\par\\noindent $\\ddagger$ Kenney et al.\\ (1993) \\par\\noindent $\\S$ Grosb{\\o}l (1985) \\end{table*} ",
        "conclusions": "In this paper, we presented the distribution and kinematics of the molecular gas in the barred spiral galaxy, NGC 3504. The results are summarized as follows:  (1) The molecular gas is highly concentrated in the central region of NGC 3504. The molecular gas is also elongated along the leading edges of the bar. The molecular ridges trace H\\,{\\footnotesize II} regions along the bar.  (2) There is a large velocity gradient of 50 km s$^{-1}$ at the molecular ridge, which corresponds to 130 km s$^{-1}$ in the galactic disk. The large velocity gradient indicates that the velocity component perpendicular to the bar decreases abruptly at the ridge. The rate of deceleration of 84 km s$^{-1}$ kpc$^{-1}$ is comparable with that in the arms of the grand-design spiral galaxy M 51.  (3) Since the position angle of the bar and the line of nodes of NGC 3504 are almost the same, we can derive an upper limit to the pattern speed of the bar from the radial velocity at the molecular ridge where molecular gas moves nearly along the bar. The upper limit of the pattern speed is 41 km s$^{-1}$ kpc$^{-1}$, which is much lower than that derived assuming that the location of the corotation radius is the bar end (77 km s$^{-1}$ kpc$^{-1}$).  (4) The derived pattern speed of the bar shows that there are two ILRs in NGC 3504. It is suggested that the intense star formation in NGC 3504 is occurring at the IILR, as suggested by numerical simulations. \\par \\vspace{1pc}\\par"
    },
    "0010/astro-ph0010278_arXiv.txt": {
        "abstract": "We determine the quantitative morphology and star formation properties of galaxies in six nearby X-ray detected, poor groups using multi-object spectroscopy and wide-field $R$ imaging.  The mean recessional velocities of the galaxy groups range from 2843 to 7558 \\kms.  Each group has 15 to 38 confirmed members ranging in luminosity from dwarfs to giants ($-13.7\\geq M_R-5$log$h\\geq -21.9$).  We measure structural parameters for each galaxy by fitting a PSF-convolved, two component model to their surface brightness profiles.  To compare directly the samples, we fade, smooth, and rebin each galaxy image so that we effectively observe each galaxy at the same redshift (9000\\kms) and physical resolution (0.87\\hi kpc).  The structural parameters are combined with [OII] measurements to test for correlations between morphological characteristics and current star formation in these galaxies.  We compare results for the groups to a sample of field galaxies.  We find that: 1) Galaxies spanning a wide range in morphological type and luminosity are well-fit by a de Vaucouleurs bulge with exponential disk profile.  2) Morphologically classifying these nearby group galaxies by their bulge fraction ($B/T$) is fairly robust on average, even when their redshift has increased by up to a factor of four and the effective resolution of the images is degraded by up to a factor of five.  3) The fraction of bulge-dominated systems in these groups is higher than in the field ($\\sim 50$\\% vs. $\\sim20$\\%). 4) The fraction of bulge-dominated systems in groups decreases with increasing radius, similar to the morphology-radius ($\\sim$ density) relation observed in galaxy clusters.  5) Current star formation in group galaxies is correlated with significant morphological asymmetry for disk-dominated systems ($B/T<0.4$).  6) The group galaxies that are most disk-dominated ($B/T<0.2$) are less star forming and asymmetric on average than their counterparts in the field. ",
        "introduction": "How the evolution of galaxies is affected by environment continues to be a fundamental question in astronomy.  Recent work by Schade $et~al.$ (1995, 1996), Abraham $et~al.$ (1996a, 1996b), Ratnatunga $et~al.$ (1999), and Simard $et~al.$ (1999, 2000) has focused on quantifying the morphologies of galaxies and relating them to environment and spectral star formation indicators.  These efforts concentrate on field or cluster galaxies.  Little has been done to quantitatively characterize galaxies in the intermediate, and common, environment of poor groups where the factors that might influence galaxy evolution differ from those in clusters or the rarefied field. One puzzle is that some poor groups with extended X-ray emission share certain properties with rich clusters, such as the shape of the galaxy luminosity function (Zabludoff \\& Mulchaey 2000) and a high early-type galaxy fraction (Zabludoff \\& Mulchaey 1998), but have lower velocity dispersions and galaxy number densities.  By fitting 2D surface brightness profiles and examining the residual light after the model is subtracted, it now is possible to derive structural parameters for a large sample of group galaxies that may then be compared with results from other environments and with the galaxies' spectral properties.  In particular, we can quantitatively and uniformly type galaxies by their bulge fraction, determine the fraction of galaxies with significant asymmetry, measure whether groups have a morphology-radius ($\\sim$ density) relation like clusters, compare group galaxy morphologies with the field, and assess whether the current star formation level in group galaxies is different than that in field galaxies of the same type.  As group galaxy number densities are closer to that of the field than of rich clusters, do the group galaxy populations differ significantly from the field?  Early-type galaxies are the minority in the field, making up only $\\sim$23-29\\% of the field population (van den Bergh $et~al.$ 1996; Driver $et~al.$ 1995; Shapley \\& Ames 1932).  Given the morphology-density relation and the low galaxy number densities and velocity dispersions of groups, it is surprising that high early-type fractions consistent with those of rich clusters have been found in some X-ray detected groups (Mulchaey \\& Zabludoff 1998).  Others (Hickson, Kindl, \\& Auman 1989; Williams \\& Rood 1987) also have found that approximately 50\\% of galaxies in compact groups are bulge-dominated systems.  These studies show that the relationship between environment and galaxy type is still unclear.  A limitation to understanding the link between galaxy morphology and environment in all these studies is how the galaxies are typed.  Typing was usually done by eye and information like star-forming regions, tidal tails, and shells were noted but not included in a quantitative analysis.  In addition, not only are there variations dependent on the author ({\\it e.g.} Fabricant, Franx, \\& van Dokkum 2000) but visual typing is difficult to do consistently for a sample with hundreds of galaxies. An important first step in improving our understanding of what determines galaxy morphology in different environments is to type galaxies in a uniform and reproducible way so that we can compare these galaxies in a quantitative manner.  An important aspect of understanding how galaxy morphology depends on environment is observing how the early-type fraction in groups varies, if at all, with position {\\it within} the group.  Although the morphology-density relation has been well-established for clusters (Dressler 1980; Whitmore \\& Gilmore 1991) and has been studied for groups found in redshift surveys (Ramella $et~al.$ 1999; Postman \\& Geller 1984), the former is defined for regions of much higher density and larger physical volume than that of poor groups and the latter is defined by the {\\it average} galaxy density of a group rather than as a function of position {\\it within} the group.  The morphology-radius ($\\sim$ density) relation has not been examined in poor groups, particularly those with extended X-ray emission that are likely to be real, bound systems.  Our large sample of galaxies projected within the virial radii of the poor groups, coupled with the quantitatively derived morphologies, makes it possible to test if a morphology-radius ($\\sim$ density) relation also exists in these groups.  We extend our analysis by testing how environment affects galaxy formation and/or evolution by observing whether the star formation levels and degree of morphological disruption in these galaxies change with environment.  We address these questions in a detailed manner by comparing only galaxies of the same morphological type.  In past studies, it was not possible to make this clean a comparison since we lacked a uniform classification scheme and a sample large enough to be cut into statistically meaningful subsamples.  With our large sample of spectroscopically-confirmed group members and quantitative measurement of physical galaxy characteristics, however, we now can use the bulge fraction to roughly isolate group galaxies of a particular, narrow class and to then compare their current star formation and asymmetry to their counterparts in the field.  In this paper, we characterize 171 galaxies in six nearby, X-ray detected, poor groups and 18 galaxies in the field using a new automated program that fits a PSF-convolved 2D surface brightness profile to an image of each galaxy and searches $\\chi^2$ space for the best model (GIM2D; Simard $et~al.$ 2000; Simard $et~al.$ 1999; Marleau \\& Simard 1998).  For each group, we have 15 to 38 spectroscopically confirmed members ranging in luminosity from dwarfs to giants ($-13.7\\geq M_R-5$log$h\\geq -21.9$) and in Hubble type from irregulars and Sd's to S0's and ellipticals.  By fitting two component profiles, we measure galaxy properties such as bulge fraction, half-light radius, and the fraction of light in an asymmetric component.  These structural parameters allow us to type the galaxy as bulge or disk-dominated and to define its asymmetry in a quantitative and reproducible manner.  Assuming a simple model, {\\it e.g.} a de Vaucouleurs (de Vaucouleurs 1948) bulge and exponential disk, we determine the goodness of each fit and check the robustness of our results for galaxies imaged at different effective resolutions.  We then compare the distributions of structural parameters and their relationship to current star formation (as measured by the 3727 \\AA ~[OII] emission line) for the group and field samples.  The sections of this paper are organized in the following manner: The data and the fitting technique are described in \\S2.  Also included in \\S2 is an explanation of how we move all the galaxies to a common redshift ($cz_0 = 9000$\\kms) and effective resolution (0.87$h^{-1}$ kpc) so that the samples may be directly compared.  In \\S3, we present the models and structural parameters fit by GIM2D for the group and field galaxies at their original and common redshift. We also test for relationships among the bulge fraction, asymmetry, and the current star formation, discussing differences between the group and field populations.  In \\S4, we discuss possible explanations for these differences.  Our conclusions are summarized in \\S5.  In this paper, we use H$_0=100 h$\\kms~Mpc$^{-1}$ and q$_0=0.5$. ",
        "conclusions": "Using multi-object spectroscopy and wide-field CCD imaging of six nearby X-ray detected, poor galaxy groups, we characterize the morphology and current star formation of the group members and of a comparison sample of field galaxies.  The large number of spectroscopically confirmed members in each group allows us to sample a wide range of galaxy types and luminosities, whereas most previous group studies have been limited to only the few brightest members.  By objectively classifying the galaxies into early and late-types using their bulge fractions, quantitatively measuring the asymmetry of each galaxy, and determining the [OII] emission line strength, we also investigate the relationship between galaxy morphology and current star formation for galaxies in these common, yet poorly understood, environments.  We measure the structural parameters and morphologically classify 171 group members and 18 field galaxies. Surface brightness models are fit to the galaxies at their observed ($cz_0$) and at a common ($cz=9000$ \\kms) redshift.  By fitting 2D models to the galaxies at these two redshifts, the composite of the group sample can be compared directly to the field, and the robustness of the models is tested.  Our conclusions are as follows:  I.  A de Vaucouleurs bulge with exponential disk model fits well the surface brightness profile of the group and field galaxies in our sample.  Of the 189 galaxies in our sample, 172 (91\\%) have residual fractions $<0.1$ once the model is subtracted (at $cz=9000$\\kms).  Our ability to fit models to the data does not depend strongly on morphological type or luminosity.  The majority of galaxies with high residuals have bars, spiral arms, tidal tails, star-forming regions, and dust lanes, features that cannot be fit with any smooth profile.  II.  Morphologically classifying galaxies by bulge fraction ($B/T$) is fairly robust on average for the galaxies in our sample of nearby groups, even when the galaxy's redshift is increased by up to a factor of four and the effective resolution degraded by up to a factor of five.  The average asymmetry ($R_A$) of these galaxies decreases slightly at the higher redshift ($0.02\\pm0.01$).  We note, however, that both $B/T$ and $R_A$ can change in either direction by as much as 59\\% and 30\\%, respectively, for individual galaxies.  III.  The fraction of early-type ($B/T\\geq0.4$) galaxies in X-ray detected, poor groups is higher than the field ($\\sim 50$\\% vs. $\\sim 20$\\% at $cz=9000$\\kms).  A common parent galaxy distribution for the group and field samples can be ruled out with $>95$\\% certainty.  IV.  Groups have a morphology-radius ($\\sim$ density) relation.  The average $B/T$ for the combined group sample decreases from the core out to $0.6$\\hi Mpc.  The anti-correlation between between $B/T$ and radius is $>95$\\% significant using the Spearman rank test.  V.  Current star formation, as characterized by [OII] emission, is correlated with high asymmetry ($R_A\\geq0.05$) for disk-dominated ($B/T<0.4$) systems in the groups.  Two possible explanations for this correlation are: (1) A physical connection between morphology and star formation exists such that when group galaxies are disturbed by an external force, their star formation also is boosted.  (2) The fact that asymmetrically distributed HII regions will cause a high $R_A$ (and that our code cannot distinguish between this case and, for example, a tidally disrupted galaxy). While we cannot always identify the sources of high asymmetry in disky, star-forming galaxies, there is less ambiguity for quiescent early-types.  There is a non-negligible fraction of early-type group galaxies (10/86=12\\%) that are both asymmetric and quiescent -- these are likely to be morphologically disturbed and/or to have a close companion.  VI.  For the most disk-dominated galaxies ($B/T<0.2$), the fraction with a high degree of asymmetry ($R_A\\geq0.05$) is significantly higher in the field than in groups (6/13=46\\% vs. 7/52=13\\%, respectively), the fraction that are forming stars is significantly higher in the field (9/13=69\\% vs. 16/52=31\\%, respectively), and the average star formation rate is significantly higher in the field ($14.3\\pm0.5$\\AA~vs. $7.4\\pm0.4$\\AA, respectively).  These points suggest that although there are some highly disturbed-looking group galaxies in our sample, very late-type galaxies in X-ray detected groups are less star forming and asymmetric on average than similar galaxies in the field."
    },
    "0010/astro-ph0010422_arXiv.txt": {
        "abstract": "The Hubble diagrams in $B$, $V$, and $I$ of a complete sample of 35 SNe\\,Ia with $(B-V)<0.06$ and $1200 < v \\la 30\\,000\\kms$ have a scatter of only $0\\fm1$, after small corrections are applied for differences in decline rate $\\Delta m_{15}$ and color $(B-V)$. The tightness of the Hubble diagrams proves blue SNe\\,Ia to be the best ``standard candles'' known. Their absolute magnitudes $M_{B,V,I}$ are calibrated by eight SNe\\,Ia with Cepheid distances from $HST$. Combining this calibration with the appropriate Hubble diagrams yields a large-scale value of $H_0=58.5\\pm6.3$ at the 90-percent confidence level.  The Hubble diagram of SNe\\,Ia has so small scatter that it seems feasable to determine $\\Lambda$ ``locally'', i.e. within $z\\la0.12$, once 100-200 SNe\\,Ia with good photometry will be available. Such a local determination would minimize evolutionary effects and K-term corrections.  Clusters of galaxies have provided useful Hubble diagrams through brightest cluster members, TF distances, and D$_n-\\sigma$ and fundamental plane distances, but with significantly more scatter ($\\sigma=0\\fm2-0\\fm3$) than SNe\\,Ia. The zeropoint calibration of these Hubble diagrams is an additional problem, which is aggravated by the high weight of any adopted distance of the Virgo cluster and by selection effects of clusters with only few well studied members. If a Virgo cluster modulus of $(m-M)=31.60\\pm0.20$ is adopted for calibration, --- a value which is well secured by Cepheids, SNe\\,Ia, and the TF method, and which also agrees with less definitive distances from the globular cluster luminosity function, novae, and the D$_n-\\sigma$ method, --- one finds $H_0=55\\pm5$. The reasons are explained why some authors have found higher values from clusters.  The determination of $H_0$ from field galaxies is beset by selection effects of magnitude-limited samples (Malmquist bias). Authors who have properly allowed for bias have consistently obtained $H_0\\approx 55\\pm5$ within $v\\la5000\\kms$ based on the TF and other methods. The calibration rests only on Cepheids, independent of any adopted Virgo cluster modulus.  Giving highest weight to SNe\\,Ia it is concluded that $H_0=58\\pm6$. ",
        "introduction": "$HST$ has brought an enormous progress in determining extragalactic distances by providing 26 Cepheid distances to late-type galaxies. This is of paramount importance not only for the determination of the Hubble constant $H_0$, but increasingly also for the physical understanding of individual galaxies whose linear sizes, luminosities, masses, radiation densities etc. depend on distance. Unfortunately the progress is confined to late-type galaxies, the distances of early-type galaxies having profited only indirectly.  An overview of the Cepheid distances from $HST$ is given in Table~1. Cepheids carry by far the largest weight for the foundation of the extragalactic distance scale. Their zeropoint is based on an adopted LMC modulus of $18.50$, which is confirmed by various distance indicators to within $\\pm0\\fm10$. Cepheid distances are uncontroversial to a large extent with possibly remaining small metallicity effects discussed in Section~2.5.  \\begin{table} \\begin{center} \\caption{Cepheid distances from $HST$.} \\label{tab:1} \\begin{tabular}{llc} \\noalign{\\bigskip} No. of Galaxies & Authors & No. of SNe\\,Ia \\\\ \\hline 18 Cepheid distances & Freedman 2000          & 1 \\\\ 1 Cepheid distance   & Tanvir et~al. 1995     & 1 \\\\ 7 Cepheid distances  & Saha et~al. 1999, 2000a & 8 \\\\ \\end{tabular} \\end{center} \\end{table}  The 26 galaxies with Cepheid distances from $HST$, all beyond the Local Group, yield a mean value of $H_0 = 65\\pm4$ [km\\,s$^{-1}$\\,Mpc$^{-1}$]. Yet these galaxies lie within $v=1200\\kms$ which is too local for this determination having any cosmic significance.  An additional distance indicator is therefore needed which can be calibrated by means of the available Cepheids and which carries the distance scale out to $\\ga10\\,000\\kms$, i.e. well beyond the influence of peculiar and streaming motions. This distance indicator must be {\\em proven\\/} to be reliable and its intrinsic dispersion must be known in order to control selection effects (cf. Section~4).  Many distance indicators have been proposed, but the proof of their reliability is extremely difficult. The demonstration that they can reproduce a limited number of Cepheid distances, which carry themselves individual errors of up to $\\sim0\\fm2$, is not good enough, and the internal dispersion remains ill defined. Just because various distance indicators are said to agree amongst themselves is no proof that they form a correct distance scale. If they have similar intrinsic dispersion and if each is not corrected for systematic effects of observational bias errors, their agreement is spurious.  The only satisfactory way to prove potential distance indicators to be useful for the determination of $H_0$ is the demonstration that they define a linear relation of slope $0.2$ (corresponding to linear expansion) in the Hubble diagram out to $\\ga 10\\,000\\kms$. The scatter about this line provides in addition the intrinsic dispersion of the method if proper allowance is made for observational errors and for the influence of peculiar motions which, however, at $10\\,000\\kms$, are negligible for all practical purposes.  This requirement of reliable long-range distance indicators is very well met by supernovae of type Ia (SNe\\,Ia), and they can be calibrated by Cepheids. SNe\\,Ia offer therefore the optimum route to $H_0$ and are discussed in Section~2. Cluster distances follow in Section~3; they also define a useful Hubble diagram, but their zeropoint calibration is still under debate. The most difficult and least satisfactory way to $H_0$ by means of distances of field galaxies is discussed in Section~4. The conclusions are given in Section~5. ",
        "conclusions": "The exceptionally tight Hubble diagram of blue SNe\\,Ia out to $\\sim\\!30\\,000\\kms$ combined with their mean absolute magnitude, which is determined from eight galaxies whose Cepheid distances have been determined with $HST$, offers the ideal instrument for the calibration of the large-scale value of $H_0$. The result is -- after allowance is made for the relatively weak dependence of SNe\\,Ia luminosities on decline rate $\\Delta m_{15}$ and color and for some small systematic effects -- $H_0=58.5\\pm6.3$.  The quoted external error is mainly determined by the calibrating Cepheid distances. Although Cepheids are generally considered to be the most reliable distance indicators, certain questions remain as to their zeropoint calibration through LMC and the effect of metallicity variations.  Phillips et~al.\\ (1999) and Suntzeff et~al.\\ (1999) have adopted a steeper relation between the SNe\\,Ia luminosity and the decline rate $\\Delta m_{15}$, which is not supported by the present larger sample. If their steep slope is taken at face value $H_0$ would be increased by 5 percent. Gibson et~al.\\ (2000) have increased the Cepheid distances of Saha et~al.\\ and Tanvir et~al.\\ in Table~2 by $0\\fm11$ on average on the basis of additional Cepheids of lower weight. This would increase $H_0$ by another 5 percent. If one choses to cumulate the two effects one could defend a value of $H_0=64$ (cf. Freedman 2000).  The calibration of $H_0$ through the Hubble diagram of clusters and through field galaxies in Section~3 and 4 has given values around $H_0=55$, i.e. somewhat smaller, if anything, than the value from SNe\\,Ia. Since the latter offer the most direct way to $H_0$, they are given the highest weight. A value of \\begin{equation}\\label{eq:12} H_0 = 58 \\pm 6 . \\end{equation} is therefore adopted as the best estimate available at present.  The two most frequent errors in the extragalactic distance scale, which can lead to values of $H_0 > 65$, are too small a Virgo cluster distance and Malmquist bias.  The distance of the Virgo cluster is still sometimes equated to the mean Cepheid distance of three Virgo spirals which are visibly on the near side of the cluster due to their high resolution into stars, -- in fact they had been preselected on grounds of their high resolution to facilitate work with $HST$. Thus, -- even if the larger cluster distance of $(m-M)_{\\rm Virgo}=31.60\\pm0.20$ (cf. Table~4 and 5) was not required by a fourth Cepheid distance of a bona fide cluster member (NGC\\,4639) and in addition by the three SNe\\,Ia and the TF relation, -- the three Cepheid distances of highly resolved galaxies could set only {\\em a lower limit on the cluster distance}, i.e. $(m-M)_{\\rm Virgo}>31.05$.  The distance of the Virgo cluster has entered with high or even full weight into the zeropoint calibration of the D$_n-\\sigma$, FP, and SBF methods and occasionally even of the turnover magnitude of the GCLF. (The significance of the GCLF, although not yet proven beyond doubt, is just that the calibration rests on the Cepheid distance of M\\,31 {\\em and\\/} on the RR~Lyr distances of Galactic globular clusters [cf. Tammann \\& Sandage 1999]). If one inserts in these cases too low a Virgo cluster distance, one obtains incorrect values of $H_0\\approx70$, or even more, by necessity.  Locally calibrated distance indicators with non-negligible scatter can exclusively be applied to distance-limited samples (or volume-limited samples in case of clusters), but only apparent-magnitude-limited samples are available, which frequently are not even complete to any specific magnitude limit. The statistical differences between a distance- and magnitude-limited sample are known for almost eighty years (Malmquist 1920, 1922), but this insight is violated again and again, always leading to too high values of $H_0$. Malmquist's message is that control of the objects missing out to a fixed distance from a magnitude-limited catalog is equally important as the catalog entries themselves.  The future will see high-weight determinations of $H_0$ from the Sunyaev-Zeldovich effect, from gravitationally lensed quasars, and from the CMB fluctuations. At present these methods yield values of $50\\la H_0 \\la 70$, which is not yet competitive with the solution of SNe\\,Ia. In the case of the CMB fluctuations one can solve for $H_0$ only in combination with other (poorly known) cosmological parameters. Therefore an {\\em independent}, high-accuracy determination of $H_0$, to be used as a prior, is as important as ever."
    },
    "0010/astro-ph0010108_arXiv.txt": {
        "abstract": "We present predictions of the evolution of the light elements, Li, Be, and B, in the early epochs of the Galactic halo, using a model of supernova-induced chemical evolution based on contributions from  supernovae (SNe) and cosmic rays (CRs), as recently proposed by Tsujimoto \\etal and Suzuki \\etal.  This model has the great advantage of treating various elements self-consistently, even under inhomogeneous conditions, as might arise from stochastic star formation processes triggered by SN explosions.  The most important prediction from our model is that the abundances of light elements in extremely metal-poor stars might be used as age indicators in the very early stages of an evolving halo population, at times when the abundances of heavy elements (``metallicity'') in most stars are dominated by local metal enrichment due to nearby SN events, and is poorly correlated with age.  Plots of the expected frequency distribution of stars in the age {\\it vs.} elemental abundance diagram show that the best ``cosmic clock'' is the $^6$Li abundance.  We have derived relationships among various cosmic-ray  parameters such as energy input to CRs by SNe, the spectral shape of the CRs, and the chemical composition in CRs, and find that we can reproduce very well recent observations of $^6$Li, Be, and B in metal-poor stars. Although our model is successful for certain sets of cosmic-ray parameters, larger energy should be absorbed by energetic particles from each SN than required to the current situation of Galactic disk.  We discuss an alternative hypothesis of AGN activity in the early Galaxy as another possible accelerator of CRs. ",
        "introduction": "\\label{sec:intr}  Recent observations of elemental abundances in extremely metal-poor halo stars (Mcwilliam \\etal 1995, Ryan \\etal 1996) uncovered evidence which suggests that the chemical evolution of the early Galactic halo was quite inhomogeneous.  For example, the abundances of various elements exhibit a large scatter, even for stars having the same metallicity (as quantified by [Fe/H]), in contradiction to the predictions of simple one-zone models which, under the assumption that gas of the interstellar medium (ISM) is well-mixed, suggest that overall stellar metallicity should be strongly correlated with age.  The amount of ejected heavy elements from supernovae (SNe) shows a strong dependence on the mass of SN progenitors, and furthermore, this dependence is different for different elements (Woosley \\& Weaver 1995; Tsujimoto \\& Shigeyama 1998).  Audouze \\& Silk (1995) have further argued that the observed abundance patterns for metal-poor stars reflect the elemental abundance ratios of ejected elements by SNe with a discrete range of progenitor masses.  As a result of the limited mixing of the ISM in the early stages of the Galactic halo, the chemical compositions of these stars are apparently providing information on SN events that occurred in their local environment, during the first epochs of star formation.  The situation described above can be understood by a simple order of magnitude argument as follows.  We define a volume of Galactic disk as $V_{\\rm d}=2\\pi R^2 z$, where $R$ is the radius of the disk and $z$ is the half thickness. Then it follows that $V_{\\rm d} \\simeq 200 \\; {\\rm kpc}^3$, when adopting $R=15\\;{\\rm kpc}$, and $z=200\\;{\\rm pc}$. The SN-rate in the current disk is about one SN per 30 yrs, and a typical SN remnant (SNR) shell can expand to $\\sim$ 100 pc before diffusing away in about 3 Myr (Tsujimoto, Shigeyama, \\& Yoshii 1999; hereafter TSY) if its expansion is not stopped by merging with other SNR shells. Then, the swept-up volume of ISM by all the coeval SNR shells is about $V_{\\rm sw}\\simeq$400 kpc$^3$, exceeding $V_{\\rm d}\\simeq$ 200 kpc$^3$, which implies that the SNR shells easily merge with one another.  On the other hand, because the halo extends to 50 kpc, its total volume, $V_{\\rm h}$, is expected to be larger than that of current disk by three to four orders of magnitude.  Provided that the total SN-rate in the Galaxy at early epochs was less than $\\sim$ 50 times of that of today, which is acceptable according to the analysis of star formation history by Madau \\etal (1998), it follows that $V_{\\rm sw}\\ll V_{\\rm h}$ at early epochs. This indicates that each SNR in the halo could survive without merging with other SNR shells. The heavy elements synthesized and ejected from a SN are confined in its SNR shell, and do not mix easily with those originating from other SNe, or already existing in the interstellar medium.  TSY presented a SN-induced chemical evolution model based on the assumption of incomplete mixing of the heavy elements in the halo. In their model {\\it all} of the stars at early epochs were born in SNR shells, under the assumption that these shells, due to their relatively high densities, are suitable sites for star formation.  Their model explains the observed large scatter of heavy element abundances in metal-poor halo stars very well, especially the trend of europium abundances.  An important prediction of this model is that the metallicity of metal-poor stars has no one-to-one correspondence with age. Suzuki, Yoshii, \\& Kajino (1999; hereafter SYK) extended this model to include an analysis of light elements, such as $^6$Li, $^9$Be, and B ($^{10}$B+$^{11}$B), which are mainly produced by reactions involving Galactic cosmic rays (GCRs), and demonstrated that this model also reproduces the observed trends of $^9$Be and B data. SYK proposed a new scenario that GCRs originate directly from acceleration of the SN ejecta, and also from acceleration of particles in the ISM swept up by SNR. SYK pointed out that the one might expect that the abundance of light elements, $^6$LiBeB, in metal-poor stars should exhibit a much better correlation to their age, in sharp contrast to the situation for their heavy element abundances.  There are (at least) two significant issues concerning the production of light elements in the early Galaxy which must be confronted.  The first is that the observed linear trend of Be and B with metallicity contradicts with that derived from the usual spallation processes of energetic protons or $\\alpha$-particles impinging on CNO elements in the ISM.  The production rate of the light elements appears {\\it constant} with respect to metallicity, when we take the observations at face value, while the usual spallation processes predict that the production rate should be {\\it proportional} to metallicity, so that a quadratic trend of Be and B with metallicity is expected.  As one possible solution, Duncan \\etal (1992) and Yoshii, Kajino, \\& Ryan (1997) proposed an ``inverse'' spallation process applied in the early Galaxy, with energetic CNO nuclei originating from freshly synthesized SN-ejecta impinging on protons and $\\alpha$-particles in ISM producing Be and B.  The production rate predicted from these inverse processes is almost constant as a function of metallicity, because SNe of metal-poor progenitors synthesize a comparable amount of CNO elements to metal-rich SNe.  The condition for these processes to work is that synthesized CNO elements in SNe should be efficiently injected into the acceleration region around the SNR shock.  However, based on a reanalysis of elemental abundances of present-day GCRs and an examination of acceleration process thought to be provided by the SNR shock, Meyer \\& Ellison (1999) and Ellison \\& Meyer (1999) have recently argued that most GCRs were accelerated out of the ISM or circumstellar material instead of SN ejecta, although they suggested that, owing to the lack of CNO in the ISM at early epochs, energetic CNO out of fresh SN ejecta might still play a role in light element production.  As a compromise solution, SYK claimed that if $\\sim 2$\\% of GCRs originate from SN-ejecta, in agreement with the estimate by Meyer \\& Ellison (1999) and Ellison \\& Meyer (1999), the linear trend of BeB with metallicity can be achieved for stars with [Fe/H]$<-1.5$.  Fields \\& Olive (1999) argued that oxygen, instead of iron, should represent the ``metallicity,'' because BeB elements are mainly produced by spallation of oxygen in the early stages of the Galaxy.  They claimed that a quadratic BeB trend with O results, when using the increasing trend of [O/Fe] inferred from the observations of Israelian \\etal (1998) and  Boesgaard \\etal (1999) for the stars of extremely low metallicity.  However, this trend of oxygen relative to iron is still controversial, and also differs from the trend of other $\\alpha$-elements such as Mg (Fields \\etal 1999 also mentioned this problem).  The second issue concerns the shortage of available energy to accelerate GCRs in order to produce the observed level of abundances of light elements. Ramaty \\etal (1997) pointed out the importance of the ``energy budget'' of GCRs when investigating the evolution of light elements, and they concluded that GCRs should be more metal-rich than the ambient matter to account for the observed Be abundance in metal-poor stars.  The energetics of the production of light elements in SNRs has been intensively studied by Parizot \\& Drury (1999a,b; hereafter PDa,b).  They also concluded that energy input by individual SNe is about one order of magnitude lower than that required to reproduce the observations.  To solve this discrepancy, the a superbubble model was suggested by Higdon \\etal (1998) as the source of metal-rich energetic particles (EPs).  Since the material in superbubbles becomes metal rich due to collective explosions of SNe, enough CNO nuclei can be accelerated by successive SNe, with no need for the artificial selective acceleration of CNO nuclei.  From extensive calculations, Parizot \\& Drury (1999c; hereafter PDc) concluded that superbubbles are capable of producing enough of the light elements to satisfy the observed level.  It should be noted, however, that the superbubble model cannot simultaneously explain the large scatter of heavy elements in metal-poor stars (see more discussion of this point in \\S \\ref{sec:sbmd}). In this sense the second problem still persists.  In this paper we formulate the model of SN-induced chemical evolution in \\S \\ref{mdlgce}, and the model of EPs in \\S \\ref{sec:gcr} and \\S \\ref{sec:leps}. We argue that an acceptable relation among three unknown parameters (input energy to EPs per SN, the spectral shape of EPs, and composition of EPs) can be found that reproduces the observed abundance of light elements (\\S \\ref{sec:ccpr}).  The results of our model are presented in \\S \\ref{sec:rslt}. We compare the predicted evolution of light elements with recent observations in \\S \\ref{sec:cmpobs}, and we predict the evolution of the composition of GCRs in the early stage of the Galactic halo in \\S \\ref{sec:cmpcr}.  In \\S \\ref{sec:tmr} we discuss the feasibility of using the abundances of each of the light elements as age indicators for old metal-poor stars.  Other related topics are discussed in \\S \\ref{sec:dscs}. Our model is compared with an alternative model of superbubbles as a source of light elements in \\S \\ref {sec:sbmd} in terms of energetics of GCRs, and the possibility of considering AGNs as another accelerator of EPs is discussed in \\S \\ref{sec:agn}. ",
        "conclusions": "\\label{sec:dscs}  \\subsection{Cosmic Ray Energetics and Light-Element Production --- Comparison with the Superbubble Model} \\label{sec:sbmd} As discussed in \\S \\ref{sec:ccpr}, in order to explain the observed $^6$Li and Be abundances in metal-poor stars, one SN had to provide $E_{\\rm gcr}\\sim 4 \\times 10^{50}$ erg to cosmic rays, even though we adopted a rather softer spectrum, $\\gamma=2.7$, which can produce light elements more efficiently. This value of GCR energy per SN is higher than that required today, and is often discussed as being problematic (PDa,b; Ramaty \\etal 1997, 2000).  The superbubble model was introduced (Higdon \\etal 1998; PDc; Ramaty et al. 2000) as one possible way to achieve high efficiency of light-element production while not violating the energy requirements.  According to detailed calculations by PDc, there are two reasons why superbubbles are suitable sites for effective production of the light elements.  First, CNO elements are more efficiently accelerated, because heavy elements ejected by successive explosions of many (up to $\\sim 100$) SNe accumulate inside the superbubble, and the gas becomes much more metal rich (approaching [Fe/H]$\\sim -1$) as compared to the gas in the outer region ([Fe/H]$\\sim -4$).  These energetic CNO nuclei will be spalled to produce light elements.  Second, the spectrum of EPs in the superbubble is expected to be $E^{-\\alpha}\\exp (-E/E_0)$, where $\\alpha \\simeq 1\\sim 1.5$ and $E_0$ is a few hundred MeV (Bykov, \\& Toptygin 1990; Bykov 1995).  As a result, they more effectively produce light elements than the so-called momentum spectrum predicted from shock acceleration of the first-order Fermi process (Blandford \\& Ostriker 1978).  An important feature of this spectrum is an enhancement of EPs in the low-energy region, around $10-100$ MeV/A, where the cross sections to produce light elements, especially $^6$Li, have their maximum value (the bottom panel of Fig.1).  \\begin{figure}[p] \\epsfxsize=15cm \\epsfysize=12cm \\epsfbox{fig8.ps} \\caption{Number of Be and $^6$Li atoms produced per unit GCR energy. Solid lines show the results of our inhomogeneous halo model with $f_{\\rm cr}=0.007$ and $\\gamma=2.7$. Dashed lines show the results of the SN-ejecta-enriched superbubbles by Ramaty \\etal (2000).} \\end{figure}  Here, we would like to compare the superbubble model with our inhomogeneous halo model with respect to GCR energetics.  In Fig. 8, adopting the spectral index $\\gamma$=2.7, and the parameter $f_{\\rm cr}$=0.007, which determines the primary heavy GCRs (\\S \\ref{sec:ccpr}), we show the number of Be and $^6$Li atoms produced per unit GCR energy (atoms/erg) for our model, overlayed with recent results of the superbubble model by Ramaty et al. (2000).  There are two primary differences between their superbubble model assumptions and our individual SN scenario with respect to the nature of the GCRs.  One is the spectrum of the GCRs: Ramaty et al. (2000) use the spectrum of $q \\propto p^{-\\gamma} \\exp(-E/{E_0})$, where $p$ is the momentum of a particle and $E_0$ ($\\sim 10$ GeV/A) is the cut-off energy, calculated on the basis of multiple shock acceleration in superbubbles (Bykov 1995), while ours is a so-called momentum spectrum (eq.(\\ref{eqn:crfx})) predicted from the Fermi acceleration mechanism in the shock of individual SNRs. The other difference is the assumed chemical composition of GCRs: the CNO abundance by mass adopted in Ramaty \\etal (2000) is almost as high as 0.1, while that adopted in our model lies in the range 0.005--0.01 during the entire epoch of early Galactic halo (Fig. 4).  It turns out that their adopted superbubble GCR spectrum yields a similar production quantity of the light elements as ours, since in the energy region 10--100 MeV/A, in which the light elements are produced the most efficiently, it is {\\it identical} to the momentum spectrum.  This is a result of Ramaty \\etal (2000) adopting a higher cut-off energy ($\\sim$ 10 GeV/A) than that in PDc.  Therefore, both the models yield almost the same number of $^6$Li atoms per unit GCR energy, much of which is produced by $\\alpha$ + $\\alpha$ fusion reactions in addition to the spallation of CNO elements.  On the other hand, the number of Be atoms produced per unit GCR energy based on our model is as low as 10\\% of the results calculated by Ramaty \\etal (2000), because the CNO abundance in GCRs adoped in our model is lower by about one order of magnitude than that adopted in their superbubble model.  The efficiency of Be production per unit GCR energy derived from our model is $\\sim$ 10$\\%$ of the superbubble model by Ramaty \\etal (2000). Therefore, we adopted a higher energy input, $E_{\\rm gcr}=4\\times 10^{50}$ erg, to GCRs per SN, but this value is only a factor of three higher than that adopted in their model ($E_{\\rm gcr}=1.5\\times 10^{50}$).  This is because the derived Be abundance from their model is higher by a factor of three at the same [Fe/H] than the results of our best-fit model to the observations.  Our adopted value of $E_{\\rm gcr}=4\\times 10^{50}$ erg, though higher than the value of $10^{50}$ erg per SN required to maintain the energy density of cosmic rays in the current disk (e.g., I \\S 4 in Berezinski$\\breve{\\rm i}$ \\etal 1990), is not unacceptably high, given that the nature of SNR in the early stage of an evolving halo might be completely different from those in the current disk, for the following reasons.  First, metal-poor SNRs suffer little from radiative losses because of the absence of metals, so they could survive for a longer period, and should be able to accelerate EPs before losing power and dissipating.  Second, the SNR shell could keep its identity until the late SNR phase because merging with other shells could be avoided, owing to the low SN density in the early Galactic halo---See \\S \\ref{sec:intr} for order of magnitude estimates. It is therefore expected that the acceleration of EPs in the early halo should have been more efficient than for the current disk of the Galaxy.  The efficiency of acceleration of GCRs is expected to decrease from the early Galaxy to the present level, as the abundance of heavy elements in the ISM increases.  Therefore, in more sophisticated models, $E_{\\rm gcr}$ should be regarded as a variable with respect to cosmic time, or more precisely, with respect to the metallicity of the SNR shells.  The abundance pattern of heavy elements in metal-poor stars can distinguish between the superbubble model and our inhomogeneous halo model, because the prediction from the superbubble model is completely different from ours. According to PDc, the superbubble is surrounded by a very metal-poor shell composed of ISM that is swept up by the expanding bubble. In the bubble, heavy elements ejected by SNe with various progenitor masses are mixed with the ISM evaporating from the shell.  Since the timescale of mixing ($\\sim$ 1 Myr), estimated by the size of the bubbles divided by the sound velocity, is smaller than the timescale of evolution of the bubble ($\\sim$ 30 Myr), the chemical composition of the bubbles are expected to be quite homogeneous.  An important result of superbubble model by PDc is that stars formed from the material of the bubble, diffusing into the metal-poor ISM, should exhibit a constant abundance ratio for any combination of elements, for instance, Be/Fe, and O/Fe, although their elemental abundances may vary, reflecting the progress of diffusion.  On the other hand, our model predicts that stars formed in each SNR shell, originating from various-mass SN progenitors, should exhibit a remarkable scatter in the abundance ratios of heavy elements.  A realistic situation may not be so simple, because some stars may have formed in the superbubble environment, and some may have formed from individual SNR shells, as recently discussed by Parizot \\& Drury (2000).  We suggest that observations of the element Eu is one means by which these two models could be distinguished.  The synthesized mass of Eu in a SN decreases with increasing SN progenitor mass, an {\\it opposite} trend to the behavior of the majority of heavy elements, including Fe (Shigeyama \\& Tsujimoto 1998). Therefore, if {\\it all} stars at early epochs were formed in individual SNR shells, they must be distributed along a decreasing line of log(Eu/Fe) as a function of [Fe/H] (see Fig.3 of Shigeyama \\& Tsujimoto 1998).  On the other hand, if superbubbles are the dominant site of star formation in the early Galaxy, most stars should have an identical Eu/Fe ratio, and be distributed along a horizontal line in the [Fe/H]-log(Eu/Fe) plane.  Relevant data obtained by various observers (Mcwilliam \\etal 1995, Ryan \\etal 1996, Luck, \\& Bond 1985, Gilroy \\etal 1988, and Magain 1989) seem to favor our model. However, this cannot yet be taken as definitive evidence, because the reported data still have quite large errors.  Future observations, especially those with high accuracy, will hopefully clarify this picture.  \\subsection{An AGN in Our Galaxy As Another Accelerator} \\label{sec:agn} \\subsubsection{Activities of Galactic Nucleus} An active galactic nucleus (AGN), which might have existed in the center of our Galaxy, could have contributed to the early acceleration of EPs. (Production of light elements in AGN was studied by Baldwin \\etal 1977 and  Crosas \\& Weisheit 1996.)  It has been argued from the observed number of QSOs that these objects were very active at redshift $z\\sim2-3$ and stopped their activities by the present time (Shaver \\etal 1996).  Moreover, evolutionary trends of bright QSOs slightly precedes (Richstone et al. 1998) the trend of cosmic star formation history (Madau \\etal 1996; Madau, Pozzetti, \\& Dickinson 1998).  Richstone \\etal (1998) interpreted this fact in such a way that the birth of QSOs was associated with the formation of the spheroidal components of galaxies, and pre-dated the active phase of star formation through the evolution of galaxies. If this scenario is correct, the total energy of a galaxy was supplied first by an AGN, then gradually by SNe. It is naturally supposed, therefore, that most EPs at early epochs were accelerated by an AGN/QSO source.  Observations of stellar motions in the region of the center of the Galaxy show that the mass density at the very center is quite high ($\\ge 2.2\\times 10^{12}\\Msun\\;{\\rm pc^{-3}}$), enough to accommodate a compact object like a massive blackhole (Genzel \\etal 1997), possibly providing evidence of an early AGN in the Galaxy.  There are two mechanisms which might conceivably supply sufficient energy to an AGN/QSO.  One is a so-called fuel (accretion-powered) engine, and the other is a so-called fly-wheel (rotation-powered) engine (Nitta et al. 1991, Nitta 1999).  While the energy supplied by the fuel engine is extracted from the gravitational energy released from an essentially {\\it infinite} amount of the accreting matter, the energy supplied by the fly-wheel engine originates from {\\it finite} amount of rotation energy of the central blackhole.  In this scenario, the luminosity of AGN/QSO is expected to evolve as follows: An AGN/QSO is extremely luminous just after being formed, because either engine is active, but after a certain period the fly-wheel engine ceases its activity when the rotation energy is exhausted.  After that, the AGN/QSO becomes fainter, to a level as low as or less than the Eddington luminosity. Nitta (1999) discussed the statistical properties of the evolution of AGN/QSOs using the model of a Kerr-blackhole engine.  According to his work, a typical AGN with mass of $ 10^8\\Msun$ has a lifespan for the fly-wheel engine of $\\sim$ 1 Gyr, which is, interestingly, similar to a typical time scale of the evolution of the Galactic halo.  The luminosity of such a typical QSO/AGN is about $10^{46}$ erg s$^{-1}$, which is much larger than the energy input by SN explosions into the current disk ($\\sim 10^{42}$ erg s$^{-1}$).  Even if the SN rate, or almost equivalently the SFR, was higher by a factor of several factors of ten in the past (Madau \\etal 1998), a typical galactic nuclei could supply at least by 2 orders of magnitude more energy than the total SNe in the Galaxy at that time.  Other interesting results concerning activities of the Galactic center have been obtained from X-ray observations (Koyama et al. 1996), which detected the existence of a hot plasma containing $\\sim 10^{54}$ erg in the central region.  Judging from the lifetime of the plasma ($\\sim 50000$ yrs), these authors concluded that continuous energy generation of $10^{41-42}$ erg s$^{-1}$ is required, although a SN origin is implausible because of various observational facts.  They also reported an emission line of iron in the cold molecular clouds near the Galactic center, possibly due to irradiation from the center which was brighter in the very recent past ($\\sim$300 yrs).  Koyama et al. (1996) argued from these observations that the Galactic nucleus still has intermittent activity, with a time-averaged luminosity $\\sim 10^{41-42}$ erg s$^{-1}$, that might result from the the activity of a fuel engine (accretion-powered) blackhole.  It should be noted that such a value is quite high, as compared to energy suppliers in our Galaxy today, being almost comparable to the energy input from the total SN explosions in the present Galactic disk ($\\sim 10^{42}$erg s$^{-1}$).  \\subsubsection{Light Element Production by AGN} If the Galactic nucleus was once very active, with the energy supplied from the rotation of the central blackhole and/or the accreting matter onto the blackhole, it is natural to imagine that EPs accelerated in the shock around the AGN would have enhanced the production of light elements at early epochs. Adopting typical values of particle velocity $v\\sim (0.5-1)\\times 10^{10}$ cm$\\;$s$^{-1}$ and path length against ionization loss $\\Lambda_i \\sim 1-10$ g$\\;$cm$^{-2}$ (Northcliffe \\& Schilling 1971) for GCRs with $E_{\\rm gcr}\\sim 10-100$ Mev/A, which most effectively produce the light elements, we estimate the lifetime of these GCRs as \\begin{equation} \\tau_{\\rm gcr}=\\frac{\\Lambda_i}{\\rho v}\\sim 1-10 {\\rm Myr}, \\end{equation} where $\\rho$ is the density of ambient gas, having a typical value of $\\sim 10^{-24}$ g$\\;$cm$^{-3}$.  With the diffusion coefficient $D_{\\rm GCR}\\sim 10^{29}$ cm$^2$s$^{-1}$ for the Galactic halo today (see eq. (\\ref{eq:difcr})), these GCRs propagate over the distance of $1-3$ kpc away from the Galactic center and are expected to enhance the production of light elements there.  We now consider how these AGN-accelerated GCRs could affect the chemical evolution of ``clumpy'' clouds which make up the entire Galactic halo at early epochs.  Studies of the kinematics of stars in the solar neighborhood indicate that metal-poor halo stars have a variety of orbital eccentricities spanning from $e\\sim 0$ to 1, whereas metal-rich disk stars rotate in almost circular orbits of $e\\sim 0$ in the Galactic disk (Yoshii \\& Saio 1979; Norris, Bessell, \\& Pickles 1985; Chiba \\& Yoshii 1997).  About 70\\% of nearby halo stars with [Fe/H]$\\le-2.2$ have orbits with $e>0.5$ (Chiba \\& Beers 2000) which are accessible to the region of central few kiloparsecs, given that the solar distance is 8.5 kpc from the Galactic center.  The observed orbital eccentricities of halo stars more or less reflect the initial motion of clouds from which such stars were born, because their orbital angular momentum is preserved, with only rare interactions between the gas and other stars.  Even if dissipation processes have played a role, initial clouds were dynamically in a more chaotic state, having more chances to pass through the central Galaxy in the past.  Accordingly, many halo stars observed in the solar vicinity today were born from clouds which could pass through the central Galaxy at least once in an orbital period $T_p$.  Since $T_p$ should be comparable to the dynamical timescale of the Galaxy $\\tau_{\\rm dyn,Gal}\\sim 10^8$ yr, we obtain the following inequalities: \\begin{equation} \\label{eq:agnev} \\tau_{\\rm halo}(\\sim 10^9 {\\rm yr}) > T_p(\\sim 10^8 {\\rm yr}) > \\tau_{\\rm evol,c}(\\sim 2\\times 10^7 {\\rm yr}) , \\end{equation} where $\\tau_{\\rm halo}$ is a typical duration of the formation of the Galactic halo, and $\\tau_{\\rm evol,c}$ is the typical timescale of cloud evolution, characterized by the lifetime of $10M_\\odot$ stars inducing the formation of new stars (\\S \\ref{sec:gcr}).  The first inequality in eq. (\\ref{eq:agnev}) indicates that the clouds could have passed near the center of the Galaxy at least 10 times during the halo phase.  Given that AGN activity lasts for $\\sim 1$ Gyr, the chemical evolution of the clouds is affected by AGN-accelerated GCRs about 10 times, and the production of light elements is intermittently enhanced once in every $\\sim 10^8$ yr.  Because of the random and chaotic motion of the clouds, the light elements originating as a result of AGN activity are distributed over the entire halo, although their production site is confined in the central few kiloparsecs.  As a result, the light-element abundance of halo stars would have been increased more rapidly, compared to the case of an exclusive SNe origin.  The trend of Be and B abundances with metallicity would be flatter than quadratic, and the trend of $^6$Li would be even flatter than linear.  These expected consequences are similar to the model prediction by Yoshii et al. (1997), under the assumption that the GCR flux was higher in the past, owing to more effective confinement of GCRs in the early Galaxy.  The second inequality in eq. (\\ref{eq:agnev}) indicates that the AGN-accelerated GCRs affect the production of light elements intermittently on longer timescales, compared to that of the stellar generations induced by SNe.  As a result, the trend of light-element abundance might not be expected to be a a simple function of metallicity.  Although precise estimates of this extra AGN effect required more detailed modeling of the evolution of clumpy clouds in the Galactic halo, it is expected that the light-element abundance could still be used as an age indicator, because GCRs originating from the AGN propagate as uniformly in the clouds as those from SNe (\\S \\ref{sec:gcr})."
    },
    "0010/astro-ph0010614_arXiv.txt": {
        "abstract": "We have calculated fast direct spectral model fits to two early-time spectra of the Type-II plateau SN 1999em, using the SYNOW synthetic spectrum code. The first is an extremely early blue optical spectrum and the second a combined \\emph{HST} and optical spectrum obtained one week later.  Spectroscopically this supernova appears to be a normal Type II and these fits are in excellent agreement with the observed spectra. Our direct analysis suggests the presence of enhanced nitrogen. We have further studied these spectra with the full NLTE general model atmosphere code \\texttt{PHOENIX}. While we do not find confirmation for enhanced nitrogen (nor do we rule it out), we do require enhanced helium. An even more intriguing possible line identification is complicated Balmer and He~I lines, which we show falls naturally out of the detailed calculations with a shallow density gradient. We also show that very early spectra such as those presented here combined with sophisticated spectral modeling allows an independent estimate of the total reddening to the supernova, since when the spectrum is very blue, dereddening leads to changes in the blue flux that cannot be reproduced by altering the ``temperature'' of the emitted radiation. These results are extremely encouraging since they imply that detailed modeling of early spectra can shed light on both the abundances and total extinction of SNe II, the latter improving their utility and reliability as distance indicators. ",
        "introduction": "SN 1999em was discovered by the Lick Observatory Supernova Search (IAUC 7294) on Oct 29.44 UT and confirmed by the BAO supernova group on Oct 29.7 UT. It was followed spectroscopically and photometrically by ground-based observatories and was the subject of a Supernova Intensive Studies (SINS) \\emph{Hubble Space Telescope} (\\emph{HST}) observation on Nov 5, 1999. SN 1999em is in the SABc galaxy NGC 1637, with a heliocentric recession velocity of 717 \\kmps. The first spectrum taken on Oct 29, 1999 is extremely blue and displays strong Balmer features typical of Type II supernovae with a characteristic velocity of around 10,000 \\kmps.  SN 1999em appears to have been a normal Type IIP, likely the result of a core-collapse in a star with a massive hydrogen envelope \\citep{leon99em00}. The early and weak radio emission (IAU Circulars 7318 and 7336) hints that the star did not experience significant mass loss shortly before the explosion, which is also thought to be the normal situation for SNe IIP \\citep[see][and references therein]{weilermd00}. The lack of circumstellar interaction makes SNe IIP ``clean'' cases to model, resulting in more reliable atmospheric modeling which in turn makes SNe IIP ideal for cosmological use. Because SN 1999em was observed earlier, and over a wider wavelength range than any other normal bright SN IIP, it is a perfect test bed for our understanding of SNe IIP.  These supernovae are of particular interest as distance indicators. ",
        "conclusions": "We have shown that both direct synthetic spectral fits and detailed NLTE models do a good job of reproducing the observed optical+UV spectra of SN~1999em. We have shown that detailed NLTE spectral modeling of very early spectra of Type II supernova can provide an independent estimate to the total reddening as well as abundance information. Specifically we find that $E(B-V) \\approx 0.05-0.10$ and $E(B-V) \\la 0.15$~mag, and we find strong evidence for helium enhanced by at least a factor of 2 over the solar value. We obtain the striking result that observed features are likely due to ``complicated P-Cygni'' lines of hydrogen and helium, and thus one needs to be especially careful with line identifications in hot differentially expanding flows. That the observed complicated profiles fall right out of the \\texttt{PHOENIX} calculations is remarkable and lends strong support to the reliablity of the modeling.  The fact that planned supernova search programs will preferentially find supernovae very early, combined with our results that the extinction can be determined by modeling the very early spectra, lends promise to the use of Type II supernovae as distance indicators, through the use of phenomenological applications like the ``Expanding Photosphere Method'' \\citep{baadeepm,kkepm,branepm,schmkireas92,esk96} or through sophisticated modeling of the ``Spectral-fitting Expanding Atmosphere Method'' \\citep{b93j1,b93j2,b93j3,b93j4,b94i1}."
    },
    "0010/hep-ph0010321_arXiv.txt": {
        "abstract": "We evaluate several quantities appearing in the effective lagrangian for the color-flavor locked phase of high density QCD using a formalism which exploits the approximate decoupling of fermions with energy negative with respect to the Fermi energy. The effective theory is essentially two-dimensional and exhibits a Fermi velocity superselection rule, similar to the one found in the  Heavy Quark Effective Theory. Within the formalism we reproduce, using gradient expansion, the results for the effective parameters of the Nambu-Goldstone bosons. We also determine the dispersion laws for the gluons. By coupling the theory to fermions and integrating over the two-dimensional degrees of freedom we obtain the effective description of in-medium fermions.  \\noindent ",
        "introduction": "QCD at high density is expected to be in a chirality breaking color superconducting phase due to formation of quark Cooper pairs \\cite{a},\\cite{alford}. In QCD with three massless quarks one expects simultaneous breaking of color and flavor simmetries into a residual diagonal color-flavor locked symmetry (CFL). The physical description and consequences have been discussed by a number of authors \\cite{b,casalbuoni,rho,son}. A formalism has been recently proposed \\cite{hong} and \\cite{beane} to deal with the color superconducting phase. It takes advantage of the approximate decoupling of the states of a massless fermion of energy distant with respect to the Fermi surface. We shall first summarize the main consequences of such a description, with emphasis on its implied velocity superselection rule, similar to that familiar in the Heavy Quark Effective Theory (HQET) \\cite{HQET}, and on its essentially two-dimensional features, due to the special relevance of momenta parallel to the Fermi velocity. We shall then calculate in this formalism the effective parameters of the goldstones, reobtaining results derived by other methods. Next we shall calculate the dispersion laws for the gluons and discuss the various masses that can be defined for the scalar, longitudinal and transverse gluons, i.e. their rest mass,  the effective mass and the inverse of the penetration length. After having coupled the theory to fermions we obtain the effective description of in-medium fermions, by integrating over the two-dimensional degrees of freedom.  In section 2 we discuss the effective theory on the Fermi surface. In section 3 we write down the effective lagrangian for the fermions and the goldstones. In section 4 we perform the gradient expansion on the generating functional. We discuss the gluon polarization tensor in section 5. In section 6 we treat the effective description of in-medium fermions. ",
        "conclusions": "After a presentation of the basic points of the formalism, we have discussed the couplings of the goldstones and their effective lagrangian. We have written down explicitly the needed trilinear couplings of the goldstones to the fermions and the quadrilinear couplings of two goldstones to two fermions. To explicitly obtain the parameters of the effective action for the goldstones we have performed a derivative expansion of the generating functional. In this way we obtain the known results at one loop order. The value of the velocity of the goldstones appears explicitly as following from the integration over the Fermi velocity. We have then discussed the  in-medium gluon polarization tensor. At the one loop order the formalism reproduces the correct values for the Debye and Meissner masses. The dispersion laws for both scalar and vector gluons follow from the gluon field equations of motion in the limit of high density; from them the various masses of the scalar, longitudinal and transverse gluons have been computed, i.e. the rest mass, the  effective masses and the inverse of the penetration lenght. A main development has been the derivation of the effective description for in-medium fermions. We have introduced fermion external fields, both left- and right-handed, to be coupled to the dynamical fermion fields. After integrating over the dynamical fermions we have obtained an effective action and the fermionic dispersion relation in terms of the gap parameter. This provides for a value of the in-medium fermion mass and shows again the role played by the integration over the Fermi velocity.  The theory of color-flavor locking, as developed and studied by many authors, has undoubtedly an intrinsic beauty and the high density fermionic formalism used here provides for a simple and appealing description. It remains to be seen to which physical situations of matter under extreme conditions the theory can usefully be applied, most probably by including additional perturbations. \\vskip.3cm\\noindent {\\bf{AKNOWLEDGMENTS}} We would like to thank Krishna Rajagopal for an illuminating discussion about a previous version of this paper."
    },
    "0010/nucl-th0010050_arXiv.txt": {
        "abstract": "We show that nuclear states lying at the edge of stability may show enormously enhanced response to small perturbations. For example, a 0.1\\% change in the strength of the strong nucleon-nucleon interaction can cause almost a hundred times bigger change in the resonance energy of the $0^+_2$ state of $^{12}$C. ",
        "introduction": "Recently, the birth of radioactive nuclear beams has made it possible to perform extensive studies of the structure and reactions of nuclei lying at the edge of stability. It was found that these nuclei posses some really unusual properties, such as, for example, halo structure \\cite{Hansen}, which were largely unknown in stable systems. Here we would like to show that some nuclear systems close to the point of stability may exhibit a hitherto unrecognized interesting phenomenon, which we call nonlinear quantum amplifying. As the system moves toward the edge of stability, its response to small perturbations can become hugely amplified. This effect may find interesting applications in halo nuclei or in astrophysical processes.  We consider nuclei which have strong 2-body or 3-body clustering nature. This means that their wave functions contain 2- or 3-cluster structures with large weight, therefore the most important degrees of freedom are the relative motions between the clusters. For instance, the low-lying states of $^7$Li and $^{12}$C are known to have strong $^4{\\rm He}+{^3{\\rm H}}$ and $^4{\\rm He}+{^4{\\rm He}}+{^4{\\rm He}}$ cluster structures, respectively. In addition, we assume that these systems are below or slightly above the threshold of the lowest possible breakup channel. If the energy of such a state, relative to the 2-body or 3-body breakup threshold, is close to zero then one may call it a weakly coupled few-body system. It is either a bound state or a long-lived resonance. In any case, the system is barely held together by the residual interactions acting between the clusters.  It is interesting to see how the residual interaction between the clusters behaves as the system goes from a weakly-bound state through the breakup point to a resonance state. While this happens, the size of the system increases. If the interaction between the clusters is very short-ranged, then one can imagine that the residual interaction drops down to zero as fast as the binding energy itself. Then nothing special happens. However, one can imagine situations where the residual force only slowly decreases but does not vanish, while the binding energy becomes zero. If this happens, then the binding energy becomes very sensitive to small perturbations, for example, in the nucleon-nucleon (N-N) force.  In order to see if such situations are really realized, we study three nuclear systems: the deuteron, the ground state of $^7$Li, and the $0^+_2$ state of $^{12}$C. ",
        "conclusions": "We have shown that certain nuclear states lying at the edge of stability may behave as nonlinear amplifiers: a tiny change in the N-N interaction can get enormously amplified in the binding energy or resonance energy. This behavior is mainly caused by the fact that the residual interaction between the 2-3 clusters of the nucleus goes down to zero more mildly than the binding energy itself.  We can envision several possible applications of the nonlinear amplification process discussed here. One application was exemplified through the $0^+_2$ state of $^{12}$C. Because of the amplification phenomenon, some astrophysical processes can be very strongly fine-tuned. Halo nuclei are also natural candidates to show this phenomenon, as the halo effect itself is strongly connected to the disappearing binding energy.  In summary, we have presented an interesting property of some nuclear systems lying at the edge of stability. Whether one can find any useful applications of this feature, beyond the one example of the $0^+_2$ state of $^{12}$C, remains to be seen."
    },
    "0010/hep-ph0010103_arXiv.txt": {
        "abstract": "The phenomenological possibilities of the Randall-Sundrum non-compact extra dimension scenario with the AdS horizon increased to approximately a millimeter length, corresponding to an effective brane tension of $(\\rm TeV)^4$, are investigated. The corrections to the Newtonian potential are found to be the only observationally accessible probe of this scenario, as previously suggested in the literature.  In particular, the presence of the continuum of KK modes does not lead to any observable collider signatures.  The extent to which experimental tests of Newtonian gravity can distinguish this scenario from the scenario of Arkani-Hamed, Dimopoulos, and Dvali with one and two millimeter size extra dimensions is explicitly demonstrated. ",
        "introduction": "To fix our notation, let us review the construction of the four-dimensional effective action presented in Ref.~\\cite{RSII}. Consider the perturbed metric \\begin{equation} ds^{2}=f(z)\\left[\\eta_{\\mu \\nu } +h_{\\mu \\nu }(x)\\right]dx^{\\mu }dx^{\\nu }-dz^{2}, \\end{equation} where the transverse dimension is parameterized by \\( z \\) (the Planck brane is located at $z=0$), and the warp factor is \\begin{equation} f(z)=e^{-2k|z|}. \\end{equation} We then find the effective Lagrangian for the metric perturbation to be \\begin{equation} \\Delta [\\sqrt{g}(R+2\\Lambda )]=-\\frac{f}{2}\\left[\\frac{-1}{2}h^{\\mu \\nu }h_{\\mu \\nu \\, \\, \\, ,\\alpha }^{\\, \\, \\, ,\\alpha }+h^{\\mu \\nu }h^{\\lambda }_{\\, \\, \\nu ,\\lambda \\mu }-h^{\\mu \\nu }h_{,\\mu \\nu }+\\frac{1}{2}hh_{,\\mu }^{\\, \\, \\, ,\\mu }\\right] + \\frac{f^2}{4}\\left[ h^{\\mu \\nu}_{\\, \\, \\, ,4} h_{\\mu \\nu ,4} - h_{,4}^2 \\right] \\end{equation} where $R$ is the five-dimensional Ricci scalar, $\\Lambda$ is the bulk cosmological constant, and the indices (which are contracted with the Minkowski metric $\\eta_{\\mu\\nu}$) run over $0,1,2,3$, the coordinates parallel to the Planck brane.  In the transverse, traceless gauge, we simply have \\begin{equation} \\Delta [\\sqrt{g}(R+2\\Lambda )]=\\frac{-f}{4}h^{\\mu \\nu }h_{\\mu \\nu \\, \\, \\, \\, \\, ,\\alpha }^{\\, \\, \\, \\, ,\\alpha }\\end{equation} Hence, the graviton action $S_G$ coupled minimally to the matter action \\( S_{M} \\) is given by \\begin{equation} S=S_{G}+S_{M},\\end{equation} where \\begin{equation} S_{G}=\\frac{M^{3}}{4}\\int d^{5}xfh^{\\mu \\nu }h_{\\mu \\nu \\, \\, \\, \\, \\, , \\alpha }^{\\, \\, \\, \\, ,\\alpha }.\\end{equation} Let us decompose the graviton in terms of 4-dimensional modes: \\begin{equation} h_{\\mu \\nu }=\\xi \\sqrt{\\frac{\\pi }{L}} \\sum_n H^{(n)}_{\\mu \\nu }\\chi ^{(n)},\\end{equation} in which \\begin{equation} \\partial _{z}(f^{2}\\partial _{z}\\chi ^{(n)})+m_{n}^{2}f(z)\\chi ^{(n)}=0\\end{equation} \\begin{equation} \\partial _{z}\\chi ^{(n)}(0)=\\partial _{z}\\chi ^{(n)}(L)=0\\end{equation} with the normalization \\begin{equation} \\int ^{L}_{-L}dzf(z)\\chi ^{(n)}(z)\\chi ^{(m)}(z)=\\frac{L}{\\pi }\\delta_{nm}. \\end{equation} Note that $\\xi$ is an arbitrary normalization parameter.  Hence, we find for the four dimensional effective action of gravity \\begin{equation} S_{G}=\\frac{\\xi^2 M^{3}}{4}\\int d^{4}x\\{H^{\\mu \\nu }_{(n)} H_{\\mu \\nu ,\\alpha }^{(n)\\, \\, \\, \\, ,\\alpha } +m_{n}^{2}H^{\\mu \\nu }_{(n)}H_{\\mu \\nu }^{(n)}\\}.\\end{equation} We can also expand the matter action to give \\begin{eqnarray} S_{M} & = & S_{M}^{(0)}+\\int d^{4}x\\frac{\\delta S}{\\delta g_{\\mu \\nu }(x)}h_{\\mu \\nu }(x,z=0)+\\ldots\\\\ & = & S_{M}^{(0)}-\\sum_n \\int d^{4}x\\frac{\\sqrt{g(z=0)}}{2}T^{\\mu \\nu }_{brane}H_{\\mu \\nu }^{(n)}\\chi ^{(n)}(z=0)\\xi \\sqrt{\\frac{\\pi }{L}}+\\ldots \\end{eqnarray} where the $\\ldots$ include comparable coupling terms induced from the fact that the brane is bent in the presence of matter in the transverse traceless gauge. The effective action is given by \\begin{equation} \\delta S=\\frac{\\xi^2 M^3}{4} \\int d^{4}x\\{H^{\\mu \\nu }_{(n)}H_{\\mu \\nu ,\\alpha }^{(n)\\, \\, \\, \\, ,\\alpha }+m_{n}^{2}H^{\\mu \\nu }_{(n)}H_{\\mu \\nu }^{(n)}\\}-\\frac{1}{2}\\int d^{4}xT^{\\mu \\nu }_{brane}H_{\\mu \\nu }^{(n)}\\chi ^{(n)}(z=0)\\xi \\sqrt{\\frac{\\pi }{L}} + \\ldots. \\end{equation} To see the Planck scale suppression of the couplings, it is convenient to normalize the kinetic term for the gravitons such that $H_{\\mu \\nu}$ has dimension 1 and the propagator resembles a scalar propagator.  Hence, we let \\begin{equation} \\xi^2=\\frac{4}{M^3} \\end{equation} Now, we have for \\( z>0 \\), \\begin{equation} \\chi ^{(n)}(z)=\\frac{1}{N_{n}}[J_{2}(a_{n}e^{kz})+\\alpha _{n}Y_{2}(a_{n}e^{kz})]e^{2kz}\\end{equation} with \\begin{equation} \\alpha _{n}=\\frac{-J_{1}(a_{n})}{Y_{1}(a_{n})}\\end{equation} where \\begin{equation} a_{n}=\\frac{m_{n}}{k},\\end{equation} and the mass condition is given by \\begin{equation} \\frac{J_{1}(a_{n}e^{kL})}{J_{1}(a_{n})}=\\frac{Y_{1}(a_{n}e^{kL})}{Y_{1}(a_{n})}.\\end{equation} To get the normalization constant \\( N_{n} \\), we use \\begin{equation} \\frac{2}{N^{2}_{n}}\\int ^{L}_{0}dz [ J_{2}(a_{n}e^{kz})+\\alpha _{n}Y_{2}(a_{n}e^{kz})]^{2}e^{2kz}=\\frac{L}{\\pi}.\\end{equation} Writing \\begin{equation} Z_{n}(x)\\equiv J_{n}(x)+\\alpha Y_{n}(x),\\end{equation} for any constant $\\alpha$ and real argument $x$, we can use the Bessel function identity \\begin{equation} \\int dxxZ^{2}_{n}(ax)=\\frac{x^{2}}{2}[Z^{2}_{n}(ax)-Z_{n+1}(ax)Z_{n-1}(ax)]\\end{equation} for any integer $n$ and the mass shell condition to get \\begin{equation} N^{2}_{n}=\\frac{\\pi}{kL}\\left[e^{2kL}Z^{2}_{2}(a_{n}e^{kL})-Z^{2}_{2}(a_{n})\\right].\\end{equation} Therefore, the factor \\( \\chi ^{(n)} \\) which determines the matter coupling to gravity is given by \\begin{eqnarray} \\chi ^{(n)}(0) & = & \\sqrt{\\frac{kL}{\\pi }}\\frac{1}{\\sqrt{\\left[ \\frac{Y_{1}(a_{n})}{Y_{1}(a_{n}e^{kL})}\\right] ^{2}-1}} \\label{eq:wavefn} \\end{eqnarray} in which we used the Bessel function identity \\begin{equation} J_{\\nu }(x)Y_{\\nu +1}(x)-J_{\\nu +1}(x)Y_{\\nu }(x)=-\\frac{2}{\\pi x},\\end{equation} and the definition of \\( \\alpha _{n} \\). The coupling of the gravitational modes to matter is then given by \\begin{equation} \\delta S_{M} = \\int d^{4}xT^{\\mu \\nu }g_{c}^{(n)}H_{\\mu \\nu }^{(n)}, \\end{equation} where we have defined the coupling \\begin{equation} g^{(n)}_{c}\\equiv \\frac{-\\xi}{2} \\sqrt{\\frac{\\pi}{L}} \\chi^{(n)}(z=0)=\\frac{-1}{M^{3/2}}\\sqrt{\\frac{\\pi}{L}} \\chi^{(n)}(z=0). \\end{equation} ",
        "conclusions": ""
    },
    "0010/astro-ph0010600_arXiv.txt": {
        "abstract": "A new model for the, so called, {\\it cluster radio relics} is presented (En{\\ss}lin \\& Gopal-Krishna 2000). Fossil radio cocoons, resulting from the former activity of radio galaxies, should contain a low energy relativistic electron population and magnetic fields. Electrons with an age of even up to 2 Gyr can be re-accelerated adiabatically to radio emitting energies, if the fossil radio plasma gets compressed in an environmental shock wave. Such a wave can be caused by a merging event of galaxy clusters, or by the accretion onto galaxy clusters.  An implication of this model is the existence of a population of diffuse, ultra-steep spectrum, very low-frequency radio sources located inside and possibly outside of clusters of galaxies, tracing the revival of aged fossil radio plasma by the shock waves associated with large-scale structure formation. ",
        "introduction": "\\vspace{-0.5em} The radio cocoons of radio galaxies become rapidly undetectable after the central engine of the active galactic nucleus ceases to inject fresh radio plasma.  Although undetectable, the aged radio plasma should still be an important component of the inter-galactic medium (IGM). We investigate the possibility of reviving patches of such fossil radio plasma (also called {\\it radio ghosts}, En{\\ss}lin (1999)) by compression in a shock wave produced in the IGM by the flows associated with cosmological large-scale structure formation. Such shock waves re-energize the electron population in the fossil radio cocoons, which can lead to observable synchrotron emission. This is probably the mechanism responsible for the, so called, {\\it cluster radio relics}, which are patches of diffuse, sometimes polarized radio emission typically found at peripheral locations in clusters of galaxies. These cluster radio relics can not be simply relic radio galaxies, as their name suggests.  The spectral ages of the electron population are {usually} too short to admit even the nearest galaxy to have been the parent radio galaxy, which has moved to its present location with a velocity typical for cluster galaxies. Therefore, a recent enhancement of the nonthermal radio output of the cluster relic sources is mandatory.   The connection between the presence of a shock wave and the appearance of the cluster radio relic phenomenon was assumed to be due to Fermi-I shock acceleration of electrons, in En{\\ss}lin et al. (1998) and Roettiger et al. (1999). While this process might work within the normal IGM, several arguments favor the possibility that indeed old fossil radio plasma is revived in the case of cluster relic sources: \\begin{itemize} \\setlength{\\topsep}{0em} \\setlength{\\partopsep}{0em} \\setlength{\\parskip}{0em} \\setlength{\\parsep}{0em} \\setlength{\\itemsep}{0em} \\setlength{\\itemindent}{0em} \\setlength{\\leftmargin}{0em} \\setlength{\\leftmargin}{0em} \\item Cluster radio relics are extremely rare, whereas shock waves should be very common within clusters of galaxies. The {dual} requirement of a shock wave and fossil radio plasma, {for producing} a cluster radio relic, {would} be an attractive explanation for the rareness of the relics. \\item Fossil radio plasma with existing relativistic electron population and fairly strong magnetic field appears to have ideal properties for being brightened up during the shock's passage. \\item The radio relic 1253+375 near the Coma cluster of galaxies appears to be fed with radio plasma by the nearby galaxy NGC 4789 (En{\\ss}lin et al. 1998). \\end{itemize} But, if indeed the fossil radio plasma and not the normal IGM were to become radio luminous at a shock wave, the expected very high sound velocity of that relativistic plasma should forbid the shock in the ambient medium from penetrating into the radio plasma. Thus, shock acceleration is not expected to occur there. Instead, the fossil radio plasma would get adiabatically compressed, and the energy gain of the electrons is expected to be mainly due to adiabatic heating. It is the purpose of this work to demonstrate that this process is sufficient to account for the cluster radio relics. \\vspace{-1em} ",
        "conclusions": "} \\vspace{-0.5em} We have argued here that adiabatic compression in cluster shocks can revive fossil radio plasma to radio detection, even up to 2 Gyr after the cessation of the nuclear activity and thus explain the observed cluster radio relics.  Below, we summarize some merits of this model (also, see Sect 1.): \\begin{itemize} \\setlength{\\topsep}{0em} \\setlength{\\partopsep}{0em} \\setlength{\\parskip}{0em} \\setlength{\\parsep}{0em} \\setlength{\\itemsep}{0em} \\setlength{\\itemindent}{0em} \\setlength{\\leftmargin}{0em} \\setlength{\\leftmargin}{0em} \\item The observed connection of cluster radio relics to shock waves arises naturally in this model. \\item The presence of the radio galaxy NGC 4789 and the morphological connection of its radio tails to the relic 1253+275 seems to be the {\\it smoking gun} of cluster radio relic formation by compression of fossil radio plasma. \\item The expected very high sound speed within the radio plasma should virtually forbid shock waves of the ambient medium to penetrate into radio cocoons. This renders adiabatic compression as a more plausible means of reviving the fossil plasma. \\item Cluster radio relics are rare. Not only a shock wave and fossil radio plasma are both required in this model, in order to produce a cluster radio relic, but also the fossil radio plasma cannot grow too old (of the order of 0.1 Gyr in the center of clusters and 1 Gyr at their boundaries). Otherwise, the depletion of the high energy end of the electron population would be too severe for the adiabatic gains during the compression phase to be able to shift them to radio emitting energies. Since radio fossils should be very common (En{\\ss}lin \\& Kaiser 2000), these requirements help to explain the observed rarity of cluster radio relics. \\item The observed tendency of relics to appear at peripheral locations in clusters follows in this model from the much shorter synchrotron lifetimes of the radio emission and, consequently, much shorter ages which the revive-able fossil radio cocoons can have near the centers. \\end{itemize} We conclude that the model presented here provides a fairly natural and promising explanation for the phenomenon of cluster radio relics. It predicts the existence of a population of diffuse, ultra-steep, very low frequency radio sources inside and possibly also outside of clusters of galaxies, due to the age dependence of the upper frequency cutoff of the radio emission arising from the revivable fossil radio plasma. The ongoing rapid improvements in the sensitivities at low radio frequencies (Giant Meterwave Radio Telescope: (GMRT), Low Frequency Array (LOFAR)) may therefore open a new window on cosmological structure formation by detecting shock waves marked by such relic radio sources.   \\def\\nat{Nature} \\def\\aj{AJ}"
    },
    "0010/astro-ph0010436_arXiv.txt": {
        "abstract": "The dynamical evolution of galaxies in clusters is modified if dark matter is self-interacting.  Heat conduction from the hot cluster halo leads to evaporation of the relatively cooler galactic halos.  The stellar distribution would adiabatically expand as it readjusts to the loss of dark matter, reducing the velocity dispersion and increasing the half-light radius.  If the dark matter content within that radius was $f_{dm} = 25-50\\%$ of the total, as indicated by current observations, the ellipticals in clusters would be offset from the fundamental plane relation beyond the observational scatter.  The requirement that their halos survive for a Hubble time appears to exclude just that range of the dark matter cross-section, $0.3 \\lesssim \\sigma_p/m_p \\lesssim 10^4$ cm$^2$ g$^{-1}$, thought to be optimal for reducing central halo cusps, unless $f_{dm} < 15\\%$.  If the cross-section is allowed to vary with the relative velocity of dark matter particles, $\\sigma \\propto v^{-2\\delta}$, a new problem of evaporation of dark matter arises in the dwarf galaxies with low velocity dispersion.  The halos of large galaxies in clusters and dwarf galaxies in the Local Group can both survive only if $\\delta < 1.1$ or $\\delta > 1.8$.  In either case the problem of central density cusps remains. ",
        "introduction": "Self-interacting dark matter has been proposed recently to alleviate some of the problems of galaxy formation arising in the standard cold dark matter models \\cite{SS:00}, \\cite{Wetal:00}.  By allowing for heat conduction into the cooler galactic center, the collisional dark matter model reduces the density cusp, which is predicted by numerical simulations in the CDM model \\cite{Moore:99}, and seems to be in conflict with observations.  A drawback of such a scenario is that the desired heat conduction takes place in all dark matter-dominated systems, producing undesired phenomena in some situations.  In clusters of galaxies, the hot cluster halo would tend to evaporate the relatively cooler galactic halos, giving strict limits on the allowed range of the self-interaction cross-section.  Depending on the value of the cross-section per particle mass, $\\sigma_p/m_p$, collisional dark matter can act either as a fluid or as a rarefied gas.  The mean free path of dark matter particles is $\\lambda \\approx m_p/(\\rho \\sigma_p)$, where $\\rho$ is the dark matter density. In the fluid regime, $\\lambda(r) \\ll r$, the heat diffuses out of the hot region (cluster halo) to the cool region (galaxy halo) on the conduction time scale.  In the opposite optically-thin regime, $r < \\lambda$, the interaction is an elastic scattering of the particles.  Numerical simulations in both fluid \\cite{Metal:00}; \\cite{Yetal:00} and scattering regimes \\cite{B:00}; \\cite{KW:00} have shown that in isolated halos the central cusp is indeed replaced by a soft core, as desired, for a suitable range of cross-section.  However, a large range of cross-section appears to be already ruled out by various observations. The experimental limits and the required rates of core evolution are estimated to narrow the currently allowed values to the range $0.1 < \\sigma_p/m_p < 6$ cm$^2$ g$^{-1}$ \\cite{Wetal:00}.  Theoretical arguments seem to be most powerful for the clusters of galaxies.  Self-interaction tends to make clusters more spherical than follows from the lensing observations.  For example, \\cite{ME:00} estimated that this constrains the cross-section to be $\\sigma_p/m_p < 0.02$ cm$^2$ g$^{-1}$.  However, there are serious uncertainties associated with this limit due to statistical and other issues.  In this paper we consider the effect of thermal evaporation of dark matter halos in clusters.  Significant evaporation of dark matter halos of the elliptical galaxies would introduce a systematic offset in their fundamental plane relations, compared to the field ellipticals.  In \\S\\ref{sec:fluid} we calculate the dynamics of spherically-symmetric galaxy halos in the fluid regime solving the diffusion equation.  In \\S\\ref{sec:scat} we estimate the evaporation time analytically in the scattering regime.  In \\S\\ref{sec:tempdep} we consider the case of a temperature-dependent cross-section that changes the dynamics of the inner parts of the halo.  Finally, in \\S\\ref{sec:ell} we show that the evaporation of dark matter halos is probably inconsistent with observational scatter of the fundamental plane relation. ",
        "conclusions": "We have considered astrophysical constraints on the possible cross-section of self-interacting dark matter.  The fundamental plane relation of elliptical galaxies requires that their dark halos largely survive the heating from the hot cluster halos.  This condition applies to galaxies with the half-light dark matter fraction $f_{dm} > 0.15$. For a velocity-independent cross-section, the evaporation time of such halos is shorter than the Hubble time if $0.3 \\lesssim \\sigma_p/m_p \\lesssim 10^4$ cm$^2$ g$^{-1}$.  The parameter space can be enlarged by assuming $\\sigma_p/m_p = 0.1$ cm$^2$ g$^{-1}$ and allowing for a velocity dependence of the cross-section.  The large halos would be unaffected, as long as they remain in the optically-thin regime, but the dwarf halos within larger halos would evaporate if the optical depth rises towards unity.  The presence of dwarf spheroidals in the Local Group excludes the range of the power-law slope $1.1 < \\delta < 1.8$.  Combined with other constraints in the literature \\cite{Wetal:00}, this appears to exclude the most interesting range of the cross-section for the purpose of lowering the core densities of galactic halos.  In order to reduce the central density cusp by scattering, the optical depth at the scale radius of the halo $R_{g,0}$ should be close to unity \\cite{SS:00}.  For a Hernquist model of the halo, $\\tau(R_{g,0})$ exceeds the optical depth at the tidal radius, $\\tau_t$, by a factor $(1+c)^2/4 \\sim 10 - 30$.  We can therefore expect the optimum regime to be $\\tau_t \\sim 0.03 - 0.1$, but as Figure \\ref{fig:cross} demonstrates, this regime would lead to evaporation of the halo in less than a Hubble time.  Thus, the original motivation for considering self-interacting dark matter may require collisional cross-sections too large to be consistent with observations of galaxies in clusters.  However, the cross-section in the optically-thin regime ($\\sigma_p/m_p \\sim 10^{-1}$ cm$^2$ g$^{-1}$) may still have very interesting consequences for the growth of central black holes \\cite{O:00}."
    },
    "0010/astro-ph0010093_arXiv.txt": {
        "abstract": "A comparison is presented between the observed surface density of extremely red objects (EROs) and the densities of red high-$z$ elliptical galaxies predicted by Pure Luminosity Evolution (PLE) and by some CDM hierarchical models. The analysis is based on  the widest field survey for EROs available to date (covering in total $\\sim$850 arcmin$^2$), and it takes into account the effects of ERO clustering. Good agreement is found between the observed surface densities of EROs and those predicted by PLE models for ellipticals selected using the same color and luminosity thresholds. Since there is evidence that the bulk of EROs are passively evolving ellipticals, this result implies that most field ellipticals were fully assembled at least by $z\\sim 2.5$. Existing hierarchical models predict instead a surface density of red ellipticals that is much smaller than that of the EROs. ",
        "introduction": "The formation of elliptical galaxies remains one of the most controversial issues of galaxy evolution and structure formation. While there is now wide consensus that the bulk of stars in ellipticals formed at very high redshifts (see e.g. Renzini 1998 and the introduction of Schade et al. 1999 for recent reviews), opinions diverge as to the epoch at which ellipticals have been assembled to their present size and mass. In semianalytical simulations of hierarchical Cold Dark Matter (CDM) cosmologies (e.g. Kauffmann 1996; Baugh, Cole \\& Frenk 1996; Baugh et al. 1998) massive ellipticals form at relatively low redshift ($z<1$) through the merging of spiral galaxies.  Empirical support for this scenario has been claimed from an alleged deficit of old ellipticals at $z\\sim 1$ compared to pure luminosity evolution (PLE) models (Kauffmann, Charlot \\& White 1996, Zepf 1997, Franceschini et al. 1998, Barger et al. 1999, Menanteau et al. 1999, Treu \\& Stiavelli 1999). However, others found evidence for a constant comoving density of ellipticals up to $z\\sim 2$ (Totani \\& Yoshii 1997, Benitez et al. 1999, Broadhurst \\& Bowens 2000, Schade et al. 1999; Scodeggio \\& Silva 2000), which would instead favor an early assembly of ellipticals, no matter whether in clusters or in the field. One obvious way to solve the issue is to search for putative high-$z$ ellipticals over much larger fields than in previous studies. Passively evolving ellipticals in the redshift range $1\\simlt z\\simlt 2$ are in fact characterized by very red colors ($R-K_{\\rm s} \\gsim 5-6$), which qualify them as extremely red objects (ERO).  In a recent study based on the largest survey for EROs to date, Daddi et al. (2000) reported the discovery of strong EROs angular clustering. This finding suggest that the origin of such discrepant results were definitively in the small fields covered by previous studies (ranging from $\\sim 1$ to $\\sim60$ arcmin$^2$), which are likely to be affected by field-to-field variations in the number of intercepted large scale structures, hence in the number of red, color selected galaxies. Moreover, the strong clustering argues for the bulk of EROs being made by elliptical (Daddi et al. 2000) rather than starburst galaxies, that can have very red $R-K_{\\rm s}$ colors as well (e.g. Cimatti et al. 1998). This agrees with independent indications that ellipticals are at least a fraction of $\\sim70$\\% among EROs (Moriondo et al. 2000, Stiavelli et al. 2000 private comunication, Cimatti et al. 1999). Under this assumption, the surface density of EROs can therefore set strong constraints to models for the formation and evolution of elliptical galaxies.  In this Letter we present the results of the comparison between the observed surface density of EROs and the densities predicted by PLE and hierarchical models for the evolution of the elliptical galaxy population. Such a comparison is based on the results of our wide-field survey for EROs (Daddi et al. 2000) and on the Thompson et al. (1999) survey, covering a total area of about 850 arcmin$^2$ resulting in the selection of a total sample of several hundreds of EROs. In addition, the measurements of the angular clustering allow us to quantify for the first time the uncertainties on the surface density of EROs due to cosmic variance. For these reasons, our study represents a significant improvement respect to previous attempts. For example, the widest field analysis most recently done was based only on 16 EROs selected from an area of 60 arcmin$^2$ (Barger et al. 1999). ",
        "conclusions": "A robust surface density of EROs has been obtained from a field that is at least ten times larger than in previous determinations, thus greatly reducing the influence of the strong field-to-field variations (cosmic variance), that we estimate given the known clustering properties of EROs (Daddi et al. 2000).  This observed surface density of EROs is in fine agreement with the density of $z>1$ ellipticals with the same color and luminosity thresholds as predicted by a PLE model in which the bulk of field ellipticals were already fully assembled at least by $z = 2.5$, and evolved passively thereafter. The agreement with the predictions of the PLE models remains good even if some 30\\% of EROs are not elliptical galaxies (e.g. dusty starburst galaxies). This result argues for much of the merging being already completed by $z\\gsim 2$, in such a way to make elliptical galaxy formation mimicking the old monolithic collapse models.  A more thorough comparison with current realizations of hierarchical models of galaxy formation is hampered by the lack of published predictions for the surface density of EROs. In these models the number density of massive elliptical galaxies drops rapidly with redshift. It does so catastrophically in a $\\tau$CDM model (Kauffmann \\& Charlot 1998b), which therefore can be readily excluded, unless only a very small fraction of EROs is made of passively evolving ellipticals. However, this would be in contradiction with the reported strong clustering properties of EROs (Daddi et al. 2000). Less dramatic is the comparison with a $\\Lambda$CDM model, in which the number of ellipticals drops more gently with redshift. Nevertheless, we argue that also this model can hardly account for the observed number of EROs unless merging does {\\it not} result in significant star formation, which is not the case in existing realizations. Future semianalytical models including specific predictions concerning the number densities of EROs would offer an interesting opportunity, as a comparison with the observed densities is expected to provide a stringent test for such models."
    },
    "0010/astro-ph0010570_arXiv.txt": {
        "abstract": "s{ Active Galactic Nuclei are fueled from material (gas or stars) that are in general far away from the gravitational influence of the central black hole, the engine thought to be responsible for their activity. The required material has a lot of angular momentum, that is a priori quite difficult to evacuate. The various dynamical mechanisms that may play a role in this game are reviewed, including $m=2$ perturbations (bars and spirals),   $m=1$ perturbations (spirals, warps, lopsidedness), and tidal interactions between galaxies and mergers. In the latest stages of the merger, a binary black hole could be formed, and its influence on the dynamics and fueling is discussed. Starbursts are often associated with AGN, and the nature of their particular connection, and their role in the nuclear fueling is described. Evolution of the fueling efficiency with redshift is addressed. } ",
        "introduction": "Some galaxies are dubbed ``active'', which means that they radiate much more energy than can be sustained by a normal evolution rate, or average  gas consumption rate, given their size and gas reservoir mass. Their activity could be due to a starburst, or to a material-accreting compact nucleus. Both are often related or simultaneous.  The activity has a short duration with respect to the galaxy life-time, but can be recurrent. A key issue is to understand the mechanisms that trigger the activity, and how a large amount of material can be funneled to the central regions, to fuel this activity.  A first possibility is provided by the dense star clusters that are present near the nucleus. These nearby stars can provide gas fuel to the nucleus through their mass loss rates, or even the tidal disruption of stars can liberate their whole gas mass. We will present and discuss the various physical phenomena involved, and their respective time-scales in section 2: only those stars that have low angular momentum orbits are available to fuel the activity, but these can be replenished through dynamical diffusion when depleted. A bigger problem arises when the nearby stars are not numerous enough to sustain the activity: mass has then to come from the whole galaxy in a short time-scale, and this rises the problem of angular momentum transfer, that must involve internal or externally-trigerred gravitational instabilities. Note that the mass fueling can occur in two steps: a first instability drives the gas towards the center, giving rise to a nuclear starburst and the formation of dense nuclear star clusters. Then, the active nucleus can be fueled by the evolution of the dense star cluster.  The gravitational instabilities of a galaxy disk are described in the next sections. Two-fluid instabilities are considered, and the critical role of gas is emphasized (section \\ref{interact}). Because of its dissipative character, the gas can cool down as soon as instabilities heat it, and maintain non-axisymmetric features like spiral structure.The corresponding gravity torques are the tool to tranfer angular momentum. The most wide-spread instabilities are the $m=2$ spirals and bars: their formation mechanisms, their family of orbits, their resonances, etc... are essential to better understand the dynamical gas fueling (section \\ref{bars}). It will be shown how bars can be destroyed, or re-born, how bars within bars develop, and how material could be driven towards the nucleus more efficiently through a hierarchy of bars.  Also very frequent are the $m=1$ instabilities, lopsidedness and off-centring. Several possible mechanisms will be described for these instabilities, that can also favor material infall to the center (section \\ref{m1}). These include counter-rotation, warps or peculiar gaseous instabilities in a near-keplerian disk, a situation common in the neighborhood of central supermassive black holes.  Finally, external triggers through galaxy interactions and mergers should also be emphasized. Observational evidence of internal and external fueling will be discussed and compared. The role of gas and bars in minor and major mergers is investigated through numerical simulations. These external mechanisms are favored by hierarchical scenarios of galaxy formation, where massive galaxies are the result of the merging of several smaller entities. In this frame, it is expected that the central black hole mass increases in parallel to the mass of the galaxy, as observations seem to confirm.  One consequence of galaxy merging is the possibility of binary black hole formation. The evolution and history of the black hole population will be investigated through the main dynamical mechanisms (section \\ref{binbh}), and implications on the fueling will be explored. The Starburst-AGN connection (competition or collaboration) is then described, in a cosmological perspective. ",
        "conclusions": "The source of fueling depends on the strength and luminosity of the AGN:  -- for low luminosity AGN and Seyferts, only stars from a dense nuclear cluster are sufficient (through tidal distorsions and stellar collisions),  -- for  high luminosity AGN and quasars, large accretion rates are required, which involve large-scale gravitational instabilities. Those drive gas towards the center that trigger big starbursts, and the coeval compact cluster just formed can provide the fuel through mass loss of young stars and supernovae. This gas must have been driven from the galactic disks, through internal gravitational instabilities (bars, spirals), more generally the consequence of interactions and mergers.  Galaxy disks are in general far from stationary equilibrium, most often subject to $m=2$ and $m=1$ instabilities. These non-axisymmetric instabilities produce gravity torques, that drive the gas inwards. When the mass concentration towards the center is large enough, a secondary bar can decouple and rotate with a higher pattern speed. Embedded bars or non-axisymmetric structures can take over the gas flow to fuel the nucleus. The fueling is therefore favored in early-type objects.  Interactions and mergers also produce bars and non-axisymmetric structures, that fuel the nucleus through their gravity torques. They first trigger huge starbursts in the merger center, that could afterwards fuel the AGN. Since internal instabilities, external trigger and hierachical merging both produce the bulges and fuel the nucleus, it is natural to expect a proportionality ratio in their masses."
    },
    "0010/astro-ph0010216_arXiv.txt": {
        "abstract": "We present $UBVRIJHK$  polarization and flux density observations of the quasar 3C 273 obtained during a time of outburst over two weeks in 1988 February. We have modelled these data with two power law components, each with wavelength-independent position angle. These components are roughly perpendicular. The steeper-spectrum component has higher infrared polarized flux density, with the electric vector approximately transverse to the projected direction of the VLBI jet. The K-band polarized flux density and position angle, and possibly the spectral index of the two components, are correlated with a time lag of less than a day. We explain our results in terms of a shocked jet model with two nearly co-spatial components: a shock component with magnetic field approximately perpendicular to the jet and the other with magnetic field approximately parallel to the jet. ",
        "introduction": "Our data consist of polarization position angle (E-field), percentage polarization, and total flux at $UBVRIJHK$ for 3C 273 for each night of UT 10, 12-18 February 1988.  The data for a representative night (12 Feb.) are shown in Figure 1, where the solid lines represent the best-fitting model for that night (see \\S 2).  The data were taken at UKIRT with the Mark II Hatfield Polarimeter (Hough, Peacock, \\& Bailey 1991), which makes simultaneous optical and near-infrared observations, important for this variable blazar. The data have been corrected for interstellar polarization and Galactic extinction. A brief discussion of these data was presented by Wills (1991).  At the time of our observations 3C 273 was in an outburst state (Courvoisier et al. 1988). This outburst has since been associated with the ejection of VLBI component C9 ($t_0=1988.1\\pm0.1$, B\\aa \\aa th et al. 1991). At this time the polarization was higher than usual.  In our data both the percentage polarization and the position angle are very frequency-dependent, and this dependence changes from night to night. The frequency dependence of $\\theta$, particularly a 90\\deg\\ flip on 14 Feb., is suggestive of the presence of two perpendicular polarized components of similar amplitude but different spectral shape. ",
        "conclusions": "We explain the polarization of 3C 273 in outburst with two synchrotron components. Since these components are nearly perpendicular they tend to cancel each other out, resulting in low polarized flux. This blazar-like polarization is further diluted by a strong, unpolarized non-blazar continuum, resulting in the low observed percentage polarization for this object.  The outburst emission of blazars is often explained in terms of a shocked jet model like that of Marscher \\& Gear (1985). Such models normally involve a quiescent component and a shock component. We instead require two components associated with the shock. These components must be nearly co-spatial to account for the very short time lag between their polarizations. The simplest model would involve an underlying magnetic field that is parallel to the jet, but which is compressed so as to be perpendicular in the region of the shock. One of our components, which has polarization parallel to the jet and  a flatter spectrum, would originate from the region of the perpendicular B-field, as expected from the Marscher \\& Gear model. The other component, with perpendicular polarization, would be due to high energy electrons that have leaked from the shocked region and are accelerated where the B-field is parallel to the jet. We see these two components superimposed."
    },
    "0010/astro-ph0010399_arXiv.txt": {
        "abstract": "We have made observations of the four hyperfine transitions of the $^{2}\\Pi_{3/2}$, J=3/2 ground state of OH at 1612, 1665, 1667 and 1720 MHz and the related 1.6-GHz continuum emission, towards NGC 6334 using the Australia Telescope Compact Array. The observations covered all the major radio continuum concentrations aligned along the axis of NGC 6334 (V, A to F). We have detected seven OH masers plus a possible faint eighth maser; two of these masers are located towards NGC 6334-A. Absorption at 1665 and 1667 MHz was detected towards almost all the continuum distribution. All transitions showed non-LTE behaviour. The 1667-/1665-MHz intensity ratios ranged from 1.0 to 1.2, significantly less than their LTE value of 1.8. The results of the OH `Sum Rule' suggest that this discrepancy cannot be explained solely by high optical depths. The 1612- and 1720-MHz line-profiles showed conjugate behaviour whereby one line was in absorption and the other in emission. In addition, the profiles commonly showed a flip from absorption to emission and vice versa, which has been interpreted as a density gradient. The OH line-to-continuum distribution, optical depth and velocity trends are consistent with a bar-like shape for the molecular gas which wraps around the continuum emission. ",
        "introduction": "NGC 6334 is a prominent \\HII\\ region/molecular cloud complex located in the Sagittarius arm of the Milky Way at a photometric distance of 1.7 kpc \\cite{Neckel78} ($1 {\\mathrm pc} \\approx 2$ arcmin). The complex contains several recent and current star-forming sites which are embedded in an elongated giant molecular cloud (GMC) extending over about 45 arcmin (\\citeNP{Dickel77}, \\citeNP{Kraemer99}).  Far-infrared observations at wavelengths 80--250 microns by \\citeN{McBreen79} revealed six strong continuum concentrations, the peaks of which are designated FIR-I to FIR-VI. Sources I to V are present along a ridge of extended continuum emission which runs northeast--southwest, parallel to the plane of our Galaxy, while the weak, extended source VI is situated further south. Observations at 4.9 GHz by \\citeN{Rodriguez82} yielded six continuum concentrations also located along the ridge. Designated A to F, some are coincident with the far-infrared concentrations. The continuum ridge is coincident with a dark dust lane seen at optical wavelengths (e.g \\citeNP{Gardner75}). The dust lane appears to be obscuring the optical ionized distribution, suggesting that it is overlying the nebulosity. The radio continuum concentrations have been individually observed in several projects (see discussion in Section 3.1). The results show that most of them have associated outflows, sites of maser emission, or embedded protostars, all indicative of very active sites of star formation.  To further investigate the physical conditions of the NGC 6334 complex, we have carried out a series of observations of the 1612-, 1665-, 1667-, and 1720-MHz hyperfine transitions of the $^{2}\\Pi_{3/2}$, J=3/2 ground state of OH using the Australia Telescope Compact Array (ATCA)\\footnote{The Australia Telescope is funded by the Commonwealth of Australia for operation as a National Facility managed by CSIRO.}.  These four transitions are valuable in investigations of molecular clouds because they can exhibit extended emission and absorption and maser emission.  A direct result of the statistical weights and transition probabilities of the `main lines' (at 1665 and 1667 MHz) and `satellite lines' (at 1612 and 1720 MHz) is that, under conditions of local thermodynamic equilibrium (LTE), their optical depths, $\\tau_{1612}$, $\\tau_{1665}$, $\\tau_{1667}$, $\\tau_{1720}$, are in the ratio 1:5:9:1. Previous studies of OH have shown that both the main-line and satellite-line intensities often show non-LTE ratios.  The satellite lines can also exhibit a conjugate-type behaviour whereby emission of one line is accompanied by absorption of the other. This behaviour is explained in detail by \\citeN{Elitzur92} and references therein. The satellite-line behaviour can be used as a diagnostic for the type of OH excitation process taking place in the region and may also provide density constraints.  For example, radiative excitation at high densities can produce 1612-MHz emission but collisional excitation at low temperatures can lead to 1720-MHz emission \\cite{Elitzur92}.  The ratios of the optical depths of the four transitions give \\begin{equation} \\tau_{1612} + \\tau_{1720} = \\frac{\\tau_{1665}}{5} + \\frac{\\tau_{1667}}{9}. \\end{equation} This relationship is commonly termed the `Sum Rule' (e.g. \\citeNP{Rogers67}) and can be used to obtain optical depth measurements even in cases where LTE is not satisfied.  For low optical depths and an extended continuum emission with a flat spectrum the Sum Rule becomes \\begin{equation} {T_{\\rm 1612}} + {T_{\\rm 1720}} = \\frac{T_{\\rm 1665}}{5} + \\frac{T_{\\rm 1667}}{9}, \\end{equation} where $T_{line}$ is the line temperature of the transition. ",
        "conclusions": "We have imaged the 1.6-GHz continuum and the four transitions of the $^{2}\\Pi_{3/2}$, J=3/2 ground state of OH towards the NGC 6334 \\HII\\ region/molecular cloud complex. Observations were made with the Australia Telescope Compact Array with angular resolutions of $4.6 \\times 7.2$ arcsec$^2$ and $25.8 \\times  19.6$ arcsec$^2$. The results are as follows:  \\begin{description} \\item (a) {\\it 1.6-GHz Continuum distribution} - The continuum emission has been detected over an area of $35 \\times 25$ arcmin$^2$. The main region shows a set of aligned concentrations in stages of evolution ranging from compact regions to well-evolved shells. They are superimposed on a faint extended area. Two additional southern regions appear as extended incomplete shells.  \\item (b) {\\it OH maser emission} - We have detected seven masers, plus a possible faint maser candidate. Including the eighth possible maser, the masers are present for all OH transitions and appear to be associated with the continuum concentrations. Two of the masers were detected towards NGC 6334-A. All maser locations include main-line transitions.  The maser velocities vary between $-8.7$ and 0.5 \\kms, with the more positive velocities located near the centre of the continuum distribution.   \\item (c) {\\it Extended OH distribution} - Main-line OH absorption was detected towards almost all of the continuum emission. The distribution of the line-to-continuum ratio at 1667-MHz ratio is uniform across the continuum and varies between $-0.4$ and $-0.9$.  The distribution is not enhanced towards the continuum emission but is concentrated in a bar extending northeast--southwest with its sharp south-eastern edge coincident with that of the optical dust lane. Our velocity and line-to-continuum results support the notion in which the molecular bar wraps around the continuum emission.   \\item(d) {\\it Non-LTE behaviour}  For the satellite lines, one transition is commonly in emission with the other simultaneously in absorption. The 1612-MHz transition is in emission near the prominent radio continuum concentrations, presumably reflecting OH excitation by the infrared radiation associated with these concentrations. In some cases, the conjugate-type behaviour is observed to reverse across a spectral-line profile. This has been interpreted as a density gradient, and we have suggested that this reflects the presence of different cloud regimes along a line of sight. The main-line transitions show non-LTE behaviour in that the $T_{1667}/T_{1665}$ intensity ratios range from 1.0 to 1.2, significantly lower than the LTE value of 1.8. The results of the OH Sum Rule suggest that this discrepancy can only be explained in terms of high optical depth if the cloud is clumping on smaller scales.  \\end{description}"
    },
    "0010/astro-ph0010350_arXiv.txt": {
        "abstract": "Do disk galaxies exist at redshifts much greater than unity, and how might they look different from local disks? How does the morphological mix of galaxies change with redshift?  What can we learn from current observations about the properties of high redshift galactic disks?  I present theoretical predictions based on semi-analytic hierarchical models, focussing on the role played by interactions, mergers, and starbursts in determining the observable properties of disk galaxies at high redshift, and discuss the interpretation of high redshift observations of possible proto-disks (damped Lyman-$\\alpha$ systems and Lyman-break galaxies) in light of these predictions. ",
        "introduction": "Very little is known about the properties of disk galaxies at redshifts larger than about unity; even whether they exist at all. The high redshift universe is a rather dangerous place for galactic disks, which are relatively fragile objects. Mergers are more frequent at early times, and the characteristic gas fractions and internal densities of early objects are high. Are violent mergers so frequent at high redshift that they destroy most or all disks? Are high redshift disks more or less susceptible to bar instabilities than local disks?  Before we begin to discuss theoretical answers to these questions, based on the currently accepted paradigm of disk formation within a hierarchical universe, let us address a semantic and practical question --- what do we mean by a ``disk'' galaxy, and how do we know one when we see it? The defining feature of a ``true'' disk is fundamentally kinematical: disks are objects that are flattened and supported by rotation. In the nearby universe, we can obtain direct observational measures of galaxy kinematics and (in principle) determine unambiguously whether an object is a disk or not. However, these observations become increasingly difficult as one pushes to high redshift.  A small sample of disk galaxies with observed rotation curves extends out to a redshift of about unity (the current record holder is at $z=1.3$; N. Vogt, private communication). For the moment, and probably for the near future, if we wish to identify disks at still higher redshifts, we will be forced to rely on more indirect observational evidence.  For example, we know that in the local universe, if we select galaxies by their morphology (spirals and irregulars), or the shape of their light profiles (exponential), or those that have blue colors or strong emission lines, we will end up with a sample that consists mostly of ``true'' disks. However, by some of these conditions, some interlopers (starbursts?) might creep in, and some ``true'' disk galaxies (early-type spirals and lenticulars) might be left out. It is unclear whether these empirical associations will remain useful at high redshift. None the less, I will discuss what they might tell us about the $z\\sim3$ Lyman-break galaxy population (e.g. Steidel et al. 1999). Another intriguing kind of observation is the kinematic data from unsaturated, low-ionization-state metal lines in high column density quasar absorption systems (damped Lyman-$\\alpha$ systems (DLAS); Prochaska \\& Wolfe 1997, 1998). These absorption profiles may provide kinematic information about gaseous disks at high redshift; however, as I discuss in Section~3, interpreting these data may be complicated. ",
        "conclusions": "Observational evidence (e.g. Le F\\`{e}vre et al. 2000) as well as theoretical studies based on dissipationless N-body simulations (Kolatt et al. 2000) support the expectation that galaxy merger rates were higher in the past.  There is a growing realization that mergers of gas-rich disks are instrumental in driving several important processes: radial gas inflows, disk destruction/spheroid production and transferring orbital angular momentum to the cold disk gas.  These processes in turn have profound implications for several observed populations at high redshift. Gas inflows probably fuel starbursts, which may be responsible for activating star formation in a large fraction of the high redshift galaxies identified in the optical and sub-mm (SPF). At the same time, these inflows may also fuel AGN activity (Kauffmann \\& Haehnelt 2000). Producing spheroids via violent mergers leads very naturally to the observed morphology-density relation, as mergers take place preferentially in dense environments. Minor mergers, which are far more common, may also produce morphological evolution towards earlier Hubble types, by triggering nuclear bursts of star formation. In addition, mergers may transfer orbital angular momentum to the cold disk gas. Initially, this gas will be launched to large distances in the form of tidal tails, but most of it will eventually fall back to form a new disk. This process may introduce an important new ingredient into the `standard paradigm' of disk formation, in which all disk angular momentum is acquired from early tidal torques.  Lines of sight passing through the extended gaseous tidal tails in gas rich, interacting galaxies could produce the large velocity width Damped Lyman-$\\alpha$ systems observed by Prochaska \\& Wolfe (1997, 1998), resolving the conflict between these observations and hierarchical models. If correct, this picture has many interesting implications, such as a decreasing fraction of large $\\delv$ profiles at lower redshift. If the same interaction also produces a starburst, this also leads to specific predictions for the probability of detecting nearby objects in emission, which we are currently investigating (Maller et al. in prep).  The prediction of a CDM-based semi-analytic model is that many disks exist at high redshift, but most have experienced recent interactions. The fraction of objects showing signatures of morphological disturbance is therefore expected to increase dramatically with redshift. There are mounting indications that such a trend is indeed observed. If confirmed, this constitutes direct evidence of hierarchial structure formation in action, and may even be utilized to obtain a measure of the merger rate as a function of redshift. One way to quantify galaxy morphology is via statistics that simultaneously measure the ``concentration'' and ``asymmetry'' of an image. This appears to be a promising way to pick out interacting galaxies and remnants of recent mergers, especially when combined with other information such as color (Conselice et al. 2000).  In order to interpret the observational results, however, it is important to calibrate the statistics by calculating them in the same way on observations and on synthetic images of simulated interacting galaxies at various redshifts. I have presented preliminary results from an ongoing program to carry out such a comparison with the HDF, using the concentration and asymmetry statistics defined by Wu (1999). An encouraging level of agreement is obtained between the statistics calculated on the real and synthetic images, supporting the hypothesis that the high redshift HDF galaxies with high asymmetries are in fact ongoing interactions or recent mergers. With further refinement of the simulations and more extensive observations, this should be a promising way to directly probe the merger history and assembly of galaxies in the present-day universe."
    },
    "0010/astro-ph0010166_arXiv.txt": {
        "abstract": "Using data obtained with the {\\it Rossi X-ray Timing Explorer}, we report the detection of a 5 Hz quasi-periodic oscillation (QPO) in the bright low-mass X-ray binary and Z source Cygnus X-2 during high overall intensities (the high intensity state). This QPO was detected on the so-called normal branch and can be identified with the normal branch QPO or NBO.  Our detection of the NBO is the first one during times when Cygnus X-2 was in the high intensity state.  The rms amplitude of this QPO decreased from 2.8\\% between 2--3.1 keV to $<$1.9\\% between 5.0--6.5 keV. Above 6.5 keV, its amplitude rapidly increased to $\\sim$12\\% rms above 16 keV. The time lags of the QPO were consistent with being zero below 5 keV (compared to the 2--3.1 keV band), but they rapidly increased to $\\sim$70 ms (140\\degr) around 10 keV, above which the time lags remained approximately constant near 70 ms.  The photon energy dependencies of the rms amplitude and the time lags are very similar to those observed for the NBO with other satellites ({\\it Ginga}, {\\it EXOSAT}) at different (i.e. lower) intensity states. ",
        "introduction": "}  Although the low-mass X-ray binary (LMXB) and Z source (Hasinger \\& van der Klis 1989) Cygnus X-2 is one of the best studied LMXBs, its complex behaviour is far from understood. On time-scales of hours to days, the source traces out its characteristic Z shape pattern in the X-ray colour-colour diagram (CD). The branches of this Z are called, from top to bottom, the horizontal branch (HB), the normal branch (NB), and the flaring branch (FB). On time-scales of weeks to months, the overall intensity varies smoothly by a factor of about 4 (see, e.g., Fig.~\\ref{fig:asm_lightcurve}). The exact morphology of the Z track and the position of this track in the CD (and hence the overall hardness of the X-ray spectrum) varies significantly when the overall intensity changes (see, e.g., Kuulkers, van der Klis, \\& Vaughan 1996 and Wijnands et al. 1997 for overviews of these changes).  The first few months of the {\\it Rossi X-ray Timing Explorer} ({\\it RXTE}; Bradt, Rothschild, \\& Swank 1993) All Sky Monitor (ASM) data of Cygnus X-2 suggested an $\\sim$78 day period in the long-term variations (Wijnands, Kuulkers, \\& Smale 1996). Although this period is still one of the strongest periods, analyses of all {\\it RXTE}/ASM data now available (up to 2000 October 9) show that the long-term variations are more complex than only one single periodicity (see also Kong, Charles, \\& Kuulkers 1998 and Paul, Kitamoto, \\& Makino 2000).   \\begin{table*} \\caption{Log of the {\\it RXTE} observations of Cygnus X-2\\label{tab:log}} \\begin{flushleft} \\begin{tabular}{llll} \\hline Observation ID & Start & End  & On source time\\\\ & (UT)  & (UT) & (ks)\\\\ \\hline 30418-01-01-00 & 1998 July 2 10:48 & 1998 July 2 14:30 & 8.2 \\\\ 30418-01-02-00 & 1998 July 3 11:37 & 1998 July 3 12:42 & 3.3 \\\\ 30418-01-02-01 & 1998 July 3 13:13 & 1998 July 3 14:15 & 3.7 \\\\ 30418-01-03-00 & 1998 July 4 11:38 & 1998 July 4 14:15 & 7.3 \\\\ 30418-01-04-00 & 1998 July 5 11:38 & 1998 July 5 14:14 & 7.3 \\\\ 30418-01-05-00 & 1998 July 6 09:07 & 1998 July 6 14:14 & 11.7 \\\\ \\hline \\end{tabular} \\end{flushleft} \\end{table*}  The physical mechanism behind the long-term variations in Cyg X-2 is unknown. Several mechanisms have been proposed, such as variations in the mass accretion rate onto the neutron star, a precessing accretion disc, or a precessing neutron star (see Wijnands et al. 1997 and references therein for a discussion of these possible mechanisms). Although variations in the mass accretion rate are unlikely to cause the long-term variability (these variations are thought to produce the motion of Cyg X-2 along the Z track; see, e.g., Hasinger \\& van der Klis 1989, Kuulkers et al. 1996, and Wijnands et al. 1997), a precessing accretion disc or a precessing neutron star are still possible. Besides affecting the X-ray count rate and the X-ray spectrum, both mechanisms would most likely also affect the rapid X-ray variability, although it is unclear how exactly the variability would be affected. A detailed study of the rapid X-ray variability during different overall intensities might give new insights into the physics behind the long-term variability of Cyg X-2. However, such a detailed study has not yet been performed. Below we briefly list the results so far available.  On the HB, quasi-periodic oscillations (QPOs) between 15 and 55 Hz are observed (the horizontal branch QPOs or HBOs).  These QPOs are also often observed on the upper part of the NB. A detailed comparison of the HBO properties during different overall intensities (hereafter referred to as different intensity states) has not yet been made. Although Wijnands et al. (1997) reported that the HBO is always observed on the HB regardless of the overall intensity of Cygnus X-2, the HBO properties could differ in detail between different intensity states.  On the NB, QPOs with frequencies near 5 Hz can also be observed (the normal-branch oscillations or NBOs; sometimes observed simultaneous with the HBO). Also the NBO properties have not yet been compared between different intensity states, although Wijnands et al. (1997) did not detect the NBO (on the lower NB; with typical upper limits of $<$1\\% rms on the NBO amplitude) during the high intensity states observed using {\\it Ginga} data of Cyg X-2. The NBOs {\\it were} observed during lower intensity states.  Another type of QPO (with frequencies between 300 and 900 Hz) was recently discovered (Wijnands et al. 1998) in Cygnus X-2 when this source was on the HB during a medium intensity state: the kHz QPOs. These QPOs were not observable during a higher intensity state (Smale 1998) with upper limits which were smaller than the measured values.  Here, we report the first detection of a NBO in Cyg X-2 during a high intensity state demonstrating that this type of QPO is so far always observable regardless of the overall intensity of the source.  The energy dependence of the rms amplitude and the time lags of this NBO are very similar to those observed during lower intensity states. ",
        "conclusions": "}  We have detected the NBO on the upper part of our observed normal branch of Cygnus X-2 when this source was at relative high overall intensities (Fig.~\\ref{fig:asm_lightcurve}). The shape of the Z track in the CD and the HID was very similar to those observed with {\\it Ginga} for Cygnus X-2 in the high intensity state (Wijnands et al. 1997; i.e., compare their Fig.~1 with our Fig.~\\ref{fig:cd_hid}). Therefore, we conclude that Cygnus X-2 was in a high overall intensity state during the observations reported here. This conclusion is strengthened by the fact that our power spectra on the lower normal branch resemble those obtained by Wijnands et al. (1997) in the high intensity state, especially the non-detection of the NBO on the lower part of the NB (compare their Figs.~8{\\it g--i} with our Figs.~\\ref{fig:powerspectra_z}{\\it c} and {\\it d}).  Wijnands et al. (1997) were unable to search for the NBO further upwards on the NB (the middle and upper part of the NB) due to the fact that they did not have enough high time resolution data at those positions on the Z track. Our {\\it RXTE}/PCA data are the first high time resolution data available on the middle and upper part of the NB of Cygnus X-2 during the high intensity state.  Our first clear detection of the NBO in the high intensity state of Cygnus X-2 shows that all power spectral components which are observed below 100 Hz in Cygnus X-2 are present at high overall intensities. So far, the only clear difference in the timing properties of Cygnus X-2 during different intensity states are the detection of the kHz QPOs on the HB during a medium intensity state (Wijnands et al. 1997) but not during high intensity states (Smale 1998 and this paper). More detections of the kHz QPOs in Cygnus X-2 are needed in order to determine the exact reason(s) why the kHz QPOs are only present during preferred intensity states.  Wijnands et al. (1997) pointed out that a precessing accretion disc, one of the proposed explanations for the long-term X-ray variations of Cyg X-2 (e.g., Kuulkers et al. 1996; Wijnands et al. 1996), could not easily account for the non-detection of the NBO during the high intensity state. Our detection of the NBO during this state now shows that the arguments used by Wijnands et al. (1997) are not valid any more when only considering the NBO. However, when also considering the kHz QPOs, it is clear that the precessing accretion disc model has still difficulties explaining the difference in rapid X-ray timing variability (i.e., the kHz QPOs) when Cyg X-2 is at different intensity states.  In the precessing accretion disc scenario, it is possible that more matter is blocking the emission region during low intensity states (hence the lower overall intensity) than during high intensity states. If this the case then during these low states, the amplitudes of the kHz QPOs should be significantly lower than in the high states due to more scattering of the radiation. However, we see exactly the opposite: no kHz QPOs are detected during the high intensity state. It is also possible that the precession of the accretion disc causes changes in the projected area of the inner disc which could result in the different intensity states. However, then the kHz QPOs would be expected in all intensity states, i.e., also during the high intensity state.  The non-detections of the kHz QPOs during the high intensity state constitute a serious problem for the precessing accretion disc model.  We now compare our results obtained for the NBO to those which have been published previously. The best cases to compare our results with are the {\\it Ginga} June 1987 observations of Cygnus X-2 reported by Mitsuda \\& Dotani (1989; see also Wijnands et al. 1997) and the {\\it EXOSAT} November 14/15 1985 observations which are reported by Dieters et al. (2000; see also Hasinger et al. 1985,1987). Wijnands et al. (1997) classified the intensity state of Cyg X-2 during the {\\it Ginga} observations as intermediate between the high and the medium intensity state. Kuulkers et al (1996) classified the state during the {\\it EXOSAT} observations as the medium state (see Wijnands et al. 1997 for the classification of the different states of Cygnus X-2).  During the {\\it Ginga} observations (Mitsuda \\& Dotani 1989; Wijnands et al. 1997) and the {\\it EXOSAT} observations (Dieters et al. 2000), the NBO was best visible on the middle part of the NB. The NBO was much less clear on the upper and lower parts of the NB (although excess power on top of a power-law function was present near 5 Hz). The exact position on the Z track where the NBO is present during those observations is very similar to the positions on the track where we detect the NBO.  Although we detected the NBO up to the upper most part of our observed NB and Mitsuda \\& Dotani (1989) and Dieters et al. (2000) did not detect the NBO on the upper part of their observed NB, this difference is most likely due to the fact that we did not observe the NB all the way to the HB/NB vertex (see Fig.~\\ref{fig:cd_hid}{\\it b}) and most likely they did.  If we would have observed the NB all the way to the HB/NB vertex most likely we would not have detected the NBO either on this part of the Z track. Wijnands et al. (1997) also reported that the NBO was best visible at the middle of the NB during the medium state and that it tends to get broader when the source moves further down the NB. The energy dependence and time lags of our NBO are also nearly identical to those obtained for the NBO by Mitsuda \\& Dotani (1989) and Dieters et al. (2000). This indicates that the physical mechanism behind the long-term variation of the X-ray flux of Cygnus X-2 does not significantly affects the properties of the NBO.   \\begin{figure} \\begin{center} \\begin{tabular}{c} \\psfig{figure=Fig6.ps,width=8cm} \\end{tabular} \\caption{The count rate of the upper part of the observed NB ({\\it a}), the rms amplitude of the NBO ({\\it b}), and the time lags of the NBO ({\\it c}) as a function of photon energy. The errors on the photon energy range from the averaged mean photon energy in the energy channels to the boundary of these channels. \\label{fig:timelags}} \\end{center} \\end{figure}  From the presently available results, we conclude that so far no significant difference is present in the NBO properties during the high intensity state of Cygnus X-2 and during its intermediate and medium intensity state.  Recently, Kuulkers, Wijnands, \\& van der Klis (1999) reported on {\\it RXTE} observations of Cyg X-2 when this source was in a very low intensity state. No NBOs were detected. However, no complete Z track was traced out during these observations but most likely only the lower part of the NB and part of the FB. It is possible that when a complete Z track would have been traced out, the NBO would have been present at the middle part or upper part of the NB.  To explain the NBO in Z sources, it has been proposed that, when those sources are accreting close to the Eddington accretion limit, part of the accretion will occur in an approximately spherical radial inflow. In this radial flow a radiation pressure feed-back loop can be set up that causes 5--7 Hz oscillations in the optical depth of the flow (Fortner et al. 1989; Lamb 1989; Miller \\& Lamb 1992). This radiation-hydrodynamic model is able to explain the combination of the minimum in the rms amplitude spectrum of the NBO and the $\\sim$140--150\\degr~phase shift in the phase lags at the same photon energy (Miller \\& Lamb 1992). Miller \\& Lamb (1992) used the results reported by Mitsuda \\& Dotani (1989) to constrain their model. Because our results are so similar to those published by Mitsuda \\& Dotani (1989), we are unable to constrain this radiation-hydrodynamic model further. In order to constrain this model, data are needed with a higher energy resolution than our {\\it RXTE}/PCA data and/or data have to be obtained for the energy range below 3 keV, a region which we are unable to probe with our {\\it RXTE}/PCA data.  However, it has already been showed that this model cannot easily explain the differences in the NBO properties observed for Cyg X-2 and those observed for GX 5--1 and Sco X-1 (see, e.g., Dieters et al. 2000).  The recent discovery of a QPO near 7 Hz in the atoll source 4U 1820--30 (Wijnands, van der Klis, \\& Rijkhorst 1999) might even impose a more serious problem to all NBO models.  The properties of this QPO in 4U 1820--30 (i.e., its frequency and its presence only at the highest observed mass accretion rates in this source) are similar to the NBO in the Z sources. However, if indeed these QPOs are the same phenomenon, then the models explaining the NBO in Z sources which require near-Eddington mass accretion rate, will not hold because the highest mass accretion rate observed in 4U 1820--30 (i.e., the accretion rate at the time of the 7 Hz QPO) is significantly lower than the critical Eddington mass accretion rate.  This is not only true for the radiation-hydrodynamic model of Fortner et al. (1989), but also for alternative models, such as the sound wave model of Alpar et al. (1992). Clearly, the NBO models have to be significantly adjusted in order to explain all NBO properties and to explain the similarities between the NBO and the 7 Hz QPO in 4U 1820--30."
    },
    "0010/astro-ph0010417_arXiv.txt": {
        "abstract": "{ The OGLE data base is used in conjunction with Kurucz atmosphere models to generate sets of period, effective temperature and luminosity for fundamental and overtone Magellanic Cloud Cepheids. The Florida pulsation code (with linear turbulent convection) is then used to compute masses for these stars, assuming an average composition of ($X$=0.716, $Z$=0.010) for the LMC and of ($X$=0.726, $Z$=0.004) for the SMC.  The average $M$--$L$ relation for the fundamental Cepheids matches closely that for the first overtone Cepheids for each Magellanic Cloud.  Neither the SMC nor the LMC average $\\Log M$--$\\Log L$ relations are straight, but have a noticeable curvature. In view of the uncertainties in distance and reddening we have adopted three different choices for these quantities. The results based on the 'long' distance scale to the clouds give a better agreement between theory and and observations than the 'short' one. All the current evolutionary tracks predict systematically larger masses for given luminosities than our observationally derived ones, especially at the high end.  Moreover, our study confirms that the evolutionary tracks of the low mass stars in SMC are not in agreement with the observations as they do not extend sufficiently blueward and do not penetrate deep enough into the instability strip, or not at all. The inference of masses directly from the observational database yields a novel and strong constraint on evolutionary calculations. \\keywords {stars : oscillations -- stars: Cepheids -- Stars: Evolution, Magellanic Clouds, distance moduli}} ",
        "introduction": "In the last few years, high quality data on large numbers of Cepheid variables in the Small and Large Magellanic Clouds have been made available by the EROS and OGLE microlensing projects (Beaulieu \\etal\\ 1995, Afonso \\etal\\ 1999, Udalksi \\etal\\ 1999abc).  In particular the OGLE Project has provided standard colors in addition to periods and magnitudes for the largest samples published to date.  In this paper we examine some of the constraints that the MC Cepheids impose on stellar evolution and stellar pulsation theories.  \\begin{figure*} \\centerline{\\resizebox{9.7cm}{!}{\\includegraphics{fig1a.ps}}\\hskip -20pt \\resizebox{9.7cm}{!}{\\includegraphics{fig1b.ps}}} \\hfill \\vskip -10pt \\parbox[b]{17cm}{ \\caption[]{\\small Mass--Luminosity relations for SMC  and LMC as derived from the OGLE data adopting choice (A) for distance modulus and reddenings in the left panel and our preferred choice (B) in the right panel; fundamental Cepheids are shown as dots and overtones as open circles. As a guide to the eye we have superposed the evolutionary $M$-$L$ relations of Girardi \\etal. } \\label{mcab} } \\end{figure*}  We use the full catalogue of publicly available of LMC and SMC single mode Cepheids and SMC double mode Cepheids produced by OGLE in BVI (Udalski \\etal, 1999abc, with the zero point corrections as suggested in the April 2000 OGLE web site, U99 hereafter).  The single mode Cepheid catalogues contain 1435 LMC and 2167 SMC stars. We keep the objects classified as fundamental mode pulsators or first overtone pulsators, with reliable photometry in both V and I. We exclude stars whose magnitudes are most likely to be strongly contaminated by companions or blending in V or I.  The remaining stars form our working sample of OGLE Cepheids.  It consists of 670 LMC fundamentals, 426 LMC overtones, 1197 SMC fundamentals and 677 SMC overtones, as well as 24 F/O1 SMC double-modes and 71 O1/O2 SMC double-modes.  The OGLE data base provides intensity averaged magnitudes and colors.  With the help of distance moduli, these can then be transformed to luminosities and effective temperatures.  The Magellanic Clouds (MC) are thought to be relatively uniform in composition, and with the observed average compositional information theoretical modeling can then provide the mass of each star.   \\begin{figure} \\centerline{\\resizebox{9.7cm}{!}{\\includegraphics{fig1c.ps}}} \\hfill \\vskip -10pt \\parbox[b]{8.cm}{ \\caption[]{\\small Mass--Luminosity relations for SMC  and LMC as derived from the OGLE data with choice (C) for distance modulus and reddenings. Superposed are the $M$-$L$ relations of Girardi \\etal. } \\label{mcc} } \\end{figure}   We have to adopt distances and reddenings for the Magellanic Cepheids.  The distance of the Large Magellanic Cloud remains at the center of the current debate about the distance scale ladder. Extreme values of the distance modulus range from 18.08 to 18.70, but the distance estimates tend to cluster around 18.3 for the 'short' distance scale, and around 18.5 for the 'long' distance scale (see Walker 1999, Udalski 2000 and references therein, Cole 1998, Girardi 1998, Stanek \\etal\\ 2000, Romaniello \\etal\\ 2000, Feast \\& Catchpole 1997, Luri \\etal\\ 1998, Groenewegen \\& Oudmaijer, 2000 and references therein, Groenewegen 2000, Sakai \\etal\\ 2000, Cioni \\etal\\ 2000).  It is beyond the scope of the paper to solve this distance scale problem, but we will justify the choices we adopt for distances and reddenings.  U99 derive reddening by making the assumption that the red clump I luminosity has a weak metallicity dependence, and they use the I luminosity to map relative variations of E(B-V).  Finally they fix the zero point of the relative scales on the basis of observations of NGC1850, NGC1835, HV2274 in the LMC, and NGC416 and NGC330 in the SMC.  These assumptions lead to the determination of a 'short' distance to the LMC of $18.24 \\pm 0.05$ and $18.75 \\pm 0.05$ to the SMC.  The relative distance between LMC and SMC obtained with four different distance indicators (Cepheids, RR~Lyrae, red clump and tip of the red giant branch) but with the same reddening maps, gives a relative distance modulus of $0.50 \\pm 0.02$.  The reddening maps of U99 give a mean reddening for the LMC of 0.147, and of 0.092 for the SMC.  The consistency of the zero point of the reddening scale between these different observations is at the level of few times $10^{-2}$ (\\eg $E(B-V)=0.15 \\pm 0.05$ for NGC1850 and $E(B-V)=0.13 \\pm 0.03$ for NGC1835).  We note that these values are different from what is usually given as mean properties for the clouds (especially for the LMC, see Walker 1999 and references therein).  In his recent review, Walker (1999) noted that the median reddenings are $E(B - V) \\sim 0.1$ for the LMC and $E(B - V) \\sim 0.08$ for the SMC.  The galactic foreground reddenings along the line of sight of the clouds are known to be low, \\viz 0.06 and 0.04, respectively. The estimation of differential reddening inside the clouds based on earlier studies is quite uncertain.  In particular, heavily reddened stars ($E(B- V)=0.30$) can be found all over the LMC, but the typical range is 0 -- 0.15.  In view of these uncertainties and controversies, we consider three alternative choices for the distance moduli and for the reddening corrections for our derivation of stellar parameters from the OGLE data: \\\\ \\ni Choice (A) adopts both the distance moduli and the reddening as suggested by U99. \\\\ \\ni Choice (B) adopts the Cepheid distance modulus to the LMC of $18.55\\pm0.10$, and a relative distance between LMC and SMC of $0.42\\pm 0.05$ (Laney \\& Stobie, 1994), with the mean reddenings of $E(B- V)=0.1$ and $E(B- V)=0.08$ for LMC and SMC respectively. \\\\ \\ni Choice (C) is intermediate in that it adopts U99's relative distance between LMC and SMC of $0.50 \\pm 0.02$ and its reddening, but a 'long' LMC distance modulus of 18.55.  To summarize the differences, (B) -- (A), in distance moduli and in mean reddenings are $\\delta \\mu  = 0\\dotm 31$, $\\delta \\langle E(B - V) \\rangle \\sim 0.047$ for the LMC, and $\\delta \\mu  = 0\\dotm 22$, $\\delta \\langle E(B - V) \\rangle \\sim 0.012$ for the SMC. The differences, (C) -- (A), in distance moduli and in mean reddenings are $\\delta \\mu  = 0\\dotm 25$, $\\delta \\langle E(B - V) \\rangle = 0$ for the LMC and the SMC.    We follow Kov\\'acs (2000) in the conversion from magnitudes to bolometric, and from colors to effective temperatures.  Because of the wider range of masses that are needed for this study, we have redone the Kov\\'acs fits with the Kurucz (1995) stellar atmospheres from the BaSeL database (Lejeune et al., 1997) for the LMC and SMC chemical compositions for the temperature, luminosity and $\\log g$ range appropriate for Cepheids ranging from 100\\Lo\\ to 20,000\\Lo, with the results  \\begin{eqnarray} 2.5\\th \\Log\\th L &=& \\mu_{MC}- V +R_V\\th E(B\\moi V) + BC +4.75 \\\\ \\Log \\th g &=& 2.62 - 1.21\\th \\Log \\th P_0 \\\\ {\\rm SMC:\\hfill} \\nonumber \\\\ \\Log \\th\\teff &=& 3.91611   + 0.0055\\th \\Log \\th g \\nonumber \\\\ &&\\quad - 0.2482 (V\\moi I_c \\moi (R_V-R_I) E(B\\moi V))  \\\\ BC &=& -0.0324 + 2.01 \\Delta T - 0.0217\\th \\Log \\th g \\nonumber\\\\ &&\\quad - 10.31\\th \\Delta T^2  \\\\ {\\rm LMC:\\hfill} \\nonumber \\\\ \\Log \\th\\teff &=& 3.91545   + 0.0056\\th \\Log \\th g \\nonumber \\\\ &&\\quad - 0.2487 (V\\moi I_c \\moi (R_V-R_I) E(B\\moi V))  \\\\ BC &=& -0.0153 + 2.122 \\Delta T - 0.0200\\th \\Log \\th g  \\nonumber\\\\ &&\\quad - 11.65\\th \\Delta T^2 \\label{eqs} \\end{eqnarray}  \\noindent where $\\Delta T = \\log \\teff - 3.772$ and $L$ is in solar units.  The transformation to absolute luminosities is then made with the adopted distance moduli $\\mu_{MC}$ to the LMC or the SMC for the various choices (A), (B) and (C).  We note that compared to Kov\\'acs there are systematic shifts of 50K in \\Teff, and 0.02 in log $L$.  It may be objected that since $\\log g = \\log (G M ) /R^2$ in Eq.~(2) there is an implicit assumption about a mass--radius --period relation.  However, in Eqs.~(3 -- 6), $\\log g$ appears with a tiny multiplier, and over the period range from 1 to 10 days its contributions to Log L and Log \\Teff\\ vary by 0.013 and 0.0046, respectively.  Ultimately, there is essentially no feedback on our mass determination.    \\begin{figure*} \\centerline{\\resizebox{15cm}{!}{\\includegraphics{figres.ps}}} \\hfill \\vskip -10pt \\parbox[b]{17cm}{ \\caption[]{The two upper panels gives the residuals of the V and I $P$--$L$ relation for uncorrected reddening in LMC and SMC. The two lower are giving the residuals of the V and I $P$--$L$ relation after reddening corrections based on the red clump stars following the U99 procedure. On each panel are indicated in solid the reddening line, and in dash the depth dispersion (the diagonal) line. }  \\label{res} } \\end{figure*} ",
        "conclusions": "We have used the OGLE data base and Kurucz atmosphere models to obtain periods, effective temperatures and luminosities for fundamental and overtone Magellanic Cloud Cepheids.  With an assumed average composition of ($X$=0.716, $Z$=0.010) for the LMC and of ($X$=0.726, $Z$=0.004) for the SMC, stellar masses have been computed with the Florida pulsation code for the observationally derived $P$, $L$ and \\Teff.  The best results are obtained by adopting the 'long' distances to the clouds.  Reddening maps, as opposed to average MC reddenings, do not reduce the unexpectedly large scatter in $M$--$L$, and we conclude that individual stellar reddenings would be necessary.  However, the largest source of the scatter is in the V and I photometric errors.  The $M$--$L$ relations for the fundamental and for the first overtone Cepheids match closely for each Magellanic Cloud.  Both the SMC and the LMC $\\Log M$--$\\Log L$ relations have a noticeable curvature.  A comparison with the predictions of the recent stellar evolution calculations show a discrepancy in the $M$--$L$ diagrams, where the evolutionary tracks are underluminous.  It also confirms a discrepancy in the theoretical HR diagrams, especially for the SMC, where the low mass tracks do not extend sufficiently blue-ward to penetrate the instability strip. Our $M$--$L$ relations are useful as a new type of constraint that the evolutionary tracks have to satisfy."
    },
    "0010/astro-ph0010621_arXiv.txt": {
        "abstract": "We investigate the structure of self-gravitating polytropic stellar systems. We present a method which allows to obtain approximate analytical solutions, $\\psi_{n+\\epsilon}({\\bf x})$, of the nonlinear Poisson equation with the polytropic index $n+\\epsilon$, given the solution $\\psi_n({\\bf x})$ with the polytropic index $n$, for any positive or negative $\\epsilon$ such that $|\\epsilon|\\ll1$. Application of this method to the spherically symmetric stellar polytropes with $n\\simeq5$ yields the solutions which describe spatially bound systems if $n<5$ and the formation of a second core if $n>5$. A heuristic approximate expression for the radial profile is also presented. Due to the duality between stellar and gas polytropes, our results are valid for gaseous, self-gravitating polytropic systems (e.g., molecular clouds) with the index $\\gamma\\simeq 6/5$. Stability of such systems and observational consequences for both stellar and gaseous systems are discussed. ",
        "introduction": "Galaxies, star clusters, and other stellar systems are stationary assemblages of stars bound by a self-consistent, mean gravitational potential. Their structure is determined by the nonlinear Poisson equation, which, in most cases, may be solved only numerically. Therefore, any analytical study acquires special interest.  The simplest model of a gravitationally bound stellar system is a spherically symmetric polytrope with a power-law distribution function of stars. The gravitational potential of a polytropic sphere is determined by the Lane--Emden equation: \\beq \\frac{1}{r^2}\\frac{d}{d r}\\left(r^2\\frac{d}{d r}\\psi(r)\\right) =-\\psi^n , \\label{LE} \\eeq where $r$ is the dimensionless spherical radius, $\\psi(r)$ is the dimensionless potential, and $n$ is the polytropic index (see \\citealp{BT} for more discussion). The density is found from $\\rho=c_n\\psi^n$, where $c_n$ is a constant which depends on $n$.  The stellar (collisionless) polytropes are related to hydrostatic equilibria of self-gravitating spheres of a polytropic gas with the equation of state $P=K\\rho^\\gamma$, where $\\gamma=1+\\frac{1}{n}$, $P$ is the gas pressure, and $K$ is a constant. The gas polytropes with $\\gamma<6/5$ (i.e., $n>5$) are usually used to model the structure of molecular clouds which has a major effect on the star formation that occurs within them. Note that the $\\gamma\\to1$ ($n\\to\\infty$) limit corresponds to the case of an isothermal gas. The gravitationally stable polytropes\\footnote{ Locally adiabatic polytropes with polytropic indexes $\\gamma>6/5$ are stable if their adiabatic indexes $\\gamma_{\\rm ad}>4/3$; globally adiabatic polytropes are unconditionally stable for $\\gamma=\\gamma_{\\rm ad}>4/3$, see discussion below and, e.g., \\citet{McKH99}.} with $\\gamma>6/5$ (i.e., $n<5$) often serve as models of stars, because they can have a zero pressure boundary at a finite radius --- the star surface. Interestingly, the Lane-Emden equation appears also as an evolution equation for a scalar field in some versions of the quintessence model \\citep{LS99}. Hence the present study is also of practical interest.  There are two special cases for which nonsingular analytical solutions of equation (\\ref{LE}) are known, namely $n=1$ and $n=5$. We performed a study of (\\ref{LE}) using a powerful mathematical technique --- the Lie group analysis of differential equations \\citep{Ibragimov}. This cumbersome analysis, the details of which are omitted from this publication, yields that, for any $n$ except $n=1$ and $5$, equation (\\ref{LE}) is non-integrable in a closed form.\\footnote{ Strictly speaking, equation (\\ref{LE}) has a zero-dimensional Lie algebra, meaning that there is no a point transformations group admitted by (\\ref{LE}). This does not prove, however, the global non-integrability. A remarkable counter-example is a class of solitonic nonlinear differential equations which ``became'' integrable after the relation of these equations to the B\\\"acklund transformations group have been discovered. In most cases, however, the absence of a Lie algebra often means nonintegrability of a differential equation.} When $n=1$, equation (\\ref{LE}) becomes the linear Helmholtz equation, and when $n=5$, the solution has been discovered by \\citet{Schuster1883} and takes a simple form: \\beq \\psi(r)=\\left(1+\\frac{1}{3}r^2\\right)^{-1/2} . \\label{psi5} \\eeq For other values of $n$, the general solution may be found only numerically.  The present study is motivated by a very peculiar dependence of the radial profiles of the gravitational potential and density on the value of $n$ when $n\\simeq5$. Several solutions of equation (\\ref{LE}) with the ``core'' boundary conditions (i.e., $\\psi(0)=1,\\ \\pd{r}\\psi(0)=0$, where $\\pd{r}\\equiv\\ppd{}{}{r}$) are represented in Figure \\ref{fig1}. A small change in $n$ results in a drastic modification of the profiles. Solutions with $n>5$ exhibit a multiple core structure, --- the fact which has probably been known but never acknowledged in the literature. In this solutions, the regions where $\\psi$ falls as a power-law are interchanged with flat cores where $\\psi\\simeq\\textrm{constant}$. For $n<5$, the solutions are spatially bound, i.e., $\\psi$ and, hence, $\\rho$ vanish at a finite radius. The radius at which the second core begins to form in the $n>5$ solutions is comparable to the radius at which the density vanishes in the $n<5$ solutions.  In this paper we develop a functional series expansion technique which allows the derivation of solutions with the polytropic index $n\\pm|\\epsilon|$, given the solution with the index $n$ (for any $n$ and $\\epsilon,\\ |\\epsilon|\\ll 1$), to any accuracy in powers of $\\epsilon$ and present it in \\S\\ref{S:FORM}. We apply this technique to the $n=5$ solution (\\ref{psi5}) and {\\em analytically} derive a first-order ($\\propto\\epsilon^1$) approximate solution in \\S\\ref{S:5}. This solution perfectly agrees with the exact numerical $n<5$ solutions everywhere, and with the $n>5$ solutions at radii smaller than the second core radius. We also provide a useful approximation for $\\psi(r)$ for $n\\simeq5$ to any power in $\\epsilon$. These results are directly applicable to the polytropic gas systems with $\\gamma\\simeq6/5$. In \\S\\ref{S:STAB} we discuss stability of stellar polytropes. We discuss the results and make some observational predictions in \\S\\ref{S:DISC}. ",
        "conclusions": "}  In this paper we have demonstrated that the radial structure of polytropic self-gravitating spheres is very sensitive to the value of the polytropic index $n$ when $n\\simeq5$ and have investigated it analytically. We developed a method which allows one to obtain approximate solutions, $\\psi_{n+\\epsilon}$, of the Lane-Emden equation, provided the solution $\\psi_n$ is known and $\\epsilon$ is a positive or negative constant, $|\\epsilon|\\ll1$. We applied this method to $n=5$ solution and obtained the first-order (in $\\epsilon$) correction analytically. The obtained solution describes finite-size systems with $n<5$ and the formation of the second core in systems with $n>5$. Both effects occur at the radius $r_*\\simeq32\\sqrt{3}/\\pi|\\epsilon|$. We also provided a continuous fraction approximation for the polytropes with $n>5$, equation (\\ref{fraction}), and a simple approximate formula for those with $n<5$, equation (\\ref{finite}).  We make the observational predictions by calculating  numerically the surface brightness of a stellar polytropic system. A galaxy or a star cluster which is described by a polytrope with $n\\gtrsim5$ will be observed as a bright core surrounded by a very extended faint halo or shell. This halo is cooler than the central core and looks nearly isothermal. Because of the large spatial size and low surface brightness this halo may easily be confused with a background. A way to disentangle the two is to probe a correlation between the core radii, $r_1,\\ r_2$, and (if detected) $r_3$ and the surface brightness distribution.  Due to the duality between the stellar collisionless polytropes and the gas polytropes, our results are immediately applicable to the latter with the adiabatic index $\\gamma=1+\\frac{1}{n}\\simeq\\frac{6}{5}$. Such polytropic gas spheres are used in studies of molecular clouds and star formation. The solutions with $\\gamma<6/5$ describe extended molecular clouds bounded by external pressure. Equation (\\ref{fraction}) describes the multi-core structure of a cloud with $\\gamma\\lesssim6/5$. The density and, hence, pressure is almost constant in core/halo regions. Therefore, if an external pressure, $p_{ext}$, is comparable to the pressure in one of the halos, then even a small variation in $p_{ext}$ will drastically change the radius of the cloud. Thus, a large dispersion of masses and sizes of the clouds embedded into nearly the same environments may indicate that the gas in the cloud has $\\gamma\\lesssim6/5$, although other explanations are also possible. The gravitationally stable polytropes with $\\gamma>6/5$ are often used as models of stars because the pressure vanishes at a finite radius, --- the star's surface. If $\\gamma\\gtrsim6/5$, the radial profile is well described by equation (\\ref{finite}) and the radius of a star is given by equation (\\ref{r*})."
    },
    "0010/astro-ph0010598_arXiv.txt": {
        "abstract": "s{  The detection of the theoretically expected dark matter is central to particle physics and cosmology. Current fashionable supersymmetric models provide a natural dark matter candidate which is the lightest supersymmetric particle (LSP). The theoretically obtained event rates are usually  very low or even undetectable. So the experimentalists would like to exploit the modulation effect. In the present paper we study a specific class of non-isothermal models involving flows of caustic rings. We find that the modulation effect arising from such models is smaller than that predicted by the isothermal models. } ",
        "introduction": "In recent years the consideration of exotic dark matter has become necessary in order to close the Universe  \\cite{Jungm}.  Recent data from the Supernova Cosmology Project  ~\\cite {SPF} $^,$~\\cite {SCP} suggest that the situation can be adequately described by  a barionic component $\\Omega_B=0.1$ along with the exotic components $\\Omega _{CDM}= 0.3$ and $\\Omega _{\\Lambda}= 0.6$ (see also Turner, these proceedings).  Since this particle is expected to be very massive, $m_{\\chi} \\geq 30 GeV$, and extremely non relativistic with average kinetic energy $T \\leq 100 KeV$, it can be directly detected ~\\cite{JDV}$^-$\\cite{KVprd} mainly via the recoiling nucleus.  Using an effective supersymmetric Lagrangian at the quark level, see e.g. Jungman et al ~\\cite{Jungm} and references therein , a quark model for the nucleon ~\\cite{Dree,Chen} and nuclear  wave functions ~\\cite{KVprd} one can obtain the needed detection rates.  They are typically very low. So experimentally one would like to exploit the modulation of the event rates due to the earth's revolution around the sun. In our previous work \\cite{JDV99}$^,$ \\cite{JDV99b} we found anhanced modulation, if one uses appropriate asymmetric velocity distribution. The isolated galaxies are, however, surrounded by cold dark matter , which,  due to gravity, keeps falling continuously on them from all directions \\cite{SIKIVIE}. It is the purpose of our present paper to exploit the results of such a scenario. ",
        "conclusions": "We have calculated the the total event rates for elastic LSP-nucleus scattering including realistic nuclear form factors. We focused our attention on those aspects of the problem, which do not depend on the parameters of supersymmetry other than the LSP mass. The parameter $\\bar {R}$, normally calculated in SUSY theories, was not considered in this work. The interested reader is referred to the literature , for a review \\cite {Jungm} and references therin and, in our notation, to our previous work \\cite {JDV} $^,$ \\cite {KVprd}.  We specialized our results for the target $^{127}I$.  We considered the effects of the detector energy cutoff, by considering  two  typical cases $Q_{min}=10$ and $Q_{min}=20$ KeV. We assumed that the LSP density in our vicinity and the velocity spectrum is that of caustic rings of Sikivie et al \\cite{SIKIVIE}. \\setlength{\\unitlength}{1mm} \\begin{figure} \\vspace*{1.8cm} \\begin{picture}(190,30) \\put(10,0){\\epsfxsize=5cm \\epsfbox{T.ps}} \\put(70,0){\\epsfxsize=5cm \\epsfbox{H.ps}} \\end{picture} \\vspace*{-2.0cm} \\caption[]{The quantities $T(u)$ and $H(u)$ entering the differential amplitude. Thick solid line corresponds to $m_{chi}=30~GeV$ the intermediate thickness line to $m_{chi}=80~ GeV$, the fine line to $m_{chi}=100~GeV$. The rest correspond to larger LSP masses and fall on top of each other.f } \\label{fig.1} \\end{figure}  The total rates are described in terms of  $t$ and $h$. In TABLE I we show how they vary the detector energy cutoff and the LSP mass.  The parameters $T(u)$ and $H(u)$ entering the differential amplitude are shown in Fig. 1. The shape of $T(u)$ is analogous to that of the isothermal models except that the maximum occurs at $u=0.0$, rather than at $u=0.1$. The function $H(u)$ shows oscillations, which result in a smaller total modulation. Another way of understanding how the cancellations arize is to note that for some rings $y_{nz}>1$, while for  others $y{_nz}<1$. \\begin{table}[t] \\caption{The quantities $t$ and $h$ in the case of the target $_{53}I^{127}$ for various LSP masses and $Q_{min}$ in KeV (for definitions see text). } \\begin{center} \\footnotesize \\begin{tabular}{|l|c|rrrrrrr|} \\hline \\hline & & & & & & & &     \\\\ &  & \\multicolumn{7}{|c|}{LSP \\hspace {.2cm} mass \\hspace {.2cm} in GeV}  \\\\ \\hline & & & & & & & &     \\\\ Quantity &  $Q_{min}$  & 10  & 30  & 50  & 80 & 100 & 125 & 250   \\\\ \\hline & & & & & & & &     \\\\ t    &0.0&1.451& 1.072& 0.751& 0.477& 0.379& 0.303& 0.173\\\\ h    &0.0&0.022& 0.023& 0.024& 0.025& 0.026& 0.026& 0.026\\\\ \\hline & & & & & & & &     \\\\ t    &10.0&0.000& 0.226& 0.356& 0.265& 0.224& 0.172& 0.098\\\\ h    &10.0&0.000& 0.013& 0.023& 0.025& 0.025& 0.026& 0.026\\\\ \\hline & & & & & & & &     \\\\ t    &20.0&0.000& 0.013& 0.126& 0.139& 0.116& 0.095& 0.054\\\\ h    &20.0&0.000& 0.005& 0.017& 0.024& 0.025& 0.026& 0.026\\\\ \\hline \\hline \\end{tabular} \\end{center} \\end{table}  The maximum occurs around the 2nd of December, something already noticed by Sikivie et al \\cite{SIKIVIE}. Furthermore the modulation is small, $h=0.25$, i.e. a $5\\%$ difference betwen the maximum and the minimum (see TABLE I). It is a bit smaller than that of the symmetric models, but a lot smaller than that predicted by the  asymmetric ones \\cite{JDV99,JDV99b}, i.e $h=0.46$"
    },
    "0010/astro-ph0010551_arXiv.txt": {
        "abstract": "We present near-infrared J (1.25$\\mu$m), H (1.65$\\mu$m), and K$_s$ (2.17$\\mu$m) imaging of the entire supernova remnant IC 443 from the Two Micron All Sky Survey (2MASS), and Infrared Space Observatory (ISO) LWS observations of [O~I] for 11 positions in the northeast. Near-infrared emission from IC 443 was detected in all three bands from most of the optically bright parts of the remnant, revealing a shell-like morphology, with bright K$_s$ band  emission along the southern ridge and bright J and H along the northeastern rim. The total luminosity within the 2MASS bands is 1.3$\\times$ 10$^{36}$ erg s$^{-1}$. These data represent the first near-infrared images that are complete in coverage of the remnant.  The color and morphological structure are very different between the northeastern and southern parts.  J and H band emission from the northeast rim is comparably bright and  can be explained mostly by [Fe~II] line emission.  The hydrogen recombination lines, P$\\beta$ and Br10, should also be present in the broad-band images, but probably contribute less than 10\\% of the J and H band fluxes.  Strong [O~I] (63$\\mu$m) lines  were detected crossing the northeastern rim, with the strongest line in the northeastern shell where the near-infrared emission shows filamentary structure. In contrast, the southern ridge is dominated by K$_s$ band light exhibiting a clumped and  knotty structure. A two excitation temperature model derived from previous ISO and ground-based observations predicts that H$_2$ lines can explain most of K$_s$ band and at least half of J and H band emission. Hence, the prominent broad-band color differences arise from physically different mechanisms: atomic fine structure lines along the northeastern rim and molecular ro-vibrational lines along the southern ridge. Shock models imply a fast J-shock with v$_s$ $\\sim$ 100 km s$^{-1}$ and 10$<$n$_o$ $<$10$^3$ cm$^{-3}$ for the northeastern rim and  a slow C-shock with  v$_s$ $\\sim$ 30 km s$^{-1}$ and n$_o$ $\\sim$ 10$^{4}$ cm$^{-3}$ for the southern ridge, respectively.  The shocked H$_2$ line emission  is well known from the southern sinuous ridge, produced by an interaction with dense molecular clouds. The large field of view and color of the 2MASS images show that the K$_s$ band emission extends to the east and the northeast, suggesting  that the interaction extends to the inner part of the northeastern shell. Our new CO map of the inner part of the northeast quadrant shows good correspondence with the K$_s$-band map. The CO lines are broad, confirming that the K$_s$-band emission is due to shocked H$_2$.   \\noindent{\\it Subject headings}: infrared:shock waves - nebula:individual (IC 443) - ISM:supernova remnants ",
        "introduction": "IC 443 is the best-studied case of supernova remnant-molecular cloud interaction. The system is observed to have broad molecular lines with CO gas accelerated to 20 km s$^{-1}$ and  shocked H$_2$ emission along the southern ridge (van Dishoeck et al. 1993; Burton et al. 1988). The molecular cloud stretches across the remnant from the north to the south as projected along the line of sight  (Cornett, Chin \\& Knapp 1977). IC 443 is in the Gem OB 1 association, located close to the Galactic plane (G189.1+3.0). Optical spectroscopy of the northeastern filaments of IC 443 indicated a shock velocity of v$_s$$\\sim$65-100 km s$^{-1}$ and a preshock density of n$_o$$\\sim$10-20 cm$^{-3}$ (Fesen \\& Kirshner 1980). Shocked H~I is associated with the bright northeastern filaments and extends across the entire southern hemisphere of the remnant with velocities up to 100 km s$^{-1}$  (Giovanelli \\& Haynes 1979).  IC 443 has unusual morphology; the optical and radio morphology is shell-like, but it extends eastward beyond the main shell (referred to the ``northeastern arm\"). The optical emission shows radial filaments with different curvature in the southwest, and the X-ray emission appears both at the interior of the shell and at the edge of the shell (Petre et al. 1988; Rho et al. 1993). A superposition with another SNR G189.6+3.3 has been suggested (Asaoka \\& Aschenbach 1994).  The presence of H$_2$ line emission along the southern ridge strongly suggests that the supernova shock wave is interacting with the molecular cloud  (Burton et al. 1988; Richter et al. 1995).  A near-infrared line of [Fe~II] (1.64$\\mu$m) is detected from    the northeastern shell, 30 times as strong as that of Br$\\gamma$, which is much stronger than seen in Galactic H~II regions. The strong [Fe~II] line is produced not only by efficient excitation of Fe but also by grain destruction (Graham et al. 1987).  Infrared emission from supernova remnants, including IC 443, is not well understood.  IRAS observations show bright emission from IC 443, which  was interpreted as thermal continuum emission from heated small and large dust grains  (Arendt 1989; Saken et al. 1992; Braun \\& Strom 1986). However, recent ISO observations show no continuum between 5-14$\\mu$m with a noise level of 0.2$\\times$$10^{-4}$  erg s$^{-1}$ cm$^{-2}$ sr$^{-1}$ (Cesarsky et al. 1999). Instead, the observed mid-infrared spectrum of the shocked gas is dominated by rich emission lines of H$_2$. ISO SWS observations of the northeastern filaments (Oliva et al. 1999a) also showed detections of rich emission lines, not molecular but ionic lines such as [Ne~II], [Fe~II] and [Si~II]. Previous interpretations of the IRAS 12 and 25$\\mu$m emission are different, based on differences between the ISO SWS and ISOCAM results. In particular, Oliva et al. (1999a) suggested that 25$\\mu$m IRAS emission is virtually all ionized ion lines, while Cesarsky et al (1999) claimed that shocked H$_2$ lines  (at least at the southern region) are dominant. This is due to limited field of view of ISO observations toward IC 443, and also due to limited spatial resolution of the IRAS image.  In this paper, we present J, H, and K$_s$ imaging from the Two Micron All Sky Survey (2MASS, Skrutskie et al. 1997) covering the entire size of the IC 443 remnant, and we show which spectral lines contribute to the J, H, and K$_s$ band emission. We present detections of the [O~I] 63$\\mu$m line using ISO LWS observations. Combining the 2MASS and ISO data,  we employ  diagnostics for rating the cooling in atomic or molecular shocks, and compare the line brightnesses with  J and C shock models.  While previous near-infrared imaging and spectroscopic observations were limited to a part of the supernova remnant, the 2MASS image offers a global view of the emission, and it provides a means to understand the surrounding ISM environment. IC 443 is the first such example, and  we  expect  other supernova remnants will be revealed in the near-infrared light using 2MASS. ",
        "conclusions": ""
    },
    "0010/astro-ph0010284_arXiv.txt": {
        "abstract": "From optical R band data of the double quasar QSO 0957+561A,B, we made two new difference light curves (about 330 days of overlap between the time-shifted light curve for the A image and the magnitude-shifted light curve for the B image). We observed noisy behaviours around the zero line and no short-timescale events (with a duration of months), where the term event refers to a prominent feature that may be due to microlensing or another source of variability. Only one event lasting two weeks and rising - 33 mmag was found. Measured constraints on the possible microlensing variability can be used to obtain information on the granularity of the dark matter in the main lensing galaxy and the size of the source. In addition, one can also test the ability of the observational noise to cause the rms averages and the local features of the difference signals. We focused on this last issue. The combined photometries were related to a process consisting of an intrinsic signal plus a Gaussian observational noise. The intrinsic signal has been assumed to be either a smooth function (polynomial) or a smooth function plus a stationary noise process or a correlated stationary process. Using these three pictures without microlensing, we derived some models totally consistent with the observations. We finally discussed the sensitivity of our telescope (at Teide Observatory) to several classes of microlensing variability. ",
        "introduction": "\\subsection{Microlensing caused by the Galaxy and other spirals}  Dark matter dominates the outer mass of the Milky Way. In principle, the population of the Galactic dark halo may include astrophysical objects as black holes, brown dwarfs, cool white dwarfs, etc., i.e., MACHOs (massive compact halo objects) with stellar or substellar mass, as well as elementary particles (a smooth component). Today, from microlensing surveys, we have some information about the granular component (MACHOs). The absence of very short duration events implies that the dark halo cannot be dominated by planetary objects. A joint analysis by {\\it EROS} and {\\it MACHO} collaborations indicated that MACHOs in the mass range $10^{-7} M_{\\odot} \\leq M \\leq 10^{-3} M_{\\odot}$ make up less than 25\\% of the dark halo (Alcock et al. 1998). From a likelihood analysis, the {\\it MACHO} collaboration concluded that a population of objects of mass $\\sim$ 0.5 $M_{\\odot}$ is consistent with their first two year of data. These MACHOs with stellar mass would have an important contribution to the total mass (Alcock et al. 1997; Gould 1997; Sutherland 1999; Mao 2000). However, very recent results by the {\\it MACHO} team, based on approximately six years of observations, point to a relatively small mass fraction (Alcock et al. 2000). For a typical size halo, the maximum likelihood estimates suggest the existence of a Milky Way dark halo consisting of about 20\\% MACHOs with mass of $\\sim$ 0.6 $M_{\\odot}$. The {\\it EROS} collaboration also agrees with this small contribution to the dark halo by $\\sim$ 0.6 $M_{\\odot}$ objects (Lasserre et al. 2000). Lasserre et al. (2000) derived strong upper limits on the abundance of MACHOs with different masses. For example, $<$ 10\\% of the dark halo resides in planetary objects. Moreover, they ruled out a standard spherical halo in which more than 40\\% of its mass is made of dark stars with 1 $M_{\\odot}$. Finally, we remark that the Milky Way dark halo inferred from the likelihood method (best standard fits by Alcock et al. 2000) is consistent with the {\\it HST} (Hubble Space Telescope) detection of a halo white dwarf population (Ibata et al. 1999). A population of cool white dwarfs contributing 1/5 of the dark matter in the Milky Way could explain all new observational results, but the hypothesis presents some difficulties (e.g., Mao 2000; Alcock et al. 2000).  The information on the nature of galaxy dark haloes is still largely based on a local spiral galaxy (the Galaxy), and so, the study of other galaxies seems an interesting goal.  The Einstein Cross (QSO 2237+0305) is a $z$ = 1.69 quasar lensed by a face-on barred Sb galaxy at $z$ = 0.0394. The time delay between the four quasar images is expected to be less than a day (Rix et al. 1992; Wambsganss \\& Paczy\\'nski 1994), and so, one can directly separate intrinsic variability from microlensing signal. For this lens system, light rays of the 4 images pass through the bulge of the foreground galaxy and there is a robust evidence that microlensing events occur (e.g., Irwin et al. 1989; Wozniak et al. 2000). The observed events may be interpreted as a phenomenon caused by the granularity of the matter associated with the nearby spiral. Very recently, for providing an interpretation of the {\\it OGLE} Q2237+0305 microlensing light curve, Wyithe, Turner \\& Webster (2000) used the contouring technique of Lewis et al. (1993) and Witt (1993).  B1600+434 is another interesting gravitational mirage lensed by an edge-on disk galaxy. Koopmans \\& de Bruyn (2000) measured the radio time delay between the two images of the system and derived a radio difference light curve which is in disagreement with zero. They investigated both scintillation and microlensing as possible causes of the non-intrinsic radio variability. If microlensing is the origin of the \"anomalous\" difference light curve, then it could indicate the presence of a lens galaxy dark halo filled with MACHOs of mass $\\geq 0.5 M_{\\odot}$.  \\subsection{Microlensing in the first gravitational lens system (Q0957+561)}  A third well-known microlensed quasar is the $z$ = 1.41 double system Q0957+561A,B. The main lens galaxy is an elliptical galaxy (cD) at $z$ = 0.36. While the light associated with the image B crosses an internal region of the lens galaxy, the light path associated with the component A is $\\approx$ 5 arcsec away from the centre of the galaxy. The cD galaxy is close to the centre of a galaxy cluster, and consequently, the normalized surface mass densities $\\kappa_A$ and $\\kappa_B$ are the projected mass densities of the lensing galaxy plus cluster along the lines of sight, normalized by the critical surface mass density. Pelt et al. (1998) used the recent values $\\kappa_A$ = 0.22 and $\\kappa_B$ = 1.24, which originate from an extended galaxy halo consisting of the elliptical galaxy halo and an additional matter related to the cluster. It is possible that a considerable part of the extended halo mass does consist of a dark component, although an estimate of the stellar contribution (luminous stars) to $\\kappa_A$ and $\\kappa_B$ is not so easy as in the Milky Way. For the image B, if the fraction of mass in granular form $\\kappa_{BG}$ is dominated by normal stars and dark stars similar to the objects that have been discovered in the Galaxy (Alcock et al. 2000), and simultaneously, the main part of the halo mass is due to a smooth component ($\\kappa_{BG} << \\kappa_B$, $\\kappa_{BG} <<$ 1) and the source quasar is small, then we must expect some long-timescale microlensing event caused by one star (luminous or dark) crossing the path of this image. In that case of small source/one star approximation, the timescale of an event will be $t_o$(years) $\\approx$ 17 $\\sqrt{M(M_{\\odot})}[600/v_t(km s^{-1})]$, where $v_t$ is the transverse velocity, and the magnification of the B component has a typical duration of several years for a 0.5-1 $M_{\\odot}$ star and any reasonable choice of $v_t$. When $\\kappa_{BG}$ is high ($\\kappa_{BG} \\sim$ 1) and/or the source is large, several stars at a time must be considered and the model by Chang \\& Refsdal (1984) is not suitable. The small source/one star model by Chang \\& Refsdal (1984) was generalized in the case of a small source and a large optical depth (Paczy\\'nski 1986) and the case of an extended source and an arbitrary optical depth (Kayser, Refsdal \\& Stabell 1986; Schneider \\& Weiss 1987; Wambsganss 1990). Therefore, the formalisms by Chang \\& Refsdal (1984), Paczy\\'nski (1986) and Wambsganss (1990) as well as new analytical approximations seem useful tools for a detailed analysis of the optical microlensing history of QSO 0957+561. A long-timescale microlensing signal was unambiguously observed from 1981 to 1999; see Pelt et al. (1998), Press \\& Rybicki (1998), Serra-Ricart et al. (1999, subsequently Paper I). In this paper, we concentrate on the possible rapid microlensing signal. In a forthcoming paper, we will carry out a comparison between the annual differences (averages from January to December) $B - A$ and the predictions from different models and physical parameters.  In the past, using a record of brightness including photometric data (in the R band) up to 1995 and a time delay of 404 days, Schild (1996, hereafter S96) analyzed the possible existence of short-timescale microlensing (rapid external variability on a timescale of months) and very rapid microlensing events (with duration of $\\leq$ 3 weeks) in the double QSO 0957+561A,B. He found numerous events with quarter-year and very short timescales (a few days). S96 also claimed that the slower component (events with a width of 90 days) can be interpreted as the imprint of an important population of microlenses with planetary mass of $\\sim 10^{-5} M_{\\odot}$. Assuming an improved delay value of 417 days, Goicoechea et al. (1998, subsequently G98) showed a difference light curve corresponding to the 1995/1996 seasons in Schild's dataset. G98 obtained fluctuations which could be associated with microlensing events, in fact, our results are in agreement with the existence of strong microlensing: the fluctuations in the difference light curve are clearly inconsistent with zero and similar to the fluctuations in the quasar signal. New work by Schild and collaborators pointed in the same direction: adopting a time delay of 416.3 days, Pelt et al. (1998) found that Schild's photometry shows evidence in favour of the presence of short-timescale microlensing; Schild (1999) made a wavelet exploration of the QSO 0957+561 brightness record, and reported that the rapid brightness fluctuations observed in the A and B quasar images (whose origin may be some kind of microlensing) are not dominated by observational noise; and Colley \\& Schild (1999), from a new reduction of \"old\" photometric data (subtracting out the lens galaxy's light according to the {\\it HST} luminosity profile and removing cross talk light from the A and B images apertures), derived a structure function for variations in the R-band from lags of hours to years, a time delay of 417.4 days and a microlensing candidate on a timescale of a day, which could imply planetary MACHOs in the lens galaxy halo. So, from the photometry taken at Whipple Observatory 1.2 m telescope by Schild, one obtains two important conclusions. First, there is evidence in favor of the existence of true short-timescale microlensing signal. Second, this rapid signal seems to support the presence of MACHOs (in the halo of the cD galaxy) having a very small mass. However we note that Gould \\& Miralda-Escud\\'e (1997) have introduced an alternative explanation to the possible rapid microlensing in the double QSO 0957+561A,B, which is related to hot spots or other moving structures in the accretion disk in the quasar, and so, planetary objects are not involved.  QSO 0957+561A,B was photometrically monitored at Apache Point Observatory (Kundi\\'c et al. 1995, 1997) in the g and r bands, during the 1995 and 1996 seasons.  Schmidt \\& Wambsganss (1998, hereafter SW98) analyzed this photometry and searched for a microlensing signature. Considering the photometric data in the g band and a delay of 417 days, SW98 produced a difference light curve covering $\\approx$ 160 days and concluded that it is consistent with zero. There is no variation in the difference light curve with an amplitude in excess of $\\pm$ 0.05 mag and the total magnitude variation of a hypothetical microlensing signal is assumed to be less than 0.05 mag (see the dashed lines in Fig. 1 of SW98). They employed this last upper limit to obtain interesting information on the parameter pair MACHO-mass/quasar-size. The lack of observed fluctuations rules out a population of MACHOs with $M \\leq$ 10$^{-3}$ $M_{\\odot}$ for a quasar size of 10$^{14}$ cm (25\\%-100\\% of the matter in compact dark objects). However, other possible scenarios (e.g., a small source and a halo consisting of MACHOs with $M \\geq$ 10$^{-2}$ $M_{\\odot}$, a source size of 10$^{15}$ cm and a halo with compact dark objects of mass $\\leq$ 10$^{-3}$ $M_{\\odot}$, etc.) cannot be ruled out from the bound on the microlensing variability in the 160 days difference light curve. In short, SW98 have not found reliable evidence for the presence of rapid microlensing events.  The gravitational lens system Q0957+561 was also monitored by our group with the IAC-80 telescope at Teide Observatory, from the beginning of 1996 February to 1998 July (see Paper I). We re-reduced our first 3 seasons (1996-1998) of QSO 0957+561 observations in the R band, made the difference light curves for 1996/1997 seasons and 1997/1998 seasons and studied the origin of the deviation between the light curves of the two images. All the results are presented in this article. The plan of the paper is as follows: in Sect. 2 we present the difference light curves and report on new constraints on microlensing variability. In Sect. 3 we suggest different models that explain the difference signal. In Sect. 4 we discuss the sensitivity of the telescope to different microlensing \"peaks\". At the end of the paper (Sect. 5), we summarize our results. ",
        "conclusions": "Several $\\sim$ 1 metre class telescopes around the world are at present involved in different optical monitoring programs of quasars with the goal to detect microlensing. There are at least two \"modest\" telescopes searching for microlensing signal related to a far elliptical galaxy (which is responsible, in part, for the gravitational mirage Q0957+561). The data taken at Whipple Observatory 1.2 m telescope and at Teide Observatory IAC-80 telescope together with the photometry from a 3.5 m telescope (at Apache Point Observatory) represent a great effort in order to obtain an accurate time delay in Q0957+561, follow the long-timescale microlensing event in that system and find some evidence in favour of very rapid and rapid microlensing (Kundi\\'c et al. 1995, 1997; Oscoz et al. 1996, 1997; Pijpers 1997; Schild \\& Thomson 1997; Paper I; S96; Pelt et al. 1998; SW98; G98).  With respect to the very rapid (events with a timescale $\\leq$ 3 weeks) and rapid (events with a duration of 1-4 months) microlensing, the previous results (before this article) are puzzling. The combined photometries (CLCs) from data taken at Whipple Observatory only can be well explained in the context of a picture including intrinsic variability, observational noise and microlensing variability on different timescales: from days to months (e.g., S96). The long-timescale microlensing does not play any role in a CLC. In particular, S96 reported on the existence of a network of rapid events with a few months timescale and an amplitude of about $\\pm$ 50 mmag (these features found by Schild are called Schild-events). However, SW98 concluded that a picture with intrinsic signal and observational noise (without any need to introduce very rapid and rapid microlensing) is consistent with the observations at Apache Point Observatory. SW98 really show a difference light curve in global agreement with the zero line, but some doubt remains on the ability of the observational noise for producing the negative and positive measured events around \"peaks\" (a \"peak\" is constituted by a set of two or more consecutive deviations which have equal sign and are not consistent with zero). In any case, SW98 observed no Schild-events.  In this paper, motivated by the mentioned intriguing results on microlensing variability, we analyzed the data from our initial monitoring program with the IAC-80 telescope (see Paper I). We focused on the possible presence of rapid microlensing events in the light curves of QSO 0957+561 and the sensitivity of the telescope (using typical observational and analysis procedures) to microlensing \"peaks\". Our conclusions are:  \\begin{enumerate} \\item Using photometric data (in the R band) for the 1996-1998 seasons, we made two difference light curves (DLCs). The total difference signal, which is based on $\\sim$ 1 year of overlap between the time-shifted light curve for the A component and the magnitude-shifted light curve for the B component, is in apparent agreement with the absence of microlensing signal. We can reject the existence (in our DLCs) of events with quarter-year timescale and an amplitude of $\\pm$ 50 mmag, and therefore, Schild-events cannot occur almost continuously. On the contrary, they must be either rare phenomena (originated by microlensing or another physical process) or, because two observatories (Apache Point Observatory and Teide Observatory) found no Schild-events, untrue fluctuations associated with the observational procedure and/or the reduction of data at Whipple Observatory. \\item From a very conservative point of view, in our data, the amplitude of any hypothetical microlensing signal should be in the interval [- 50 mmag, + 50 mmag]. The rms averages of the DLCs (global variability) are of about 22 mmag (1996/1997 seasons) and 15 mmag (1997/1998 seasons), and reasonable constraints on the possible microlensing variability lead to interesting information on the granularity of the dark matter in the main lensing galaxy (a cD elliptical galaxy) and the size of the source (QSO). Thus the set of bounds derived from 1995-1998 seasons (SW98 and this work) rules out an important population of MACHOs with substellar mass for a small quasar size (Schmidt 1999). \\item In order to settle any doubt on the ability of the observational noise for generating the global (rms averages) and local (events and other less prominent features) properties of the DLCs, we have also carried out several experiments as \"Devil's advocates\". The measured variability (the rms value, a very rapid event and some minor deviations) in the DLC for 1996/1997 seasons can be caused, in a natural way, by the observational noise process. In the absence of microlensing signal, we proposed two different models (M1 and M2; see subsection 3.1) whose associated photometries (simulations) are consistent with the observations. In addition, the DLC for 1997/1998 seasons is a quasi-featureless trend with relatively small rms average. To explain the variability in our second observational DLC, we again showed that microlensing is not necessary. Two new models (M3 and M4; see subsection 3.2) only including the reconstruction of the intrinsic signal (assumed as a polynomial or a correlated stationary process) and a Gaussian observational noise process, led to simulated DLCs in agreement with the measured behaviour. \\item We finally show that from a typical monitoring with our telescope (observing times, method of analysis, etc.) is not possible the resolution of $\\leq$ 20 days microlensing \"peaks\". The confusion with noise does not permit the separation between true microlensing features and \"peaks\" due to the observational noise. However, all hypothetical \"peaks\" with a timescale $>$ 40 days must be interpreted as phenomena which are not associated with the observational noise (e.g., microlensing fluctuations). At intermediate timescales (of about one month) the situation is somewhat intricate. Given a measured DLC, the probability of observing one noise \"peak\" (with a duration of about 30 days) is less than 10\\%. Therefore, if we search for microlensing signal and find an \"intermediate peak\", the relative probabilities that the fluctuation is a noise feature or a microlensing \"peak\" are $<$ 1:10. \\end{enumerate}"
    },
    "0010/astro-ph0010001_arXiv.txt": {
        "abstract": "We present the results of UV spectropolarimetry ($\\lambda_{\\rm rest} \\simeq 2000 - 3000$\\AA) and far-UV spectroscopy ($\\lambda_{\\rm rest} \\simeq 1500 - 2000$\\AA) of two low-redshift narrow-line radio galaxies (NLRGs) taken with the Faint Object Spectrograph onboard the Hubble Space Telescope (HST).  Spectropolarimetry of several NLRGs has shown that, by the presence of broad permitted lines in polarized flux spectrum, they have hidden quasars seen through scattered light. Imaging polarimetry has shown that NLRGs including our targets often have large scattering regions of a few kpc to $\\gtrsim 10$ kpc scale. This has posed a problem about the nature of the scatterers in these radio galaxies.  Their polarized continuum has the spectral index similar to or no bluer than that of quasars, which favors electrons as the dominant scattering particles. The large scattering region size, however, favors dust scattering, because of its higher scattering efficiency compared to electrons.  In this paper, we investigate the polarized flux spectrum over a wide wavelength range, combining our UV data with previous optical/infrared polarimetry data.  We infer that the scattering would be often caused by opaque dust clouds in the NLRGs and this would be a part of the reason for the apparently grey scattering.  In the high-redshift radio galaxies, these opaque clouds could be the proto-galactic subunits inferred to be seen in the HST images.  However, we still cannot rule out the possibility of electron scattering, which could imply the existence of a large gas mass surrounding these radio galaxies. ",
        "introduction": "At least some part of the population of narrow-line radio galaxies (NLRGs) are believed to harbor quasars in their central region, hidden from direct view. This has been demonstrated by polarimetric observations of these radio galaxies.  The observations have shown that the radiation from some of these NLRGs is linearly polarized with the E-vector perpendicular to the radio-jet axis. Also the permitted lines are broad in the polarized flux spectra \\citep{An94,Yo96a,Og97,TCG95,Yo98,Co99}, the same as in Seyfert 2 galaxies \\citep{AM85,Tr92,Ka94,Tr95}. In some cases, the broad components are discernable in total flux, but highly polarized and of low equivalent width.  The polarization degree often goes higher in the UV than in the optical due to the lower dilution by the unpolarized star light from the host galaxies.  Therefore optical, rest-UV polarimetric observations of high-redshift radio galaxies have been done extensively \\citep{Ci93,di94,di96,Ci96,De96,Ci97,Ci98b,Tr98}. The imaging polarimetry has shown that the rest-UV image is typically elongated along the radio axis, and polarized perpendicular to this elongation \\citep{di93,Tr98}. Recently, this has been also domonstrated by UV imaging polarimetry of nearby radio galaxies \\citep{Hu99}.  Thus, all these facts fit well into the Unified Model of radio-loud quasars and radio galaxies \\citep{Ba89}.  The scattered light, together with the anisotropic nuclear radiation along the radio axis, has provided the explanation of at least some part of the alignment between the UV morphology and radio jet structure in $1 \\lesssim z \\lesssim 2$ galaxies (called the alignment effect; di Serego Alighieri et al. 1989; see McCarthy 1993 for a review). In the two $z\\sim4$ galaxies observed, however, the aligned light is unpolarized starlight (4C41.17, Dey et al. 1997; 6C1909+722, Dey 1998).  While the scattering origin for much of the UV radiation has now been proven in several NLRGs, the nature of the scattering agents, i.e. hot/warm electrons and/or dust grains, still remains to be debated. The imaging polarimetry and long-slit polarimetry have revealed that the size of the scattering region in some NLRGs is of a few kpc to $\\gtrsim$ 10 kpc scale \\citep{Ci96,De96,Tr98,Hu99}. This is one or more orders of magnitude larger than found in Seyfert galaxies (e.g. Capetti et al. 1995). This large size implies a rather enormous mass for the scattering gas in the NLRGs if the scatterers are electrons \\citep{di94,Ci96,De96}. Therefore dust grains are preferred a priori to be the scattering agent, because of their much better scattering efficiency (i.e., larger scattering cross section per unit gas mass).  To address the nature of the scattering regions and environment around the hidden quasars in the radio galaxies, we need a wide wavelength coverage for the polarized flux distribution.  In this paper, we present UV spectropolarimetry of low-redshift narrow-line radio galaxies obtained by the Hubble Space Telescope (HST). Combining the data with other optical/IR polarimetric data from the literature, we seek the general characteristics of the observed scattering regions. In \\S \\ref{sec-obs}, we describe our observation and data reduction process, and the results are presented in \\S \\ref{sec-res}. We discuss the nature of the scattering region in \\S \\ref{sec-disc}, and summarize our conclusions in \\S \\ref{sec-conc}. We adopt $H_0 = 50$ km/sec/Mpc and $q_0 = 0.5$ throughout this paper. ",
        "conclusions": "\\label{sec-conc}  We have presented the results of HST UV spectropolarimetry of low-redshift narrow-line radio galaxies.  We investigated the polarized flux spectrum over a wide wavelength range, combining the UV data with the optical/IR data in the literature.  The polarized flux spectrum shape in our sample and in other high-redshift radio galaxies are found to be no bluer than the typical quasar continuum. This, and apparently no exponential decrease in the shorter wavlength, would suggest that in a lot of cases the dust scattering clouds, if any, would be opaque and thus show a grey scattering efficiency. In the high-redshift galaxies, these opaque dust scattering clouds could be proto-galactic fragments inferred to be seen in the HST images. However, the possibility of the scattering by electrons still cannot be ruled out, which might imply a large gas mass surrounding these radio galaxies."
    },
    "0010/gr-qc0010100_arXiv.txt": {
        "abstract": "A typical stellar mass black hole with a lighter companion is shown to succumb to a chaotic precession of the orbital plane. As a result, the optimal candidates for the direct detection of gravitational waves by Earth based interferometers can show irregular modulation of the waveform during the last orbits before plunge. The precession and the subsequent modulation of the gravitational radiation depends on the mass ratio, eccentricity, and spins.  The smaller the mass of the companion, the more prominent the effect of the precession. The most important parameters are the spin magnitudes and misalignments.  If the spins are small and nearly aligned with the orbital angular momentum, then there will be no chaotic precession while increasing both the spin magnitudes and misalignments increases the erratic precession. A large eccentricity can be induced by large, misaligned spins but does not seem to be required for chaos. An irregular precession of the orbital plane will generate irregularities in the gravitational wave frequency but may have a lesser effect on the total number of cycles observed. ",
        "introduction": "The Post-Newtonian (PN) expansion of the relativistic two body problem leads to a system of equations describing the fate of spinning binaries \\cite{pn}. The PN expansion converges slowly to the fully relativistic description \\cite{kww}.  For this reason, it is a poor approximation at small separations.  Despite its shortcomings, the PN expansion does give the qualitative features of a relativistic system such as nonlinearity, the existence of unstable circular orbits \\cite{kww}, homoclinic orbits \\cite{loc}, and spin precession.  Since these are the ingredients for chaotic dynamics, the qualitative behaviour should persist in a more accurate approximation, although the quantitative conclusions are subject to change (see for instance the improved technique of Ref.\\ \\cite{damour}).  It is worth emphasizing that approximations can introduce chaos when the exact system is truly regular.  One might worry that the error at 2PN order has introduced chaos which would be removed if we knew the full equations of motion without approximation.  However, the relativistic two-body problem is likely to be more irregular at higher orders as the nonlinearities of general relativity are more accurately represented, not less. One might even be inclined to take the extreme resistance of the relativistic two-body problem to solution as evidence, or at least confirmation, of nonintegrability.  The validity of the PN expansion is not questioned further and the equations are treated as a self-contained dynamical system. In the PN scheme, the orbit evolves according to the force equation \\be \\ddot{\\bf \\vec r }={\\bf \\vec a_{PN}} +{\\bf \\vec a_{SO}}+{\\bf \\vec a_{SS}}+{\\bf \\vec a_{RR}} \\label{eom1} \\ee in center of mass harmonic coordinates \\cite{kidder}. The acceleration is due to Post-Newtonian (PN) effects, spin-orbit (SO) and spin-spin (SS) coupling and radiative reaction (RR). The explicit form of ${\\bf \\vec a}$ can be found in the appendix. The spins also precess due to the relativistic frame dragging and Lens-Thirring effect.  The precession equations are \\be \\dot{\\bf \\vec S_1} = {\\bf \\vec \\Omega_1}\\times {\\bf  \\vec S_1 } \\quad  , \\quad \\dot{\\bf \\vec S_2 }={\\bf  \\vec \\Omega_2}\\times {\\bf  \\vec S_2} \\label{prec1}\\ee with \\begin{eqnarray} {{\\bf \\vec \\Omega_1}} = {1 \\over r^3} \\biggl [ && \\left (2+{3 \\over 2} {m_2 \\over m_1}\\right) {\\bf \\vec  L_N} -{\\bf \\vec  S_2 } + 3({ {\\bf \\hat n} \\cdot {\\bf \\vec S_2}}) { {\\bf \\hat n} } \\biggr ] , \\end{eqnarray} and \\begin{eqnarray} {{\\bf \\vec \\Omega_2}} = {1 \\over r^3} \\biggl [ && \\left (2+{3 \\over 2} {m_1 \\over m_2}\\right) {\\bf \\vec  L_N } -{\\bf \\vec  S_1} + 3({ {\\bf \\hat n} \\cdot {\\bf \\vec S_1}}) { {\\bf \\hat n} } \\biggr ]. \\end{eqnarray} The spins precess with constant magnitude although the total spin ${\\bf \\vec S}={\\bf {\\vec S}_1}+{\\bf {\\vec S}_2}$ may not have constant magnitude.  \\begin{figure} \\centerline{\\psfig{file=angles.eps,width=2.5in,angle=-90}} \\vskip 25truept \\caption{ A schematic drawing of the inclination angle $i={\\rm arccos} \\left ( {\\bf \\hat L_N}\\cdot {\\bf \\hat J}\\right )$ and the angle $\\theta_S={\\rm arccos}\\left ({\\bf \\hat L_N}\\cdot{\\bf \\hat S}\\right )$. The orbital plane traces out a band as in figs. \\ref{m.14r20} and \\ref{m.3r20} as the Newtonian angular momentum precesses about ${\\bf \\vec J}$. \\label{angles}}  \\end{figure}  The orientation of the orbital plane is defined by the Newtonian orbital angular momentum \\be {{\\bf \\vec  L_N}} \\equiv \\mu ({{\\bf \\vec  r} \\times {\\bf \\vec v}}) \\label{lnewt} \\ee with the reduced mass $\\mu=m_1m_2/m$ and the total mass $m=m_1+m_2$. Spin precession, generates a precession of the orbital plane (fig.\\ \\ref{angles}).  This can most easily be seen by noting that to 2PN-order the total angular momentum ${\\bf \\vec J}$ is conserved with ${\\bf \\vec J}={\\bf \\vec L}+{\\bf \\vec S}$. The orbital angular momentum  ${\\bf \\vec L}$ can be split into two pieces, ${\\bf \\vec L}=\\alpha {\\bf \\vec L_N}+{\\bf \\vec L_{SO}}$ as in eqns.\\ (\\ref{split}) and (\\ref{split3}).  The term ${\\bf \\vec L_{SO}}$ is due to spin-orbit coupling and $\\alpha $ contains Post-Newtonian corrections. To 2PN order $\\dot {{\\bf \\vec J}}=0$ and so \\be \\dot {\\bf \\vec  L}\\sim -\\dot {\\bf \\vec S}. \\ee The magnitude of the orbital angular momentum is {\\it not} constant. Further, the precession of the orbital plane can be much more complicated than the precession of ${\\bf \\vec S}$: $\\left (\\alpha {\\bf \\vec L_N}\\right )^{\\dot {}}\\sim -\\dot {\\bf \\vec S}- \\dot {\\bf \\vec L_{SO}}$. The orbital plane therefore does not just carve out a simple cone as it precesses around the direction of ${\\bf \\vec J}$.  Instead the plane tilts back and forth as it precesses. The motion becomes chaotic as the tilting back and forth becomes highly irregular in the Hamiltonian system. When radiation reaction is included, the degree of irregularity in the precession of the orbital plane can be estimated in terms of how many windings the pair spends near the chaotic region of the underlying conservative system.  Fig.\\ \\ref{m.14r20} shows an example of the trajectory for the center of mass of a BH/NS pair.  The motion is plotted in three-dimensions to illustrate the precession of the orbital plane.  \\begin{figure} \\centerline{\\psfig{file=3dm.14r20_compare.ps,width=4.in}} \\caption{A  three-dimensional view of a regular orbit with $\\beta=m_2/m_1=1.4/10$, $S_1=m_1^2$ and $S_2=0$. The initial angle $\\theta_1={\\rm arccos}({\\bf \\hat L_N}\\cdot{\\bf \\hat S_1}) = 45{}^o$.  The initial conditions for the orbit are $r/m=20$ and $r\\dot \\phi=0.209$. \\label{m.14r20}}  \\end{figure}   The precession, whether regular or irregular, modulates the emitted gravitational waves. The response of a detector on Earth to an impinging gravitational wave can be parametrized as \\be h=F_+h_++F_\\times h_\\times \\ee where $h_+$ and $h_\\times$ are the two gravitational wave polarization states and $F_+$ and $F_\\times$ are the antenna patterns.  A radiation coordinate system can be defined such that the polarization axes are fixed even in the presence of precession \\cite{chern}.  In such a coordinate system, $F_+$ and $F_\\times$ are constant.  However, the polarization states $h_+$ and $h_\\times$ depend on the inclination of the orbit and the precessional frequency in a complicated way \\cite{kidder}. For a circular orbit, the detector response can be written \\be h=A\\cos(2\\Phi-\\delta)\\label{simp}, \\ee where the higher harmonics have been ignored for simplicity. The amplitude $A$ and polarization phase $\\delta$ depend on the binary's location, orientation, and precession \\cite{{acst},{kidder}}. For an elliptical orbit, $h$ has terms of the form $\\cos(\\Phi),\\cos(2\\Phi)$, and $\\cos(3\\Phi),$ at quadrupole order so the gravitational wave spectrum shows oscillations at once, twice and three times the orbital frequency.    Precession of the orbital plane will (1) modulate the amplitude in eqn.\\ (\\ref{simp}) (2) modulate the polarization phase $\\delta $ and therefore the frequency of the gravitational waves and (3) contribute to the overall accumulated phase by changing the inspiral lifetime.  Any extreme sensitivity to initial conditions will most likely have the largest effect on the modulation of the amplitude and frequency of the gravitational waves. The overall accumulated contribution to the number of cycles in the observed waveform will certainly be effected by the general bulk precession but may be less sensitive to the irregularity of the precession until the very final stages of coalescence. The reason for this is that at the radii accessible to the interferometers, the irregularity seems to predominately effect the orientation of the orbit with a lesser effect on the net orbital velocity $\\vec \\omega = \\dot \\theta \\hat \\theta +\\sin\\theta \\dot \\phi \\hat \\phi$. LIGO/VIRGO aim to observe gravitational waves by accurately measuring the accumulated phase defined as \\be \\Psi=\\int \\omega dt=\\int {\\omega \\over \\dot \\omega}d\\omega , \\ee which may only have a small correction from the irregularity of the precession.  Irregular motion will effect the phasing, that is the gravitational wave frequency, and the amplitude of the wave. The number of cycles in the accumulated phase may therefore be determined by the average bulk behaviour of the precession. Although this conjecture requires further scrutiny, we can use the approximations  of Refs.\\ \\cite{{acst},{kidder}} to provide a rough figure for the number of wave cycles.  In Ref.\\ \\cite{acst}, circular Newtonian orbits were studied.  The effect of spin precession was isolated without including spin-orbit acceleration terms in the equations of motion in Ref. \\cite{acst}.  This separation of the spin precession equations and the equations of motion removed any possibility of chaotic coupling. However,  the gross features can be fairly represented by this approach. They estimate that the change in the total number of cycles amounts to about twice the total number of precessions.  For the binary black holes of size $15 M_\\odot +5M_\\odot$, there can be $\\sim 10 - 15$ precessions in the observable band and so there can be $\\sim 20-30$ additional cycles (depending on spin orientation, magnitude and eccentricity of the orbit). In Ref.\\ \\cite{kidder}, the contribution to the total number of orbits over the entire LIGO bandwidth was also estimated to be $\\sim {\\cal O}(20)$ for black hole binaries with total mass $\\sim 20 M_\\odot$.  The number of additional cycles could quadruple for NS/NS masses \\cite{kidder}  It has been argued that matched filtering will be a poor means by which to observe BH/BH orbits in the chaotic regime \\cite{{me1},{scott}} and that other cruder methods will be employed. We contribute to this debate only by indicating when irregularity will influence detectability.  It has also been emphasized in Ref.\\ \\cite{scott}, that the PN expansion is not sufficient to accurately model templates for $r \\lta 15 m$.  Nonetheless, the higher order contributions will incorporate stronger relativistic effects and therefore more nonlinearity which should only lead to more irregular motion. Since we are trying to provide a physical picture of the general trends and dependences, we continue to use the PN expansion. ",
        "conclusions": "It is reasonable to conclude for a typical stellar mass BH/BH system that if the primary black hole spins rapidly, the orbital plane will precess unpredictably for the final orbits before merger. The dependence of the precessional motion was tested as a function of three parameters:  $\\bullet $ $\\beta=m_2/m_1$:  As was already known, the smaller the mass ratio, the thicker the band occupied by the precessing orbit \\cite{{acst},{kidder}}. A ratio of $\\beta=1/3$ is small enough for precession to modulate observable gravitational waves.  The mass ratio may also determine the radius at which chaotic resonances occur by effecting the relative orbital and spin frequencies.   $\\bullet $ spin magnitudes:  The transition to chaos occurs as the spin magnitudes and misalignments are increased.  There can be chaotic motion even if only one body spins.  If it is a light companion spinning, then the magnitude of the spin has to be larger than maximal \\cite{maeda}.  If it is the heavier object, then chaos can occur for rapid but physically allowed black hole spins. This is true for both BH/BH and BH/NS pairs. For a given eccentricity and radial separation, a transition to chaos can occur as the spin of the companion is increased.  $\\bullet $ eccentricity:  Large misaligned spins cause eccentricity which may therefore be a feature common to chaotic orbits. However, large eccentricity alone certainly does not cause chaos. If chaos is indeed occurring near the unstable homoclinic orbits of Ref.\\ \\cite{loc}, then we might expect a fair spread in eccentricities for chaotic orbits since the homoclinic orbits occur with a broad range of eccentricities. Importantly, even for a completely regular orbit the usual relativistic precession of an elliptic orbit in the plane becomes superposed on the precession of the orbital plane.  This combination, though not an indication of chaos, does lead to a complicated modulation of the gravitational wave signal.  The most pressing question remains, will irregular orbits of coalescing binaries hinder observations? A detailed study of the modulated gravitational wave is needed.  A conjecture is that an erratic precession of the orbital plane will result in an erratic frequency or phasing of the gravitational wave but will not alter the total number of cycles much.  If this conjecture is fair, observations in the chaotic regime may be salvaged with a modification of the matched filtering method. As direct detections become more acute, we can aspire to watch the transition from regular to irregular motion.  We might then witness the signature nonlinearity of general relativity determine the fate of chaotic binaries.    \\vskip 25truept  I am grateful to V. Kalogera, V. Kaspi, S. Portegies Zwart and F. Rasio for sharing their expertise. I want to especially thank J.D. Barrow, M. Bucher, E. Bertschinger, E. Copeland, N.J. Cornish, P. Ferreira, L. Grischuk, S.A. Hughes, R. O'Reilly, and B.S. Sathyaprakash for their insightful comments and support.  I also want to thank the theoretical physics group at Imperial College for their hospitality. This work is supported by PPARC."
    },
    "0010/astro-ph0010371_arXiv.txt": {
        "abstract": "We present deep {\\it Hubble Space Telescope} optical imaging observations of two Galactic globular clusters, 47~Tuc and M15, using the {\\it Space Telescope Imaging Spectrograph} (STIS) longpass (LP) filter and the {\\it Wide Field and Planetary Camera 2} (WFPC2) F606W and F814W filters.  These globular clusters have very different metallicities (${\\rm [Fe/H] \\sim -0.7}$~dex for 47~Tuc, and ${\\rm [Fe/H] \\sim -2.2}$~dex for M15), essentially spanning the metallicity distribution of the entire Milky Way globular cluster system, and were chosen to investigate the relationship between magnitudes in the non-standard STIS LP system and the better characterized WFPC2 magnitudes.  We determine this relationship by combining our new STIS data with archival WFPC2 data, taking care to provide a robust and reliable transformation.  Examining the 47 Tuc and M15 data separately, there is no evidence for a metallicity dependence in the transformation relations.  This allows us to combine the data for both clusters, and also include the stars in the Small Magellanic Cloud that lie within the field of view of the 47~Tuc images, to determine final transformations that are valid for main sequence stars with colors in the range $0.3 \\lesssim$ \\mycolor\\ $\\lesssim 1.8$. These final relations predict STIS magnitudes within 0.05 mag (1$\\sigma$) of those measured for stars in both clusters.  In an appendix, we discuss the differences between our empirical transformation relations and those predicted by synthetic photometry. ",
        "introduction": "We have obtained deep {\\it Hubble Space Telescope} (HST) observations of the outer regions of two Galactic globular clusters, 47~Tuc (NGC~104) and M15 (NGC~7078), using the {\\it Space Telescope Imaging Spectrograph} (STIS) and the {\\it Near-Infrared Camera and Multi-Object Spectrometer} (NICMOS), as supporting observations for a project (GO~7419) with the scientific goal of determining the faint stellar luminosity function (LF) in the apparently dark-matter-dominated Ursa Minor dwarf spheroidal galaxy (UMi~dSph).  These data are supplemented by archival {\\it Wide-Field and Planetary Camera 2} (WFPC2) observations of these clusters and were obtained to aid our understanding of the deep STIS data for the Ursa Minor galaxy.  The UMi~dSph is one of the few Galactic satellites which appears to have undergone a very short-lived episode of star formation; indeed, its stellar population is remarkably similar to that of a classical halo globular cluster such as M15, being old and metal-poor \\citep{mateo}.  Thus, a direct comparison between the luminosity functions of M15 and the UMi dSph is equivalent to a stellar mass function comparison and constrains the amount of baryonic dark matter in the UMi dSph (see Feltzing, Gilmore \\& Wyse 1999 and Wyse et al.~1999 for preliminary analyses that find remarkable similarity between the LFs of the UMi~dSph, M15 and M92, another metal-poor halo globular cluster).  In this paper, we discuss the application of our multi-instrument photometry of 47~Tuc and M15 to the derivation of transformations from one photometric system to another.  Here, we use aperture photometry of stars in these clusters to derive transformations between the WFPC2 data taken with the F606W and F814W filters and STIS photometry taken with the optical longpass (LP) filter, including an investigation of the possible existence of any metallicity dependence introduced by the broad STIS LP filter.  Unless otherwise noted, we specify the sizes of all photometric apertures in terms of their radii (not their diameters) throughout this paper.  In Section~2, we describe the observational data and their reduction.  Section~3 discusses the photometry and how we selected the stars used in calculating the photometric transformations.  Section~4 presents the resulting transformation equations, and further discussion of these results is given in Section~5.  We summarize our conclusions in Section~6, and Appendix~A describes the differences between our empirical transformation relations and the relations predicted by SYNPHOT. In a complementary paper, \\citet{beaulieu} derive similar calibrations from STIS LP and WFPC2 F555W and F814W observations of NGC~6553, a metal-rich (${\\rm [Fe/H] \\sim -0.2}$) and highly reddened (E$_{\\rm B-V}\\sim 0.7$) globular cluster. ",
        "conclusions": "We have argued that  the differences between the linear fits to the complete samples of 47~Tuc and M15 color-color data shown in Figure~\\ref{colorcolor} are due to the presence of a higher-order term in the ``true,'' metallicity-independent transformation relations. Here, we discuss some other effects which could also produce this outcome and explain why we feel that these are  less important.  One possible source of differential systematic errors in the photometry is the different methodologies used to measure the WFPC2 magnitudes of 47~Tuc and M15. While both we and \\citet{pck97} performed aperture photometry, \\citet{pck97} used point-spread functions (PSFs) to first subtract nearby neighbors, which should improve the measured magnitudes, especially for the faintest stars.  In addition, PSF-fitting allows the removal of spurious stellar detections (e.g. background galaxies, bad pixels) through various goodness-of-fit selection criteria, such as $\\chi^2$ and sharpness. However, since the 47~Tuc and M15 fields observed here are not particularly crowded, the photometric improvements produced by PSF-fitting are only expected to be significant at the faintest magnitudes.  Our use of magnitude cutoffs for the WFPC2 data and our exclusion of objects which do not lie along the cluster main-sequence ridge lines have likely minimized any differences caused by the photometric algorithms employed.  Likewise, the magnitude cutoffs that we have adopted should preclude any significant biases in the photometry due to incompleteness of the WFPC2 data. For the WF3 observations of M15, \\citet{pck97} estimated their data to be 93\\% complete at \\myv=25 and 92\\% complete at \\myi=24; these levels should be approximately the same for the WF4 data, for which \\citet{dmp95a} give a completeness level of 92\\% at \\myi=24 and \\myi--\\myv$\\sim$1. We also expect the WF2 image of 47~Tuc to be more than 90\\% complete at the magnitude cutoffs used here.  Although \\citet{dmp95b} estimated their WF2 completeness level to be only $\\sim$80\\% at \\myi=23 and \\myi--\\myv$\\sim$1.65, this incompleteness was dictated by their F606W data.  We use a much more extensive set of F606W observations than they did, so our completeness level will be determined by the F814W data, which \\citet{dmp95b} show to be $\\sim$80\\% complete one full magnitude fainter than our \\myi\\ cutoff.  Could the documented  complications with WFPC2 photometry be the cause of the different linear transformation relations derived from the complete samples of 47~Tuc and M15 ridge-line stars? As discussed previously, the WFPC2 magnitudes in both clusters will be systematically too faint (about 3\\% on average) due to the CTE effect, but the differential effect between the two clusters should be negligible.  The LvS anomaly, on the other hand, while also causing the WFPC2 magnitudes to be systematically too faint, will become increasingly important as the magnitudes become fainter.  To first order, the errors in \\myv\\ and \\myi\\ will offset one another, so the \\myv--\\myi\\ colors will not be as greatly affected as the \\myv--\\mystis\\ and \\mystis--\\myi\\ colors will be.  While we have attempted to compensate for the LvS anomaly through the magnitude limits that we have applied to the WFPC2 data, the 47~Tuc and M15 stars will still have \\myv--\\mystis\\ colors which are progressively too red and \\mystis--\\myi\\ colors which are progressively too blue as fainter stars are considered. This will introduce a slight bit of curvature in the color-color relations shown in Figure~\\ref{colorcolor} and could potentially account for the differing transformation relations derived for the two clusters.  However, this ``curvature'' will not occur at the same color in 47~Tuc and M15 (because of their differing distances and metallicities) and therefore should affect the fits derived from the color-restricted samples of their members as well.  Since this is not seen, we do not expect that the CTE effect and LvS anomaly make significant contributions to the resulting transformation equations.  We have quantified  this further by performing photometry on the longest-exposure F606W and I814W images of 47~Tuc listed in Table~\\ref{hstdata}.  We then used the algorithms suggested by \\citet{whitmore} and by \\citet{cm98} to correct the measured counts for the CTE effect and for the LvS anomaly, respectively, and calculated new magnitudes for these stars.  Without applying any magnitude cutoff to these data, the same analysis procedure used for the combined images was adopted, resulting in a new determination of the 47~Tuc ridge-line stars.  This set of data contains a smaller number of stars than the data resulting from the combined WFPC2 images because, even though it extends to slightly redder colors than the magnitude-limited, combined data, most of the bright, blue main-sequence stars are no longer included because they are saturated in the 1500-sec F606W exposure.  Performing least-squares fits (with iterative 3-$\\sigma$ rejection) to the corrected, single-image colors of the 47~Tuc ridge-line stars, the resulting linear and quadratic transformation relations have the coefficients listed in Table~\\ref{singtransforms}.  Comparing these coefficients to the analogous fits from Table~\\ref{combtransforms}, we see that the linear fits have nearly identical slopes, while their intercepts differ by about 3\\%, as expected from the CTE effect.  Thus, the magnitude cutoffs that we have applied to the combined-image data appear to have reasonably precluded the LvS anomaly from significantly biasing our results.  This is reinforced by the 1$\\sigma$ agreement between the coefficients of the quadratic relations derived from the combined 47~Tuc, M15 and SMC data (Table~\\ref{combtransforms}) and those derived from the single-image 47~Tuc data alone (Table~\\ref{singtransforms}).  \\subsection{A Cautionary Note}  Note that the transformation relations that we present are only valid for magnitudes in a 0.$^{\\prime\\prime}$5 aperture. In particular, our transformations differ significantly from synthetic calculations, and we discuss this further in Appendix A, where we also describe a possible dependence of the \\mystis\\ aperture corrections on color.  We especially encourage those who use SYNPHOT to calculate color-color transformations to read Appendix A and to use care when applying the SYNPHOT (infinite-aperture) results to infer observed magnitudes."
    },
    "0010/astro-ph0010365_arXiv.txt": {
        "abstract": "We present theoretical evolutionary models for $Z=0$ stars in the mass range $4\\leq M/M_\\odot\\leq 8$ from the pre-main sequence up to the AGB phase. Contrary to previous calculations we found that these stars develop normal thermal pulses and third dredge-up episodes. Special attention is devoted to analyze the chemical enrichment in the envelope due to the above mechanisms. As a consequence, we show that these stars become carbon and nitrogen rich. U\\-sing different IMFs proposed in the literature for the Population III stars, we study their contribution to the pre-galactic chemical enrichment. It is found that $Z=0$ AGB stars could significantly contribute to $^7$Li, $^{12}$C and $^{14}$N and produce extreme non-solar $^{24}$Mg/$^{25}$Mg/$^{26}$Mg ratios. However, the net contribution is very sensitive to the IMF adopted and to the fraction of primordial matter which goes into stars. ",
        "introduction": "The standard homogeneous Big Bang nucleosynthesis predicts that the material emerged from this epoch was mainly made by $^1$H and $^4$He. The total mass fraction of heavier elements was lower than $10^{-10}$, so that the first generation of stars, the so called Population III, was built from matter essentially deprived of metals. Such a peculiarity significantly affected the star formation phase. Yoshii and Saio \\cite{YS86} found that the peak of the resulting IMF for Pop III stars ranges between 4 and 8 $M_{\\odot}$. Recently, Nakamura and Umemura \\cite{NU99} have obtained that the typical mass of Pop III stars should be $\\sim 3$ M$_\\odot$. Nevertheless, recent simu\\-lations of collapse and fragmentation of primordial clouds (see e.g. \\cite{La99}) obtain a Jean mass $\\sim 10^2-10^3$ M$_\\odot$, although it is still not clear whether these clumps will further fragment down to stellar mass values. The key role in this process is played by molecular hydrogen rather than dust or heavy molecules cooling. Therefore, early star formation with a normal present-day IMF seems very unlikely. In this framework, intermediate and/or very massive stars were the dominant constituents of the first stellar population.  The simplest explanation for the absence of \"zero metal\" stars is that they have not survived up to the present epoch, which is compatible with the idea that primordial stars were not of low mass. Thus, the very metal poor stars presently observed were formed from matter already enriched by the ashes of {\\it massive} Z=0 stars. Consequently, an indirect way to search for Pop III stars is to look for their nucleosynthetic imprint in the extremely metal poor objects now observed. As it is well known, the most important contribution of intermediate mass stars to the chemical enrichment of galaxies comes from the nucleosynthesis occurring during the AGB phase. ",
        "conclusions": ""
    },
    "0010/astro-ph0010153_arXiv.txt": {
        "abstract": "We present $u,b,r$ \\& $i$ galaxy number counts and colours both from the North and South Hubble Space Telescope Deep Fields and from the William Herschel Deep Field. The latter comprises a $7'\\times 7'$ area of sky reaching $b\\sim$28.5 at its deepest. Following Metcalfe et al. (1996) we show that simple Bruzual \\& Charlot evolutionary models which assume exponentially increasing star-formation rates with look-back time and q$_0$=0.05 continue to give excellent fits to galaxy counts and colours in the deep imaging data. With q$_0$=0.5, an extra population of `disappearing dwarf' galaxies is required to fit the optical counts.  We further find that the $(r-i):(b-r)$ colour-colour diagrams show distinctive features corresponding to two populations of early- and late-type galaxies which are well fitted by features in the Bruzual \\& Charlot models. The $(r-i):(b-r)$ data also suggest the existence of an intrinsically faint population of early-types at $z\\sim0.1$ with similar properties to the `disappearing dwarf' population required if q$_0$=0.5.  The outstanding issue remaining for the early-type models is the dwarf-dominated IMF which we invoke to reduce the numbers of $z>1$ galaxies predicted at $K<19$. For the spiral models, the main issue is that even with the inclusion of internal dust absorption at the A$_B=0.3$ mag level, the model predicts too blue $(u-b)$ colours for late-type galaxies at $z\\sim1$. Despite these possible problems, we conclude that these simple models with monotonically increasing star-formation rates broadly fit the data to $z\\sim3$.  We compare these results for the star formation rate history with those from the different approach of Madau et al. (1996). We conclude that when the effects of internal dust absorption in spirals are taken into account the results from this latter approach are completely consistent with the $\\tau$=9Gyr, exponentially rising star formation rate density out to $z\\approx3$ which fits the deepest, optical/IR galaxy count and colour data.  When we compare the observed and predicted galaxy counts for UV dropouts in the range $2\\la z\\la 3.5$ from the data of Steidel et al. (1999), Madau et al. (1996) and new data from the Herschel and HDF-S fields, we find excellent agreement, indicating that the space density of galaxies may not have changed much between $z=0$ and $z=3$ and identifying the Lyman break galaxies with the bright end of the evolved spiral luminosity function. Making the same comparison for B dropout galaxies in the range $3.5\\la z\\la 4.5$ we find that the space density of intrinsically bright galaxies remains the same but the space density of faint galaxies drops by a factor of $\\sim$5, consistent with the idea that L$^*$ galaxies were already in place at $z\\approx4$ but that dwarf galaxies may have formed later at $3\\la z\\la 4$. ",
        "introduction": "In four previous papers, Jones et al. (1991) (hereafter Paper I), Metcalfe et al. (1991) (hereafter Paper II), Metcalfe et al. (1995a) (hereafter Paper III) and McCracken et al. (2000) (hereafter Paper IV), we used photographic and CCD data to study the form of the galaxy number-magnitude relation at both optical ($B\\sim27.5$) and infra-red ($K\\sim20$) wavelengths. In this paper we extend our observations of the $B$-band field of paper III to even deeper magnitudes ($B\\sim28$ mag), and over a much larger area of $7'\\times 7'$, by exposing for a total of 30 hours over five nights with the Tektronix CCD at the prime focus of the 4.2 metre William Herschel telescope (WHT). We have also acquired $\\sim34$ hours of exposure in the $U$-band (to $U\\sim27$ mag), 8 hours in the $R$-band (to $R\\sim26.5$ mag) and 5 hours in the $I$-band (to $I\\sim25.5$ mag) on the same field, again with the WHT. We refer to this field throughout as the William Herschel Deep Field (WHDF).  Although not competing in terms of either collecting area or field of view with these ground-based observations, the high resolution of the Hubble Space Telescope offers the ability to image sub-arcsecond objects to much greater depths than can be done from the ground. The Hubble Deep Field (HDF) project (Williams et al. 1996) has provided such a dataset in the public domain, and here we present our analysis of the $F300W$, $F450W$, $F606W$ and $F814W$ preliminary release North (HDF-N) and South (HDF-S) fields, using similar data reduction techniques as for the WHT data. Roughly speaking, these passbands correspond to $U$, $B$, $R$ \\& $I$ (see section 4.2). For unresolved objects the $3\\sigma$ limit of $B\\sim29.5$ mag on the $F450W$ image is considerably deeper than our WHT frame. However, for resolved galaxies this limit is much brighter. Indeed, for sources more extended than the WHT `seeing' disk ($\\sim 1.25''$ radius) the ground-based frame is slightly deeper, making it an ideal complement to the HST images. A similar situation exists between the WHT $u$ and the HDF $F300W$ frames, where, for a source at the WHT seeing limit, the WHT is actually over $0.5$ mag deeper than the HDF. The same is not true for the $F606W$ and $F814W$ images, which are 2-3 mag deeper (for typical galaxy sizes) than the deepest previous $R$ and $I$-band galaxy counts (Smail et al. 1995, Hogg et al. 1997).  The modelling and interpretation of HDF-N data has been discussed briefly elsewhere (Metcalfe et al. 1996). Here we discuss the data reduction for both ground- and space-based data, and present the number counts, colour distributions and models in greater detail. Finally we discuss the comparison between the models and  the data.  The optical - infra-red properties of the WHDF, based on infra-red imaging data to $K\\sim20$ taken on the United Kingdom Infrared Telescope, and data to $K\\sim23$ on one smaller, $1.8'\\times 1.8'$ area, are discussed in Paper IV.  We now also have infra-red imaging data covering the whole of our WHDF to $H\\sim22.5$ mag  taken with the $\\Omega$ Prime camera on the Calar Alto 3.5m telescope. This data will be  discussed in a subsequent paper (McCracken et. al., in prep.). ",
        "conclusions": "We have produced $u_{ccd},b_{ccd},r_{ccd}$ and $i_{ccd}$ galaxy number counts and $(u-b)_{ccd}$, $(b-r)_{ccd}$ and $(r-i)_{ccd}$ colours from one $7'\\times7'$ field, on which we have the deepest ground-based $B$-band image yet taken. We also have deep $H$ and $K$-band photometry on this field (Paper IV, McCracken et al. in prep).  We have compared these data with our analysis of the HDF-N and HDF-S, produced using the same software. For the UV and blue-bands, although the HDF's are deeper for unresolved objects, in terms of surface brightness signal-to-noise our WHT image is $\\sim0.4$ mag deeper, and also covers 5 times the area. The counts from the datasets agree well, after the ground-based data has been corrected for confusion. All are still rising at the faintest magnitudes, with a slope $d\\/log(N)/dm\\sim0.25$. The HDF $F606W$ counts are over 2 magnitudes deeper than the previous equivalent ground-based limits. They also show a still-rising count at the faint end with a slope $d\\/log(N)/dm\\sim0.25$. The ground-based $U$ counts show an almost constant slope of $d\\/log(N)/dm\\sim0.4$ to $U\\sim25.5$. However the HDF $F300W$ counts flatten to a slope $d\\/log(N)/dm\\sim0.15$.  The deepest counts are the HDF-S STIS data, which reach to $STIS_{AB}\\sim30$, which is roughly equivalent to $R\\sim30$. Our evolutionary models continue to fit to the limit of these deepest optical number count data.   In terms of the comparison with evolutionary models, we summarise our conclusions  as follows:  1. We have fitted simple exponential star-formation history models to our data, and found that low $q_0$ cosmologies give a good fit from $U$ to $K$ over a range of $\\sim10$ magnitudes. To enable high $q_0$ models to fit, an extra component must be added to the galaxy population at high ($z\\ga1$) redshift. One way of doing this is by adding a `disappearing dwarf' dE population which is bright and blue at high redshift and faint and red at low redshift.  2. At very bright magnitudes ($B\\la 17$) our count models fail and overpredict the counts. Since the steep count at $B\\la 17$ seems to be reproduced at the same level at $R$, $I$, $H$ and $K$, which is unlikely for an evolutionary explanation, this  represents strong evidence that the high normalisation we have adopted for the counts is substantially correct.  3. The models provide a good fit to the distributions of colours in $(b-r)_{ccd}$ and $(F450-F606)_{vega}$ and in $(r-i)_{ccd}$ and $(F606-F814)_{vega}$, although in both $(u-b)_{ccd}$ and $(F300-F450)_{vega}$ the spiral models are too blue by $0.5$ magnitudes at $z=1$.  4. The same models produce elliptical and spiral tracks in the colour-colour planes with shapes very similar to the loci of the data. The fact that the fit of our models is so good, particularly in $(r-i)_{ccd}:(b-r)_{ccd}$ for the ground-based data and $(F606-F814)_{vega}:(F450-F606)_{vega}$ for the HDF, data lends strong support to the underlying assumption that the star-formation rate increases exponentially to beyond $z=2$ (Metcalfe et al. 1996). It is clearly possible to use $(R-I):(B-R)$ to probe the star-formation rate directly over a wide redshift range $0.5\\la z\\la3$, emphasising the power of this colour-colour diagram as a diagnostic tool for galaxy evolution.  5. In the WHDF $(r-i)_{ccd}:(b-r)_{ccd}$ plane, we detect an excess of low redshift($z\\approx 0.1$), low luminosity early-type galaxies which may have parameters consistent with the `disappearing dwarfs' invoked in our q$_0$=0.5 evolutionary model.  6. We have shown how the discrepancy with the $z=1$ SFR peak of Madau et al. (1996) comes from the different treatment of dust and photometry in the HDF. When we repeat the UV dropout SFR density measurement made in the $2\\la z\\la3.5$ range by Madau et al. we find good self-consistency with the prediction of our exponentially increasing SFR ($\\tau$=9Gyr) spiral model.  7. We have made new measurements of the SFR density in the WHDF and in the HDF-N and HDF-S. In the redshift range $1\\la z\\la2$ the WHDF SFR density lies lower than the measurement of Connolly et al. (1997) in the HDF. These measurements bracket the prediction of our $\\tau$=9Gyr spiral model. In the $2\\la z\\la3.5$ range our observed SFR density based on UV dropouts, corrected for dust and LF incompleteness, and the ground-based estimates of Steidel et al. (1999) lie close to the HDF estimates and our model. At $3.5\\la z\\la4.5$ there is a wider spread, with the brighter ground-based SFR estimates from the WHDF and Steidel et al. generally being higher and closer to the model than the HDF SFR estimates which are significantly lower than the model.   8. We have detected a population of galaxies that occupy the same part of the $(r-i)_{ccd}:(b-r)_{ccd}$ plane as $z\\sim3$ UV dropout galaxies, but which do have measurable $u$-band fluxes. We have suggested that these may be candidates for high redshift galaxies that do not show UV dropout and that statistics of QSO absorption lines may not rule out such a population at $z\\approx3$. The increase in SFR density at $z\\approx3$ will depend on the fraction which prove to be at high redshift.  9. We have created simulated CCD frames to check the apparent sizes of galaxies predicted by our evolutionary models and found they are quite consistent with the sizes of galaxies measured on the HDF-N.   10. For the early-type models, the main remaining problem is that although they produce reasonable fits to the data at optical wavelengths, they predict too many high redshift ($z\\ga1$) galaxies at $K\\sim19$ unless they have a dwarf-dominated stellar IMF. This reduces the evolutionary brightening at infra-red wavelengths whilst still allowing sufficient optical evolution.  Dynamical merging may constitute an alternative explanation and this possibility has been explored by other authors (eg Kauffmann \\& Charlot 1998).   11. The main outstanding question for the spiral models is why do they predict $(u-b)_{ccd}$ and $(F300-F450)_{vega}$ colours which are too blue in the range $0.5\\la z\\la1$? We have suggested that this problem is unlikely to be solved by the addition of more dust to the models because the amount of dust needed is so great (A$_B\\sim1.3$ mag) that it would lead to the model predicting $(B-R)$ colours that are $\\sim$0.5 magnitudes too red for a $z=0$ spiral. One possibility is that the spiral IMF is truncated above 5-6 M\\sun. It is the subject of continuing research to see if such a model can produce enough evolution to continue to fit the steep $B$-band number-count slope. Another possible explanation, suggested by the dust laws in nearby galaxies with different metallicities, is that the dust extinction law evolves, with the 2200\\AA~ dust absorption feature appearing in high metallicity galaxies  at $z\\la0.5$ and disappearing in lower metallicity galaxies at higher redshift.  12. The evolutionary q$_0$=0.05 and q$_0$=0.5 models correctly predict the sky density of Lyman break $z\\approx3$ galaxies detected by Steidel et al. (1999) from Keck spectroscopy. The models suggest that these galaxies may be spiral galaxies which have undergone an exponentially increasing SFR with look-back time. At $z\\approx4$ these models generally over-predict the space density of sub-L$^*$ galaxies at faint limits in the HDF-N and HDF-S and WHDF by a factor of 5, suggesting that we may have reached an epoch of significant galaxy formation. The $z\\approx4$ luminosity function matches best our q$_0$=0.5 model with the dE population removed, suggesting that one interpretation might be that  bright L$^*$ galaxies were already in place at $z=4$ but many fainter galaxies formed at $3\\la z\\la4$. These then quickly dimmed to become the red, dwarf population possibly detected in the WHDF $(r-i)_{ccd}:(b-r)_{ccd}$ data at M$_B=-15$ at $z\\approx0.1$."
    },
    "0010/astro-ph0010479_arXiv.txt": {
        "abstract": "We investigate the axisymmetric structure of collimating, relativistic, strongly magnetized (force-free) jets. In particular, we include the differential rotation of the foot points of the field lines in our treatment. The magnetic flux distribution is determined by the solution of the Grad-Shafranov equation and the regularity condition along the light surface. With differential rotation, i.e. the variation of the iso-rotation parameter $\\omf $, the shape of the light surface is not known a priori and must be calculated in an iterative way.  For the first time, we have calculated the force-free magnetic structure of truly two-dimensional, relativistic jets, anchored in a differentially rotating disk. Such an approach allows for a direct connection between parameters of the central source (mass, rotation) and the extension of the radio jet. In particular, this can provide a direct scaling of the location of the asymptotic jet light cylinder in terms of the central mass and the accretion disk magnetic flux distribution.  We demonstrate that differentially rotating jets must be collimated to a smaller radius in terms of the light cylinder if compared to jets with rigid rotation. Also, the opening angle is smaller. Further we present an analytical estimate for the jet opening angle along the asymptotic branches of the light surface. In general, differential rotation of the iso-rotation parameter leads to an increase of the jet opening angle.  Our results are applicable for highly magnetized, highly collimated, relativistic jets from active galactic nuclei and Galactic superluminal jet sources. Comparison to the M87 jet shows agreement in the collimation distance. We derive a light cylinder radius of the M87 jet of 50 Schwarzschild radii. ",
        "introduction": "Observations of astrophysical jet sources have now established the idea that jet formation is always connected to the presence of an accretion disk and strong magnetic fields. This holds for various scales of energy output, jet velocity and nature of the jet emitting objects. Examples are jets from active galactic nuclei (AGN), Galactic superluminal jet sources, the example of a mildly relativistic jet from a neutron star (SS 433) and the numerous class of protostellar jets (see Zensus et al. 1995; Mirabel \\& Rodriguez 1995; Mundt et al. 1990; Ray et al. 1996). Magnetic jets are believed to originate very close to the central object in the interaction region with the accretion disk. Beside observational arguments also theoretical considerations have shown that magnetic fields play an important role in jet formation and propagation (Blandford \\& Payne 1982; Pudritz \\& Norman 1983; Shibata \\& Uchida 1985; Sakurai 1985; Camenzind 1987; Lovelace et al. 1991).  If the central object is a black hole as it is the case for AGN and Galactic superluminal jet sources, the surrounding accretion disk is the only possible location for a field generation (by dynamo action or/and advection of flux). In the case of stellar objects (protostars, white dwarfs or neutron stars), the central star also carries a relatively strong magnetic field and it is not yet clear, whether the jet magnetosphere originates in the disk or in the star. However, a strong interaction between stellar field and accretion flow is evident. The jet formation process itself is not yet fully understood theoretically. In particular, for the mass transfer from the disk into the jet and the process of magnetic field generation a complete physical model is missing.  However, over the last decades the basic ideas of Blandford \\& Payne (1982) have been extended by various authors. The general picture is the following. Matter is lifted from the disk into the magnetosphere and becomes magnetically accelerated (Ferreira 1997). Toroidal magnetic fields, generated by inertial back-reaction of the plasma on the poloidal field, may collimate the disk magnetosphere into a highly collimated jet flow (Camenzind 1987; Chiueh et al. 1991; Lovelace et al. 1991). In general, due to the complexity of the MHD equations, stationary relativistic models of magnetic jets has to rely on simplifying assumptions such as self-similarity (Contopoulos 1994, 1995), some other prescription of the field structure (Li 1993, Beskin 1997) or the restriction to asymptotic regimes (Chiueh et al. 1991; Appl \\& Camenzind 1993; Nitta 1994, 1995). For highly magnetized jets the force-free limit applies. This allows for a truly two-dimensional calculation of the magnetic field structure (Fendt et al. 1995; Fendt 1997a). Relativistic magnetohydrodynamics implies that poloidal electric fields, which are not present in Newtonian MHD, cannot be neglected anymore.  From the observations we know that extragalactic jets as well as Galactic superluminal jets and protostellar jets are collimated almost to a cylindrical shape (Zensus et al. 1995; Ray et al. 1996; Mirabel \\& Rodriguez 1995). Theoretically, it has been shown that current carrying relativistic jets must collimate to a cylinder (Chiueh et al. 1991). For the asymptotic limit of a cylindrically collimated magnetic relativistic jet, Appl \\& Camenzind (1993) found a non-linear analytical solution for the trans-field force-balance in the case of a constant iso-rotation parameter. These results were further generalized for jets with differential rotation (Fendt 1997b). Such an asymptotic field distribution is especially interesting for jets emerging from (differentially rotating) accretion disks.  In previous papers, we applied the asymptotic jet model of Appl \\& Camenzind (1993) as a boundary condition for the calculation of {\\em global, two-dimensional,} force-free jet magnetospheres for rapidly rotating stars (Fendt et al. 1995) or rapidly rotating black holes (Fendt 1997a). In this paper, we continue our work on 2D force-free jet magnetospheres applying an asymptotic jet with {\\em differential rotation} of the iso-rotation parameter $\\omf$ as boundary condition for the global jet structure. Such an approach allows for a connection between the differentially rotating jet source -- the accretion disk -- and the asymptotic collimated jet. Since jet motion seems intrinsically connected to the accretion disk, differential rotation of the field lines should be a natural ingredient for any magnetic jet structure. As a principal problem for differentially rotating relativistic jet magnetospheres, position and shape of the singular light surface are not known {\\it a priori}, but have to be calculated in an iterative way together with the magnetic flux distribution.  In Sect.\\,2 we recall some basic equations of the theory of relativistic magnetospheres and discuss several difficulties with the solution of the Grad--Shafranov (hereafter GS) equation. After some comments on the numerical approach in Sect.\\,3, we discuss our results in Sect.\\,4. A summary is given in Sect.\\,5. ",
        "conclusions": "We have investigated the two-dimensional magnetic field distribution in collimating, relativistic jets. The structure of the axisymmetric magnetic flux surfaces is calculated by solving the relativistic force-free Grad-Shafranov equation numerically. In relativistic MHD, {\\em electric fields} become important in difference to Newtonian MHD. The simplifying assumption of the force-free limit has been applied as relativistic jets must be highly magnetized.  The central point of our paper is the consideration of differential rotation of the foot points of the field lines, i.e. a variation of the iso-rotation parameter $\\omf(\\Psi)$. The underlying model is that of a magnetic jet anchored in an accretion disk. Two main problems had to be solved in order to calculate a two-dimensional field distribution: a) to determine the {\\it a priori} unknown location of the light surface, b) the proper treatment of the regularity condition along that light surface. The light surface is the force-free equivalent of the Alfv\\'en surface and provides a singularity in the Grad-Shafranov equation. We summarize our results as follows.  (1) We find numerical solutions for the two-dimensional magnetic flux distribution connecting the asymptotic cylindrical jet with a differentially rotating disk. In our example solutions the asymptotic jet radius is about 2.5 times the asymptotic light cylinder radii. This is the first truly two-dimensional relativistic solution for a jet magnetosphere including differential rotation of the iso-rotation parameter $\\omf(\\Psi)$. The physical solution, being characterized by a smooth transition across the light surface, is unique for a certain parameter choice for the rotation law $\\omf$.  (2) The half opening angle of the numerical jet solution is about 60 degrees. Cylindrical collimation is achieved already after a distance of 1-2 asymptotic light cylinder radii along the jet axis. Differential rotation decreases the jet opening angle, but increases the distance from the jet origin where collimation is achieved. The ``jet expansion rate'', defined as the ratio of the asymptotic jet radius to the jet radius at the jet origin, is about 10.  (3) From the analytical treatment of the regularity condition along the asymptotic branches of the light surface we derive a general estimate for the jet opening angle. We find that the jet half opening angle is larger than $45\\degr$ and increases for a steeper profile of the differential rotation $\\omf$.  (4) Our two-dimensional ansatz, in combination with the treatment of differential rotation, allows for a connection of the asymptotic jet solution with the accretion disk. Certain disk properties can be deduced from the asymptotic jet parameters. Examples are the disk toroidal magnetic field distribution, with a maximum at half of the disk radius and the angular momentum flux per unit time and unit radius. This is interesting as a boundary condition for accretion disk models. We find that most of the angular momentum is lost in the outer part of the disk.  (5) Application of our model to the M87 jet gives good agreement qualitatively. From our numerical solution we derive an asymptotic light cylinder of the M87 jet of about 50 Schwarzschild radii. Collimation of the jet would be achieved after a distance of two asymptotic light cylinder radii from the source. This value is comparable with the observations, however, the opening angle in our model is larger by a factor of two.  In summary, we have presented the first global two-dimensional solutions for a relativistic jet magnetosphere taking into account differential rotation of the jet footpoints. From our {\\em jet} model we may determine certain physical quantities in the {\\em disk} that are not possible to observe, as e.g. the angular momentum flux distribution at the disk-jet interface. Comparison with the M87 jet shows that our model seems to be consistent with the observations, therefore allowing for a derivation of the collimation distance, the light cylinder radius and the jet expansion rate for that example. Clearly, such features as the time-dependent ejection of knots and the interaction process between disk, jet and central source cannot be answered by our approach. Time-dependent relativistic MHD simulations of the whole ``star''-disk-jet system would be necessary, however, such codes are not yet fully developed."
    },
    "0010/astro-ph0010015_arXiv.txt": {
        "abstract": "In this paper, I will take a synoptic approach to determining the intergalactic infrared radiation field (IIRF). This approach draws on both the multi-TeV $\\gamma$-ray observations and the infrared background observations and relates them {\\it via} the semi-empirical modelling method of Malkan \\& Stecker. I discuss the evidence for an intergalactic infrared background obtained by an analysis of the {\\it HEGRA} observations of the high energy $\\gamma$-ray spectrum of Mrk 501 and from constraints from Mrk 421 deduced from the Whipple air Cherenkov telescope results. I will show that this evidence is in accord with the predictions made by Malkan \\& Stecker (1998) for the intergalactic infrared spectral energy distribution produced by galaxies. The Malkan-Stecker predictions are also in excellent agreement with mid- and far infrared galaxy counts. However, there may be potential problems relating these predictions with the results of the analysis of {\\it COBE-DIRBE} far infrared data. The $\\gamma$-ray and {\\it COBE-DIRBE} observations may also need to be reconciled. I will discuss possible ways to resolve this situation including a partial nullification of the $\\gamma$-ray absorption process which can hypothetically occur if Lorentz invariance is broken. ",
        "introduction": "Shortly after the discovery that blazars can be strong sources of high energy $\\gamma$-rays, we suggested that very high energy $\\gamma$-ray beams from blazars can be used to measure the intergalactic infrared radiation field, since pair-production interactions of $\\gamma$-rays with intergalactic IR photons will attenuate the high-energy ends of blazar spectra (Stecker, De Jager \\& Salamon 1992). In recent years, this concept has been used successfully to place upper limits on the the intergalactic IR field (IIRF) (Stecker \\& De Jager 1993; Dwek \\& Slavin 1994; Stecker \\& De Jager 1997; Stanev \\& Franceschini 1997; Biller {\\it et al.} 1998). Determining the IIRF, in turn, allows us to model the evolution of the galaxies which produce it. As energy thresholds are lowered in both existing and planned ground-based air Cherenkov light detectors, cutoffs in the $\\gamma$-ray spectra of more distant blazars are expected, owing to extinction by the IIRF. These can be used to explore the redshift dependence of the IIRF.  On the other hand, by modelling the intergalactic infrared and optical radiation fields as a function of redshift, one can calculate the expected $\\gamma$-ray opacity as a function of redshift and comparison of predicted absorption with the data can yield important new information about the evolution of the IIRF (Stecker \\& De Jager 1998; Salamon \\& Stecker 1998).  Exploring the IIRF with high energy $\\gamma$-rays has both advantages and disadvantages over direct infrared observations. An important advantage is that since the absorption process only acts effectively in intergalactic space over large distances, the $\\gamma$-ray approach does not suffer from problems of subtraction of foreground emission from both zodiacal light and interstellar dust in the Galaxy. As we have seen at this meeting, these can be formidable sources of uncertainty in determining the IIRF. Also, high energy $\\gamma$-ray observations of blazar and $\\gamma$-ray burst spectra at high redshifts can be used to probe the past evolution of the IIRF whereas direct infrared observations can only tell us about the present IIRF.  On the other hand, the disadvantage of the $\\gamma$-ray approach is the uncertainty in the unabsorbed spectrum of the source. This uncertainty is somewhat ameliorated by the strong energy dependence of the absorption effect itself. Therefore, it is my contention that using both direct infrared and high energy $\\gamma$-ray observations in a ``synoptic'' approach can yield the most information about the IIRF. ",
        "conclusions": "Studies of the absorption of high energy $\\gamma$-rays in the spectra of extragalactic objects can be used to probe both the present intergalactic infrared background and the infrared and optical radiation that existed in intergalactic space at higher redshifts. The results of such observations are free of the effects of solar system and galactic foreground contamination. Such studies are most effective when combined with direct infrared observations in a synoptic approach.  High energy $\\gamma$-ray observations have already provided the best constraints on the mid-infrared background and they support a galaxy evolution scenario where emissivity evolution in the mid-infrared evolved as $\\sim (1 + z)^Q$ with $Q \\sim 3$. This is consistent with observations of galaxy counts using {\\it ISOCAM}. On the other hand, the analysis of {\\it COBE-DIRBE} observations favors a stronger evolution in far-infrared emissivity with $Q > 4$.  There are various possible interpretations of these results; among them is the possibility that special relativity is modified at high energies. Future $\\gamma$-ray and infrared observations will be needed to resolve this situation."
    },
    "0010/astro-ph0010259_arXiv.txt": {
        "abstract": "We perform evolutionary calculations of binary stars  to find progenitors of systems with parameters similar to the recurrent novae U Sco.  We show that a U Sco type--system may be formed starting with an initial binary system which has a low--mass carbon-oxygen white dwarf as an accretor. Since the evolutionary stage of the secondary is not well known, we calculate sequences with hydrogen rich and helium rich secondaries. The evolution of the binary may be devided into several observable stages as: classical nova, supersoft X--ray source with hydrogen stable burning and strong  wind phases, ending up with the formation of a massive  white dwarf near the Chandrasekhar mass limit. We follow the chemical evolution of the secondary as well as of the matter lost from the system, and we show that observed $^{12}$C/$^{13}$C and N/C ratios may give some information about the nature of the binary. ",
        "introduction": "Recurrent novae are a small class of objects which bear many similarities to other cataclysmic variable systems. They experience recurrent outbursts at intervals of 20--80 yrs.  Webbink et al. (1987) lengthily discussed the nature of the recurrent novae, and they concluded that according to outburst mechanisms there are two subclasses of these systems: (a) powered by thermonuclear runaway on the surface of the white dwarf (e.g. U Sco), and (b) powered by the transfer of a burst of matter from the red giant to the main--sequence companion.  U Sco is one of the best observed recurrent novae. Historically, its outbursts were observed in 1863, 1906, 1936, 1979, 1987 and in 1999. Determinations of the system visual luminosity at maximum and minimum, indicate a range $\\Delta m_V \\sim $9. We also note that the mean recurrent interval implied by known outbursts is $P_{rec} = 23 ~$yrs. Schaefer (1990) and Schaefer \\& Ringwald (1995) observed eclipses of U Sco in the quiescent phase, and determined the orbital period $P_{orb}$=1.23056 d.  Ejecta abundances have been estimated (from 1979 outburst) from optical and UV studies by Williams et al. (1981) and Barlow et al. (1981). They derived extremely helium rich ejecta He/H$\\sim$ 2 (by number), while the CNO abundance was solar with an enhanced N/C ratio. From the analysis of the 1999 outburst Anupama \\& Dewangan (2000) obtained an average helium abundance of He/H$\\sim$0.4$\\pm$0.06. The estimated mass of the ejected shell for 1979 and 1999 outbursts is $\\rm \\sim 10^{-7} ~M_\\odot$ (Williams et al. 1981; Anupama \\& Dewangan 2000). Spectroscopically, U Sco shows very high ejection velocities of (7.5--11)$\\rm \\times 10^3 ~km ~s^{-1}$ (Williams et al. 1981; Munari et al. 2000). Latest determinations of the spectral type of the secondary indicate a K2 subgiant (Anupama \\& Dewangan 2000; Kahabka et al. 1999). Following to  Kahabka et al. (1999) the distance to U Sco is about 14 kpc.  According to Kato model (1996), supersoft X--ray emission is predicted to be observed about 10--60 days after the optical outburst. BeppoSAX detected  supersoft X-ray emission from U Sco in range 0.2--20 keV just 19--20 days after the peak of its optical outburst in February 1999 (Kahabka et al.  1999). The fact that U Sco was detected as a supersoft X--ray source (SSS) is consistent with steady hydrogen burning on the surface of its white dwarf component.   In this paper we construct several evolutionary sequences which may lead to formation of systems like U Sco. In Section 2 the evolutionary code is briefly described. In Section 3 we discuss the effect of mass transfer on binary evolution. Section 4 contains the results of the calculations. A general discussion and conclusion follow. ",
        "conclusions": "We calculated several evolutionary sequences to reproduce orbital and physical parameters of the recurrent novae of U Sco--type. We showed that U Sco systems possibly form from binaries with a low--mass C--O white dwarf as accretor. Such a system evolves through several observable stages: recurrent nova, SSS with stable hydrogen burning and SSS with strong wind phases. In final phase of evolution a massive white dwarf near the Chandrasekar limit is formed. We propose that the evolutionary sequence B is able to produce binary system with parameters similar to U Sco. Our best fitting model has initial parameters: $M_{sg,i} = 1.7 ~M_\\odot $, $M_{wd,i} = 0.85 ~M_\\odot $ and $P_i (RLOF) = 1.61 $d. Based on evolutionary model we constracted a detailed theoretical model for U Sco. We found that the best fit parameters are: $M_{sg} = 0.94 ~M_\\odot $, $M_{wd} = 1.31 ~M_\\odot $, $\\log L_{sg}/L_\\odot = 0.42 $, $\\dot M_{sg} = 3.58 \\times 10^{-8} M_\\odot ~yr^{-1} $ for $P_{orb} = 1.23056 ~d$."
    },
    "0010/astro-ph0010290_arXiv.txt": {
        "abstract": "We review observations of molecular absorption line systems at high redshift toward red quasars and gravitational lenses. ",
        "introduction": " ",
        "conclusions": ""
    },
    "0010/astro-ph0010545_arXiv.txt": {
        "abstract": "{There are a number of theoretical and observational hints that large numbers of low-mass galaxies composed entirely of dark matter exist in the field.  The theoretical considerations follow from the prediction of cold dark matter theory that there exist many low-mass galaxies for every massive one. The observational considerations follow from the observed paucity of these low-mass galaxies in the field but not in dense clusters of galaxies; this suggests that the lack of small galaxies in the field is due to the inhibition of star formation in the galaxies as opposed to the fact that their small dark matter halos do not exist.  In this work we outline the likely properties of low-mass dark galaxies, and describe observational strategies for finding them, and where in the sky to search.  The results are presented as a function of the global properties of dark matter, in particular the presence or absence of a substantial baryonic dark matter component.  If the dark matter is purely cold and has a Navarro, Frenk \\& White density profile, directly detecting dark galaxies will only be feasible with present technology if the galaxy has a maximum velocity dispersion in excess of 70 km s$^{-1}$, in which case the dark galaxies could strongly lens background objects.  This is much higher than the maximum velocity dispersions in most dwarf galaxies.  If the dark matter in galaxy halos has a baryonic component close to the cosmic ratio, the possibility of directly detecting dark galaxies is much more realistic; the optimal method of detection will depend on the nature of the dark matter. A number of more indirect methods are also discussed.} ",
        "introduction": "It has been a long-recognized feature of hierarchical clustering models of galaxy formation, like the cold dark matter (CDM) model, that they predict the existence of a very large numbers of low-mass galaxies.  This follows directly from the shape of the fluctuation spectrum.  On the mass scales that correspond to those of galaxies, cold dark matter has a fluctuation spectrum $\\delta (k)$ whose power spectrum $\\vert \\delta (k) \\vert^2 \\sim k^n$ has index $n \\sim -2$ (Bardeen et al.~1986); such a power spectrum results in a mass function of dark-matter (DM) halos that is steep at low masses, close to $n(M) \\sim M^{-2}$ (Press \\& Schechter 1974, White \\& Rees 1978, White \\& Frenk 1991, Lee \\& Shandarin 1999). The prediction of CDM theory that there should exist many satellite galaxies per giant galaxy has recently been confirmed by numerical simulations (Moore et al.~1999, Klypin et al.~1999 and references therein).  But observations (Cowie et al.~1996, Lin et al.~1996, Ellis et al.~1996) show that field galaxies have a luminosity function $n(L)$ far shallower than $n(L) \\sim L^{-2}$.  This is normally reconciled with a steep mass function by arguing that the formation of stars in low-mass galaxies is inefficient (e.g.~Efstathiou 1992, White \\& Kauffmann 1994, Efstathiou 2000) so that low-mass galaxies have far less luminosity per unit total mass (including dark matter) than high-mass galaxies.  Recent observations have, however, revealed steep luminosity functions $n(L) \\sim L^{-2}$ in both the Virgo (Phillipps et al.~1998) and Fornax (Kambas et al.~2000) clusters.  Neither of these are particularly rich clusters, like the Coma cluster, but the important thing is that they are (unlike Coma) close enough to us that that their luminosity functions can be measured directly down to very faint limits (see Fig.~2 of Trentham 1998). On the other hand the luminosity function of galaxies in the Ursa Major Cluster (Trentham, Tully \\& Verheijen 2000), a nearby diffuse group of spiral galaxies about one-twentieth as massive as the Virgo Cluster (Tully et al.~1996), is far shallower than this. The Ursa Major and Virgo/Fornax luminosity functions are inconsistent at a high level of confidence.  The advent of mosaic CCDs on large telescopes and the consequent improved depth to which large areas of the sky can be surveyed to very faint magnitudes (the Trentham et al.~(2000) Ursa Major survey had a 1$\\sigma$ surface-brightness threshold fainter than $\\mu_R = 27$ mag arcsec$^{-2}$) means that the long-standing worry (Disney 1976) of there existing many galaxies in diffuse environments that are missing from (previously shallow) surveys because their surface-brightnesses are too low is far less serious than it used to be.  It also appears that the luminosity function of the Local Group (van den Bergh 2000 and references therein) is not steep between $M_R = -13$ and $M_R = -11$, which is where the Virgo Cluster luminosity function is found to be steep; galaxies that have been discovered recently in the Local Group, like Cetus (Whiting, Hau \\& Irwin 1999) are fainter than this, so that van den Bergh's luminosity function is probably complete down to $M_R=-11$. The shallow luminosity function that was seen in Ursa Major may well therefore be a common feature of diffuse environments. This would be in apparent contrast to the upturn in the luminosity function of dwarf irregular galaxies in the field seen by Marzke et al.~(1994), but the value of $\\alpha \\sim -1.8$ quoted by those authors comes from a Schechter (1976) function fit to the luminosity function and is only very indirectly related to the faint-end power-law index of the luminosity function as discussed above (the Schechter $\\alpha$ is determined primarily by the shape of the bright end of the luminosity function appropriate the the particular kind of galaxy being studied; see Section 2 of Trentham 1997)  Let us assume that the mass function of halos is the same in small galaxy clusters like Virgo and Fornax as it is in large galaxy groups like Ursa Major. This would be likely in any hierarchical clustering model of galaxy formation in low-density environments. It would be less true in very high-density environments like the cores of very rich galaxy clusters like Coma where dark-matter halos can be tidally disrupted (see the high-resolution numerical simulations of Ghigna et al.~1998), but in this case the variation of the mass function with density is in the {\\bf opposite} sense to what is observed in Virgo, Fornax, and Ursa. Major -- namely, small galaxies are rarer in the denser environments. It follows from this assumption about the mass function that there must be many galaxies in the diffuse environments that are completely dark and never formed stars (see also the discussion in Section 5.2 of Klypin et al~1999).  There could be many reasons for why this happened, like gas availability (Virgo and Fornax have substantial X-ray gas halos so presumably had a reservoir of gas in the past which has since been gravitationally heated --- Ursa Major does not) and the tidal creation (Barnes \\& Hernquist 1992a) and destruction (Putman et al.~1998 and references therein) of dwarfs. The lack of gas availability for low-mass halos in low-density environments may follow from the heating and ionization of the intergalactic medium by OB stars and supernovae over cosmic time (Klypin et al.~1999; Bullock, Kravstov \\& Weinberg  2000). Whatever the mechanism, it must generate a large effect: there are more than ten times as many $M_R = -11$ dwarfs in Virgo per $M_R = -20$ giant galaxy than in Ursa Major (Phillipps et al.~1998, Trentham et al.~2000). But the existence of large numbers of dark galaxies in the diffuse environments does not depend on the details of this mechanism; it depends only on the assertion that the mass function does not vary. The fact that it is also predicted by CDM theory, and the success of that theory at predicting galaxy properties and large-scale structure on many different scales (e.g.~Jenkins et al.~1998; Fontana et al.~1999; Katz, Hernquist \\& Weinberg 1999; Kauffmann et al,~1999) simply gives additional weight to the hypothesis that these dark galaxies exist. In CDM theories, these dark halos in low density environments will tend to have formed during most of the history of the Universe, except at very early times (then these halos could form stars). They probably are still forming now.  The dark galaxy hypothesis is one way to solve the problem of CDM theories overproducing large numbers of small galaxies without having to abandon the theories and their other successes.  An alternative way to perturb the theories so as to avoid this discrepancy is to force the cold dark matter to be self-interacting (Spergel \\& Steinhardt 2000, Burkert 2000; however see Miralda-Escud{\\'{e}} 2000). In this case the mass function of low-mass galaxies in the field is substantially shallower than $n(M) \\sim M^{-2}$.  If we accept this possibility of many dark galaxies existing, we can then ask the question: can we find them, and how?  This is the subject of the present paper.  The optimum methods of searching for dark galaxies depend on the nature of the dark matter, in particular whether or not any of the dark matter is baryonic. The results also depend on the total masses of the dark galaxies, which are unknown.  Given the arguments in the previous paragraphs, they will be similar to the total masses of the galaxies that created the upturn in the Virgo luminosity function.  These have absolute magnitudes $M_R > -13$, and so have stellar masses of $\\leq 10^7 {\\rm M}_{\\odot}$ (assuming an $R$-band stellar mass-to-light ratio of 1). The total masses are therefore likely to be $\\leq 10^8 {\\rm M}_{\\odot}$, given a cosmic ratio of baryonic matter to halo dark matter of 0.1 (see Section 2 and references therein) and the fact that the fraction of baryonic matter in galaxies which {\\it did} form stars is likely to be close to this number or lower; the present-day masses of the Virgo dwarfs may be lower than this due to tidal stripping of their dark matter within the Virgo cluster but here we are concerned with the initial masses of the halos. We therefore restrict the discussion in this paper to dark galaxies with masses $\\leq 10^8 {\\rm M}_{\\odot}$.  In Section 2 we review current observations that give some constraints on the dark matter composition. In Section 3 we describe ways that dark galaxies can be observed and structure this section according to the possibilities highlighted in Section 2.  In Section 4 we discuss further where are the best places in the sky to make the observations described in Section 3. Throughout this work, we assume the following cosmological parameters: $H_0 = 65\\,\\, {\\rm km} {\\rm s}^{-1} {\\rm Mpc}^{-1}$, $\\Omega_{\\rm matter}=0.3$, $\\Omega_{\\Lambda}=0.7$. The majority of the observational tests that we suggest target local dark galaxies and the strategies do not depend significantly on the cosmology. ",
        "conclusions": ""
    },
    "0010/astro-ph0010129_arXiv.txt": {
        "abstract": "The major cornerstone of future ground-based astronomy is imaging and spectroscopy at the diffraction limit using adaptive optics. To exploit the potential of current AO systems, we have begun a survey around bright stars to study intermediate redshift galaxies at high resolution. Using ALFA to reach the diffraction limit of the 3.5-m telescope at Calar Alto allows us to study the structure of distant galaxies in the near-infrared at scales of 100--150\\,pc for $z$=0.05 and at scales 1.0--1.5\\,kpc at $z$=1. In this contribution we present the initial results of this project, which hint at the exciting prospects possible with the resolution and sensitivity available using an AO camera on the 8-m class VLT. ",
        "introduction": " ",
        "conclusions": ""
    },
    "0010/astro-ph0010403_arXiv.txt": {
        "abstract": "Using FORS2, mounted on Kueyen (UT2 of the VLT), we have obtained the redshift of the lensing galaxy in the gravitational lens system \\object{HE~1104$-$1805}. We measure $z=0.729 \\pm 0.001$, in good agreement with previous estimates based on the time delay and the position of the lens on the fundamental plane.  It also coincides with one of the metal line absorption systems that are seen in high resolution spectra of \\object{HE~1104$-$1805}. ",
        "introduction": "\\object{HE~1104$-$1805} was discovered as part of the Hamburg/ESO Quasar Survey and was first identified as a gravitational lens candidate by Wisotzki et al. (\\cite{Wizot93}).  It consists of two lensed images of a radio-quiet quasar at $z$= 2.319 that are separated by $\\sim$3.2\\arcsec.  The lensing galaxy was discovered from ground based near-IR (Courbin, Lidman \\& Magain, \\cite{Courbin98a}) and HST optical observations (Remy et al. \\cite{Rem1998}; hereafter R98) and is 1\\arcsec\\, from the brighter quasar image (component A).  From a spectrophotometric monitoring program that lasted several years, Wisotzki et al. (\\cite{Wizot98}; hereafter W98) measured a time delay of 0.73 years between the two components. To convert this into an estimate of the Hubble Constant, one needs to determine the geometry of the system (astrometry and lens and source redshifts) and to model the mass distribution of the lens. Several mass models for the lens have been published (W98; R98; Courbin et al. \\cite{Courbin2000a}; hereafter C2000; Lehar et al. \\cite{Lehar00}) and precise astrometry from HST images is available. Since the source redshift is known, the remaining unknown is the redshift of the lens.  In their discovery paper, Wisotzki et al. (\\cite{Wizot93}) noted that the continuum of component A was considerably harder than that of component B, and that the broad emission lines of the two components were identical, only scaled by a factor of 2.8.  This was still the case during the spectrophotometric monitoring carried out by Wisotzki et al. (\\cite{Wizot98}) and in IR spectra covering the 1 to 2.5~$\\mu$m range (C2000). This has been interpreted as microlensing of the continuum emitting region of the quasar in component A, which is twice as close to the lensing galaxy as component B. Presumably, the broad line region is too large to be affected. It may be possible to use this information to gain insight into the geometries and sizes of the continuum emitting and broad line regions; however, detailed predictions require the lens redshift.  From high resolution spectroscopic observations (Smette et al. \\cite{Smette95}; Lopez et al. \\cite{Lopez99a}), several metallic absorption line systems have been detected: $z=0.517, 0.728, 1.280, 1.320, 1.662, 1.860, 2.220$ and $z=2.300$. The systems at $z=0.728$, $z=1.320$ and $z=1.662$ contain lines that are mostly detected in component A, the component that is closest to the lens. At one time or another, each system has been individually proposed to be the lens.  Despite its importance, the redshift of the lensing galaxy has proved elusive.  Apart from the many unpublished attempts to measure it, the redshift of the lens has been estimated by indirect means.  From IR and optical photometry, C2000 gave a redshift in the range $z=0.8$ to $z=1.2$ while from the time delay and a model of the lens, W98 estimated $z=0.79$. From the position of the lens on the fundamental plane, Kochanek et al. (\\cite{Koe20}) derived $z=0.77$. Although model dependent, the two latter estimates prove to be very close to the truth.  In this paper we describe the successful measurement of the lens redshift, $z=0.729$, using the spectral deconvolution method described by Courbin et al. (\\cite{Courbin2000b}).   \\begin{figure*}[t] \\begin{center} \\resizebox{18.0cm}{!}{\\includegraphics{VUV12.f1.ps}} \\caption[]{{\\it Top:} the 2D sky-subtracted spectrum of \\object{HE~1104$-$1805}. This spectrum is the mean of the three 1080s exposures. {\\it Bottom:} deconvolved spectrum. The spectra of the quasar images have a FWHM of 0.2\\arcsec. The pixel size is half that of the original data (hence, the height of this image is twice that of the original), i.e., 0.05\\arcsec. The main features of the lensing galaxy are marked. The scale in the spatial direction is given by the separation between the 2 quasar images, i.e., 3.14\\arcsec.} \\label{2d_spec} \\end{center} \\end{figure*} ",
        "conclusions": "The aim of our observations were modest: to measure the redshift of the lensing galaxy in \\object{HE~1104$-$1805}, one of the ever growing list of lenses with known time delays.  This was possible with only one hour of VLT time during average weather conditions. The prospects of measuring the lens redshifts in other systems are therefore very good, at least for double systems, where aligning the slit simultaneously on the quasar images and on the lens is straightforward.  With the time delay and the lens and source redshifts known and several mass models published, it is relatively straight forward to derive the Hubble constant. W98 present two models to describe the lens: a singular isothermal sphere with external shear and a singular isothermal ellipsoid without external shear. For the former (the model presented by R98 is very similar), one derives $H_0=52$ km/s/Mpc and for the latter, one derives $H_0=59$ km/s/Mpc. We have set $(\\Omega_M,\\Omega_{\\Lambda})=(0.3,0.7)$. C2000 also present two models: a singular ellipsoid without external shear and a two component model which consists of a singular isothermal ellipsoid to represent the lensing galaxy and a more extended component representing a galaxy cluster centered on the lensing galaxy.  The first model leads to $H_0=62$ km/s/Mpc, but the second leads to values for the Hubble Constant which are a factor of 1.5 to 2 lower.  \\object{HE~1104$-$1805} is somewhat unusual in that the lensing galaxy is closest to the brighter component. It has been pointed out by several authors that the lens probably does not consist of a single galaxy (W98, R98, C2000; Lehar et al. \\cite{Lehar00}).  Additional extended mass is required to match all the observational constraints: the position of the two images relative to the lens, the magnification ratio of the images, and the lens orientation and ellipticity. Deep images in the near-IR may be able to reveal the distribution of this additional mass.  As the lens appears to be associated with the absorption system seen in the quasar spectrum at $z=0.728$, one could use the velocity spread in this system to constrain the lens velocity dispersion. However, this assumes that the absorption line system and the lens are the same object or, at the very least, in the same potential well. Without further supporting evidence that links the absorption line system to lens, it is premature to use the velocity spread as a constraint on the mass models.  The lens redshift estimates of W98 and Kochanek et al. (\\cite{Koe20}) have proven to be accurate, whereas the estimate of C2000 from IR and optical photometry is less so. The optical photometry is in good agreement with the VLT spectrum; however, if the lens is an elliptical galaxy, then the IR points are over-estimated by 0.3 magnitudes. The cause of this discrepancy requires further investigation."
    },
    "0010/astro-ph0010650_arXiv.txt": {
        "abstract": "We present wide-field images of the quadruple gravitational lenses B1422+231 and MG\\,J0414+0534 obtained from global Very Long Baseline Interferometry (VLBI) observations at 8.4\\,GHz  on 23 November 1997. We present also a lens model for MG\\,J0414+0534, which reproduces the core positions and flux densities of the VLBI images, combining a singular isothermal ellipsoid with external shear, and a singular isothermal sphere to represent, respectively, the main lens galaxy and its neighbor, a faint galaxy near one of the images. ",
        "introduction": " ",
        "conclusions": ""
    },
    "0010/astro-ph0010466_arXiv.txt": {
        "abstract": "Observations of the starburst galaxy, M82, have been made with a 20-station global VLBI array at $\\lambda$18cm. Maps are presented of the brightest young supernova remnants (SNR) in M82 and the wide-field mapping techniques used in making images over a field of view of $\\sim$1 arcminute with 3 milliarcsecond resolution are discussed. A limit has been placed on the power law deceleration of the young SNR, 43.31+592 with an index greater than 0.73 $\\pm$ 0.11 from observations with the European VLBI Network. Using the global array we have resolved compact knots of radio emission in the source which, with future global observations, will enable better constraints to be placed on the expansion parameters of this SNR.  The latest global observations have also provided high resolution images of the most compact radio source in M82, 41.95+575. We determine an upper limit to the radial expansion rate along the major axis of 2000 km s$^{-1}$. However, the new images also show structure resembling that of collimated ejection which brings into question the previous explanation of the source as being due to the confinement of a supernova by a high density circumstellar medium.  It is apparent that we are now able to image the brightest supernova remnants in M82 with a linear scale which allows direct comparison with galactic SNR such as Cassiopeia A. ",
        "introduction": "M82 is generally considered to be the archetypal `starburst' galaxy. Far-infrared (FIR) and radio luminosities imply a high star-formation rate and at a distance of 3.2 Mpc \\cite{burbidge64} it lends itself well to study with radio interferometry on a parsec scale. The radio studies of M82 have revealed a population of compact radio sources (Unger et al., 1984; Kronberg et al., 1985)\\nocite{unger84}\\nocite{kronberg85} which were tentatively identified as being either radio supernovae (RSn) or supernova remnants (SNR). Subsequent observations with MERLIN at a resolution of 50 mas (0.75 pc) resolved most of the compact sources and the parsec-scale shell-like structure observed in many of the objects confirmed them to be young SNR \\cite{muxlow94}. Furthermore, recent EVN observations at 15 mas (0.2 pc) resolution \\cite{pedlar99} have further resolved the most compact sources and by comparison with a 1986 epoch an expansion speed of $\\sim$9500 kms$^{-1}$ has been measured for the compact shell-like remnant, 43.31+592. Obviously it would be preferable to measure the size evolution of this remnant on much shorter timescales than the $\\sim$10 years required by the EVN observations. A global VLBI array at $\\lambda$18cm provides a maximum resolution of $\\sim$3 mas and enables these measurements to be performed on timescales of $\\sim$2 years and also provides an opportunity to determine the deceleration of the remnants as they interact with the interstellar medium (ISM). The SNR in M82 and other nearby galaxies provide a means of understanding the star-formation processes in a starburst environment and also enable the ISM in a starburst to be probed on parsec scales.  Since the SNR are produced by massive stars with lifetimes of order 10$^{6}$ years, then these SNR trace the recent star formation history of a galaxy. If the supernova rate can be established, then by assuming an initial mass function it is possible to calculate a total star formation rate to compare with that derived from other methods, such as the total far-infrared or non-thermal radio luminosity. Previous VLBI observations of M82 have concentrated on the brightest and most compact source, 41.95+575 (Bartel et al., 1987; Trotman 1996)\\nocite{trotman96}\\nocite{bartel87}. However, we are now able to process the large amounts of data needed to image a large field of view. Therefore, it is now possible to image a field of view of $\\sim$1 arcminute with the 3 mas resolution which the global array provides at $\\lambda$18cm. ",
        "conclusions": "These global VLBI observations have produced the highest resolution images yet made of the young supernova remnants formed in the nuclear starburst within M82. At the full resolution of $\\sim$3 mas images have been made of the brightest and what are inferred to be the youngest sources. For the compact young supernova remnant, 43.31+592, we draw the following conclusions.  (1) The measurements of the shell diameter from 15 mas resolution EVN images \\cite{pedlar99} places a limit on the deceleration parameter. Since the remnant is present in the 1972 image of Kronberg \\& Wilkinson (1975)\\nocite{kronberg75}, a limit of $\\delta >$ 0.73 $\\pm$ 0.11 has been calculated. As yet, this limit does not allow for a physical analysis of the supernova model, but with future observations tighter constraints will be placed on the deceleration of this source.  (2) The higher resolution images presented in this paper will enable proper motion studies to be carried out on the compact knots of radio emission as shown in figure~\\ref{4331final}. Further global VLBI observations are scheduled to take place in late 2000, two years after the initial observations. This represents the minimum time required to detect significant proper motion of the knots assuming a radial expansion speed of 10000 km s$^{-1}$.  (3) For the first time we have been able to image the supernova remnants in M82 on a similar linear scale to that which we can image the galactic supernova remnant, Cassiopeia A.  In addition, we draw the following conclusions for the compact radio source, 41.95+575.  (1) The new image of 41.95+575 reveals structures which represent deviations from its previous description as an elongated shell and we speculate that these may either represent collimated outflow as predicted by the jet-induced explosion models of Khokhlov et al. (1999) or a prominence of the type predicted by Blondin et al. (1996).  (2) We calculate an upper limit to the radial expansion rate along the major axis of 2000 km s$^{-1}$.  The young supernova remnants in M82 fill an important gap in the evolutionary timescale of supernova remnants between the very young (e.g. SN1993J) and those with ages measured in hundreds of years (e.g. those in the Milky Way and Magellanic clouds). Therefore, M82 will continue to act as an important laboratory for the study of young SNR and any differences in evolution which are due to the starburst ISM may become apparent. \\subsection*{Acknowledgements} We acknowledge the contribution of the observatories who participate in the European VLBI Network and the VLBA. We thank Rick Perley at NRAO, Socorro for allowing us to use the VLA image of Cassiopeia A. We thank Tim O'Brien and John O'Connor for useful comments. Andrew McDonald acknowledges the receipt of a PPARC postgraduate research grant."
    },
    "0010/astro-ph0010185_arXiv.txt": {
        "abstract": "The Hilbert-Einstein equations are insufficient to describe the geometry of the Universe, as they only constrain a local geometrical property: curvature. A global knowledge of the geometry of space, if possible, would require measurement of the topology of the Universe. Since the subject was discussed in 1900 by Schwarzschild, observational attempts to measure global topology have been rare for most of this century, but have accelerated in the 1990's due to the rapidly increasing amount of observations of non-negligible fractions of the observational sphere. A brief review of basic concepts of cosmic topology and of the rapidly growing gamut of diverse and complementary observational strategies for measuring the topology of the Universe is provided here. ",
        "introduction": "Is it possible to observe the whole Universe? Is the object studied in what is claimed to be observational cosmology really all of space or just a tiny bit of space?  In order to answer these questions, it is necessary to measure the mathematical properties of {\\em local geometry} (such as the curvature) and {\\em global geometry} (such as the topology), which together describe the `shape'  and size of space, under the assumption that the curvature is nearly constant everywhere in space. Since a century ago, Schwarzschild (1900), de Sitter (1917), Friedmann (1924) and Lema\\^{\\i}tre (1958) have realised that the spatial part of our Universe could correspond to a space (a 3--manifold) which may have either a non--zero curvature and/or a non--trivial topology.  The measurement of these properties (one local and the other global) from surveys obtained at telescopes of different sorts, such as the GMRT, the AAT, the VLT, XMM, MAP and Planck Surveyor, should enable us to find out if our cosmological observations are global in the sense of measuring the whole of space, or whether they simply measure a tiny fraction of the Universe: our observable sphere.  Tests for measuring curvature or topology are dependent to differing extents on assumptions of the cosmological model adopted. Most tests are evaluated in terms of the most popular model, i.e. the `Hot Big Bang' model, or in other words, the perturbed Friedmann--Lema\\^{\\i}tre model, but as long as the cosmological expansion interpretation of redshifts is retained, many of the tests involving `standard candles' or `standard rulers' should also be valid for the quasi steady state cosmology model \\citepf{HBN93}.  In order to aid the non--specialist, some reminders on curvature and topology are provided in \\SS\\ref{s-definegeom}. The application of these geometrical concepts to the standard hot Big Bang model, to extrapolations of the standard model and to the quasi steady state cosmology model are presented in \\SS\\ref{s-models}.  What do the observations tell us? Serious observational work with what may be hoped to be sufficiently deep surveys to determine the global geometry of the Universe have only just started in the last decade, and the race is on to obtain the first significant results. A brief glance at the various strategies using different astrophysical objects or radiation sources and tentative results is described in \\SS\\ref{s-obsvns}.  Comoving coordinates are used to describe space throughout this review.  \\label{s-definegeom} ",
        "conclusions": "Is it possible to show by observations that the global Universe is observable? And that observational cosmology is something more than just extragalactic astrophysics?  For a recent summary of observational results, see table of \\citet{LR99}. The analyses using different observational data sets and different methods have so far answered these questions with `No' on scales which are much smaller than the horizon. All the observations point to the Universe being simply connected up to a scale of $\\sim 1000${\\hMpc}, i.e. about a tenth of the horizon diameter.  At the $\\sim 1000${\\hMpc} scale and larger, definitive answers have not yet been obtained. On the contrary, some candidate 3-manifolds consistent with several observational data sets have been suggested by \\citeauthor{RE97} \\citep{RE97,RBa99,Rouk00a}, \\citet{Rouk00c} and \\citet*{BPS98,BPS00a,BPS00b}.  Specific testing of these large 3-manifolds might show that one of these makes correct predictions of three-dimensional positions (celestial coordinates and redshift) of multiple topological images of previously known objects.  In parallel, developments of all the techniques described above are being actively pursued by the different groups cited above, using computer simulations and different classes of observations.  Strong hopes are put in the MAP satellite which should be launched in the next year or two to map the CMB, but the better resolution and the ability to measure polarisation information in the CMB by the Planck Surveyor might be needed to extract the signal from the noise.  Using three-dimensional methods, numerous surveys such as the 2dF and SDSS galaxy and quasar surveys which are presently underway might be sufficient to obtain significant results.  Theory may be slow in catching up. Quantum cosmology studies regarding the evolution of topology during the quantum epoch are nowhere near making predictions for the present-day topology of space, though some interesting work has begun (e.g. \\citealt{DowS98}).  Not only would a significant measurement of cosmic topology show that the Universe is spatially finite in at least one direction, but it would add topological `lensing' to the tool presently finding a great variety of useful applications: gravitational lensing. The two are similar in that the geometry of the Universe generates multiple images in both cases. They differ in that the former uses the whole Universe as a `lens', does not magnify the image and generates images at (in general) widely differing redshifts and angles, all of which contrast with the latter.  \\noindent{\\bf Acknowledgments}  This review is partially inspired from that of \\citet{Rouk99}, and the author thanks Jean-Philippe Uzan, Varun Sahni and Tarun Souradeep for providing numerous useful comments.    \\def\\apj{ApJ.}                        % \\def\\apjs{ApJ. Supp.}                 % \\def\\aj{A.J.}                           % \\def\\aanda{A\\&A}                      % \\def\\cqg{Class. Quant. Grav.}         % \\def\\mnras{Month. Not. R. Astron. Soc.}  \\newcommand\\joref[5]{#1, #5, {#2, }{#3, } #4} \\newcommand\\epref[3]{#1, #3, {#2}}"
    },
    "0010/astro-ph0010300_arXiv.txt": {
        "abstract": "The NLTE effects affecting oxygen-abundance determinations of solar-type  stars are discussed. LTE is perfectly safe for the forbidden lines. The permitted triplet at 777\\,nm is expected to show NLTE effects on the order of a few tenths of a dex (always in the sense that LTE overestimates the abundance), but the magnitude of the effects is dependent on the still very uncertain cross sections of collisional excitation by collisions with neutral hydrogen atoms. Little is known about the NLTE effects on molecular line formation. ",
        "introduction": "The approximation of local thermodynamic equilibrium (LTE) is commonly used in the analysis of stellar spectra. It simplifies the computation of atomic and molecular level populations immensely by postulating Saha-Boltzmann equilibria, thus defining the state of the gas as a function of only temperature, pressure, and chemical composition. The solution of the equation of radiative transfer for the frequencies of interest is then straightforward. The implied assumption that the radiation field does not influence the properties of the gas is, however, not realistic. Thus we must question the validity of LTE and investigate the NLTE problem of spectral-line formation. This will here for most of the time be taken to mean the solution of the coupled equations of statistical equilibrium and radiative transfer for a trace species in a stellar photosphere. NLTE effects are then simply errors caused by the LTE approximation.  When discussing the NLTE effects on a spectral line, it is often clarifying to separate between effects on the line opacity and on the line source function. The line-opacity effect is described with the NLTE departure coefficient of the atomic population of the lower energy level involved in the transition: $b_i = n_l/n_l^{LTE} $. The departures from LTE of the line source function is described by its ratio to the local Planckian, $S_{\\rm L}/B_\\nu(T)$.  In the following, NLTE effects on lines used for oxygen-abundance determinations in unevolved or moderately evolved stars of approximately solar effective temperatures will be discussed. The discussion focuses only on departures from LTE, thus disregarding for a moment all other relevant uncertainties like missing background opacities, observational errors, stellar parameters, effects of granulation, line-damping treatment, etc. ",
        "conclusions": "For those in quest of accurate oxygen abundances of solar-type stars, it should be noted that: \\begin{itemize} \\item NLTE effects are utterly unimportant for the forbidden oxygen lines. \\item The 777\\,nm triplet lines are not formed in LTE. NLTE effects will not change the slopes in the [O/Fe]--[Fe/H] diagrams very much, but they will surely influence the error bars because of the uncertainties in the collisional cross sections. \\item The NLTE effects on the triplet are always expected to be negative. Thus LTE results define an {\\em extreme} on the possible range of abundances, not a conservative mean hypothesis. \\item NLTE effects on molecular lines should be investigated. \\end{itemize}  \\ack Martin Asplund is thanked for stimulating discussions as well as for comments on the manuscript."
    },
    "0010/astro-ph0010136_arXiv.txt": {
        "abstract": "We present high quality $BVRI$ CCD photometry of the variable star V716 Monocerotis (= NSV 03775).  We confirm it to be an RR Lyrae star of variability type ab (i.e.\\ a fundamental mode pulsator), and determine its metallicity ([Fe/H] $= -1.33 \\pm 0.25$), luminosity ($M_V = 0.80 \\pm 0.06$), and foreground reddening ($E(B-V) = 0.05$--$0.17$) from the Fourier components of its light curve.  These parameters indicate a distance of $4.1 \\pm 0.3$ kpc, placing V716 Mon near the plane of the Galaxy well outside the solar circle. ",
        "introduction": "The classification of variable stars can often depend sensitively on the available observational data.  For example, the periodic rapid brightness increase displayed by RR Lyrae stars can easily be confused with typical variability of cataclysmic variables (such as the start of a dwarf nova outburst) if only limited observations are available.  A number of alleged cataclysmic variables (CVs) have been reclassified as RR Lyrae stars following more extensive observation \\cite[e.g.\\ CG Muscae,][]{LW97}.  As RR Lyrae variables are particularly useful objects for exploring Galactic structure and dynamics, owing to the availability of relationships between their observational properties and their distances, it is desirable to continue to reclaim misclassified RR Lyrae stars from the ranks of the CVs.  V716 Monocerotis (= NSV 03775) was first identified as a variable star by \\citet{hoffmeister49}.  It was categorized as a suspected CV or RR Lyrae variable by \\citet{KK82}.  \\citet{KO95} ruled out the possible CV nature of V716 Mon by obtaining a spectrum (indicative of an A-type star) and several CCD and photoelectric data sets (showing an almost complete cycle of the characteristic light curve of an RR Lyrae star).  \\citet*{haefner96} independently concluded that V716 Mon was an RR Lyrae star based on a steady decrease in its relative brightness of $\\approx0.44$ mag during a 4.75 hr CCD light curve.  The CV catalog and atlas of \\citet*{DWS97} now lists V716 Mon as a ``non-CV,'' thus providing the final word in the tale of its long-running identity crisis.  We observed V716 Mon during the telescope orientation of the 1999 Cerro Tololo Inter-American Observatory (CTIO) summer student programs.  We chose this object for two main reasons: (1) no calibrated photometry currently exists for it in the literature, and (2) it is located in a field that is not too crowded to perform reliable photometry at the plate scale of the CTIO Curtis Schmidt telescope ($2\\farcs32$ pixel$^{-1}$ -- see \\S\\ref{s-obs}). We present here the first calibrated $BVRI$ photometry of V716 Mon.  We use our multi-color light curves to calculate several fundamental parameters of this RR Lyrae star, including its distance and position within the Galaxy. ",
        "conclusions": "We have obtained extensive $BVRI$ photometry of the variable star V716 Mon.  Its period and multi-color light curves indicate that it is an RR Lyrae variable of Bailey type ab (i.e.\\ a fundamental mode pulsator). Using the empirical relations between light curve Fourier components and physical properties (JK96, KJ97), we estimate the star's metallicity to be [Fe/H] $= -1.33 \\pm 0.25$ dex, its absolute magnitude to be $M_V = 0.80 \\pm 0.06$ mag (internal error only), and its reddening to be in the range $E(B-V) = 0.05$--$0.17$ mag.  The resulting distance is $4.1 \\pm 0.3$ kpc, placing V716 Mon near the plane of the Galaxy well outside the solar circle."
    },
    "0010/astro-ph0010246_arXiv.txt": {
        "abstract": "We present a self-consistent, semi-analytical dynamical model of disk driven outflows in AGNs that are accelerated by a combination of magnetic stresses and radiation pressure. This model will make it possible to examine scenarios in which the wind is homogeneous as well as cases where it consists of dense clouds embedded in a more tenuous, magnetized medium. The various ingredients of this model will be tested through quantitative predictions from both multiwavelength spectral observations and reverberation mapping of AGNs. ",
        "introduction": "Many researchers have theorized about the structure and dynamics of the gas near the cores of AGNs, specifically in the Broad Emission Line Region (BELR), Broad Absorption Line Region (BALR), and the Warm Absorber (WA).  In addition, there are open questions as to how all of these different regions might be related, if at all.  In attacking these questions, different investigators have invoked various processes and physical effects to explain observations of AGNs.  Researchers have hypothesized that the gas may exist either as discrete clouds or as continuous winds.  The idea that the gas is partitioned into discrete clouds is the more ``traditional'' approach to BELR modeling (Bottorff et al. 1997).  However, recent spectral analysis by Arav et al. (1997, 1998) seem to show that a continuous wind might be better suited to explain high resolution spectra of broad lines.  Second, to explain the acceleration of the gas to the velocities observed in spectral lines, various researchers have suggested either radiative acceleration or magneto-centrifugal acceleration.  Radiative acceleration from a central continuum source was proposed by Arav, Li, \\& Begelman (1994) and Arav (1996) to explain the acceleration of gas in the BALR.  Later, Proga (2000) examined radiative acceleration due to continuum emitted by a circumnuclear accretion disk as well as the central source. Other researchers have invoked magneto-centrifugal acceleration (K\\\"{o}nigl \\& Kartje, 1994), which appears to explain the central dusty torus through the natural vertical density stratification in the magneto-centrifugal wind.  Of course, investigators don't usually include just one of the above possibilities; many different combinations of mechanisms have been attempted. Emmering, Blandford, \\& Shlosman (1992) modeled the cores of AGNs by using a magneto-centrifugally driven continuous outflow to simulate discrete clouds.  This model was later improved (Bottorff et al. 1997) to explain line variabilities.  Working with radiative acceleration only, Murray et al. (1995) and later Chiang \\& Murray (1996) modeled the gas as a continuous wind, driven by continuum and line radiation forces from both the central source and the disk.  They used the inner edge of the wind as a shield and therefore as a possible origin of the Warm Absorber, and also explained the Broad Emission Line (BEL) profiles with the wind itself.  Finally, de Kool \\& Begelman (1995) used a continuous wind with magneto-centrifugal driving and continuum radiation pressure to examine the effects of radiation pressure on the magnetic outflow structure.  Also, K\\\"{o}nigl et al. (1995) and Kartje et al. (1997) explained EUV absorption features in BL Lac objects using both magnetic and radiative acceleration.  They accomplish this by first lifting material vertically away from the disk using magneto-centrifugal driving (modified by including an approximate radiation pressure in the self-similar wind equations).  The material travels upward along the magnetic streamlines until it intercepts the central beaming cone of the BL Lac, where the gas is forced out radially by both line and continuum radiation pressure.  The clouds move radially outward, absorbing the central continuum at a range of velocities as they accelerate, yielding broad absorption troughs in BL Lac spectra. ",
        "conclusions": ""
    },
    "0010/astro-ph0010070_arXiv.txt": {
        "abstract": "Recent observations with the {\\it Extreme Ultraviolet Explorer} ({\\it EUVE}) suggest that at least some clusters of galaxies are luminous sources of extreme ultraviolet (EUV) radiation. It is not clear yet whether EUV emission is a general feature of clusters; for the purposes of limiting the contribution to the background radiation, we assume that it is true of most clusters. Assuming that the source of the EUV emission is inverse Compton (IC) scattering of the Cosmic Microwave Background photons by relativistic electrons, we construct a simple model for the expected average emission from clusters as a function of their mass and the redshift of interest. Press-Schechter theory is used to determine the abundance of clusters of various masses as a function of redshift. We determine the amount of background radiation produced by clusters. The total mean intensity, spectrum, and the ionization rates for \\ion{H}{1} and \\ion{He}{2} are determined at present and at a variety of redshifts. Because clusters form by the merger of smaller subclusters, the amount of EUV background radiation should be larger at present than in the past. We compare our results to the ionizing background expected from quasars. We find that while clusters do contribute a significant EUV background, it is less than a percent of that expected from quasars. ",
        "introduction": "\\label{sec:intro}  Recently, extreme ultraviolet (EUV) and very soft X-ray emission has been detected from a number of clusters of galaxies. Except for the Coma and Virgo cluster (Bowyer, Lampton, \\& Lieu 1996; Lieu et al.\\ 1996a,b,c; Bowyer, Bergh\\\"ofer, \\& Korpela 1999; Bergh\\\"ofer, Bowyer, \\& Korpela 2000), the individual detections are controversial. Other clusters which may have been detected include Abell~1795 and Abell~2199 (Bowyer, Lieu, \\& Mittaz 1998; Mittaz, Lieu, \\& Lockman 1998), Abell~4038 (Bowyer et al.\\ 1998), and Abell~4059 (Bowyer 2000). The excess EUV detections in the rich clusters (i.e., not Virgo) correspond to luminosities of $\\sim 10^{44}$ ergs s$^{-1}$. The detections of Abell~1795 and Abell~2199 as extended EUV sources have been quite controversial. For both clusters, detections with {\\it EUVE} were claimed by Bowyer et al.\\ (1998) and Mittaz et al.\\ (1998). However, Bowyer, Bergh\\\"ofer, \\& Korpela (1999) argued these detections were due to an incorrect subtraction of the background. In Abell~2199, Lieu et al.\\ (1999b) reobserved the cluster, and continue to claim detection of an extended EUV source. An associated soft X-ray excess may have been seen with {\\it BeppoSAX} (Kaastra et al.\\ 1999). Recently, {\\it XMM} appears to have confirmed the presence of very extended, EUV/soft-X-ray excess emission in Abell~1795 (Arnaud 2000). In addition to the controversies about individual clusters, Arabadjis \\& Bregman (1999) argue that all the detections may be due to uncertainties in Galactic absorption.  Thus, it is not certain at this point whether excess EUV emission is a general characteristic of clusters of galaxies. In order to limit the contribution of cluster EUV emission to the extragalactic background, we will assume that it is, in fact, a general property of rich clusters. If this is the case, then our calculations may give a determination of the cluster contribution to the EUV background. If EUV excess emission is present in only a fraction of rich clusters, then our calculations will give a strong upper limit to the contribution to the extragalactic background.  A promising theory for the EUV emission by clusters is that it is produced by inverse Compton (IC) scattering of the Cosmic Microwave Background (CMB) by relativistic electrons in the intracluster medium (Hwang 1997; Bowyer \\& Bergh\\\"ofer 1998; En{\\ss}lin \\& Biermann 1998; Sarazin \\& Lieu 1998). This requires the presence of a distinct population of relatively low energy relativistic electrons with energies of $\\sim$100 MeV (Bowyer \\& Bergh\\\"ofer 1998) beyond the population of higher energy electrons which produce radio emission in a small fraction of clusters (e.g., Feretti 1999). Such a distinct population is expected theoretically, because the lower energy electrons have lifetime which are comparable to the Hubble time, while the radio emitting electrons have very short lifetimes (Sarazin 1999; Atoyan \\& V\\\"olk 2000; Takizawa \\& Naito 2000).  An alternative theory is that the EUV emission is thermal in origin (Mittaz et al.\\ 1998; Lieu, Bonamente, \\& Mittaz 1999a). There are a number of concerns with this thermal model, particularly with the energetics (Fabian 1996; Sarazin \\& Lieu 1998). Recently, Lieu et al.\\ (1999a) have argued that the observed EUV to soft X-ray spectra of clusters decline rapidly with increasing frequency, and that this favors a thermal model. However, such a rapid decline is, in fact, what was predicted by the IC model (Sarazin \\& Lieu 1998; Sarazin 1999; Atoyan \\& V\\\"olk 2000; Takizawa \\& Naito 2000; Figure~\\ref{fig:spectrum_model} below).  In any case, we will normalize our calculations of the EUV emission from clusters to the observed EUV flux from the Coma cluster, which is the best observed rich cluster. Thus, our results are somewhat independent of the emission mechanism for the EUV emission. In detail, we need a model to scale the EUV luminosity of Coma to other clusters and other redshifts, and we need a model for the spectrum of the emission. At present, there are no detailed thermal models for the EUV emission from clusters which provide such predictions. On the other hand, the IC model provides a very simple scaling of the luminosity and spectrum to other masses of clusters and to other redshifts (\\S~\\ref{sec:emit} below). Thus, we will adopt the IC model for the emission to scale the observed flux of the Coma cluster to other clusters and redshifts.  If this radiation really is a common feature of clusters at the indicated luminosity levels, it might make a contribution to the intergalactic background radiation field at UV through soft X-ray energies. In this paper, we estimate the mean intensity of the diffuse EUV background from galaxy clusters, both at the current epoch and at higher redshifts. We will compare the radiation field from clusters with that produced by quasars and other active galactic nuclei (AGN) (e.g., Haardt \\& Madau 1996). In \\S~\\ref{sec:mass}, we use Press-Schechter (1974) theory to calculate the abundance of clusters of a given mass as a function of redshift. In \\S~\\ref{sec:emit}, the EUV emission of clusters of a given mass is calculated, based on a simple IC theory. We assume that both the thermal energy of the intracluster gas and the energy of relativistic electrons are produced ultimately by intracluster shocks, and that these shocks convert a fixed fraction of the shock energy into relativistic particles. The resulting diffuse radiation field at the present epoch and at higher redshifts are given in \\S~\\ref{sec:results}. The resulting ionization rates of \\ion{H}{1} and \\ion{He}{2} are also derived there. The cluster diffuse EUV is compared to that produced by quasars and other AGN. The conclusions are summarized in \\S~\\ref{sec:conclusion}. ",
        "conclusions": "\\label{sec:conclusion}  Observations with {\\it EUVE} indicate that some clusters of galaxies are luminous sources of extreme ultraviolet (EUV) radiation. In order to limit the contribution of EUV emission from clusters to the extragalactic background, we have assumed that this emission is a common feature of clusters. A promising theory for this emission is that it is due to IC scattering of the Cosmic Microwave Background photons by relativistic electrons. We have given a simple model for the average EUV luminosity expected from clusters based on this theory. We summed this emission over clusters of varying mass at different redshifts, using the Press-Schechter formalism to determine the mass function of clusters as a function of redshift. We determined the amount of background radiation produced by clusters through EUV IC emission. The total mean intensity, spectrum, and the ionization rates for \\ion{H}{1} and \\ion{He}{2} were determined at present and at a variety of redshifts. Because clusters form by the merger of smaller subclusters and the abundance of massive clusters increases with time, the amount of EUV background radiation should be larger at present than in the past. We compared our results to the ionizing background expected from quasars. We find that while clusters do contribute a significant EUV background, it is less than a percent of that expected from quasars. Of course, this last conclusion is strengthened if EUV emission is not a general feature of clusters."
    },
    "0010/astro-ph0010252_arXiv.txt": {
        "abstract": "We have investigated whether motion of gas in a bar-like potential can account for the peculiar but systematic velocity field observed in the nuclear region of the starburst galaxy NGC 253. This unusual velocity field with gradients along both major and minor axes was revealed in a high resolution ($1.8^{\\arcsec}\\times 1.0^{\\arcsec}$) H92$\\alpha$ recombination line observation by Anantharamiah and Goss (1996).  A simple logarithmic potential is used to model the bar. Assuming that the bulk of the gas flows along closed and non-intersecting \\zx1 (bar) and \\x2 (anti-bar) orbits of the bar potential, we have computed the expected velocity field and position-velocity diagrams and compared them with the observations.  A comparison of the integrated CO intensity maps with the spatial distribution of the \\zx1 and \\x2 orbits in the model indicates that the nuclear molecular gas in NGC 253 lies mainly on the \\x2 orbits. We also find that the velocity field observed in the central 100 pc region in the H92$\\alpha$  recombination line is well accounted for by the bar model if most of the ionized gas resides in the inner \\x2 orbits.  However, the model is unable to explain the velocity field on a larger scale of $\\sim 500$ pc observed using the OVRO interferometer with a resolution of $5^{\\arcsec}\\times 3^{\\arcsec}$. The direction of the observed CO velocity field appears twisted compared to the model.  We suggest that this perturbation in the velocity field may be due to an accretion event that could have occurred $10^7$ years ago. ",
        "introduction": "NGC 253 is a nearby, barred Sc galaxy with a nuclear starburst region (${\\rm L}\\sim 3 \\times 10^{10}{\\rm L}_{\\odot}$; Telesco \\& Harper 1980).  The galaxy has been extensively studied at different wavelengths.  The proximity of the galaxy (d$\\sim$ 3.4 Mpc) and the ongoing nuclear starburst makes it an ideal candidate to study enhanced star formation at nearby distances. The galaxy is nearly edge on ($i=78^{o}$) with a considerable amount of dust in the center which makes it bright in the infrared wave bands. Observations of the H$\\alpha$ emission line indicate possible nuclear outflows and stellar winds in the central region, arising from enhanced star formation (Ulrich 1978, Schulz \\& Wegner 1992). Radio continuum observations reveal numerous compact sources within the inner 200 pc which are either supernova remnants or HII regions (Ulvestad \\& Antonucci, 1997).  At near-infrared wavelengths (NIR), a prominent bar can be observed in the center (Scoville et al. 1985) whose position angle is tilted by about 18\\deg with respect to the major axis of the galaxy . Interferometric CO observations with a resolution of $\\sim 5''$ have shown the presence of a molecular bar of dimensions $\\rm30^{\\arcsec}\\times10^{\\arcsec}$, whose orientation is similar to the NIR bar (Canzian, Mundy \\& Scoville, 1988). HCN and CS emissions, which trace dense gas, also reveal a similar bar ( Paglione et al 1995 and Peng et al. 1996). Recent work on the bar of NGC 253 indicate that there is at least one Inner Lindblad Resonance (ILR) point for the galaxy (Arnaboldi et al. 1995).  Various types of data on NGC 253 indicate that the nuclear starburst is mainly located within a circumnuclear ring in the bar (e.g. Engelbracht et al. 1998).  The kinematics of gas in the nuclear region of \\n253 has been known to be anamolous from the early \\ha observations by Demoulin \\& Burbidge (1970). Radial velocity measurements at various position angles near the nucleus show that, in addition to solid body rotation, large non-circular motions exist within the central region (Ulrich 1978, Shulz \\& Wigner 1992, Munoz-Tunon, Vilchez \\& Castaneda 1993). However, due to problems of obscuration at optical wavelengths, results from \\ha and [NII] observations are difficult to interpret. Longer wavelength observations such as that of CO (Canzian, Mundy \\& Scoville 1988), Br$\\gamma$ (Puxley and Brand 1995) and H$_2$ (Prada et al. 1996) indicate steeper velocity gradients along the major axis and only solid body rotation in the central region. Most of the optical and IR observations, aimed at studying the kinematics of the nuclear region of \\n253 (e.g. Arnaboldi et al. 1995, Prada et al. 1996, Prada, Gutierrez \\& McKeith 1998, Engelbracht et al. 1998),  have relied on measuring the velocity gradients along chosen position angles passing though the nucleus (e.g. major axis and minor axis). However interferometric radio observations at millimeter and centimeter wavelength (e.g. Canzian et al. 1988, Paglione et al. 1995, Anantharamaiah \\& Goss 1996, Peng et al. 1996) measure two-dimensional velocity fields which provide a more complete  picture of the kinematics in the nuclear region. These two-dimensional measurements reveal a complex but systematic velocity pattern in the central region. In addition to solid body rotation, the observed velocity fields indicate motions which may be due to a bar-like potential in the center (Peng et al 1996) or a kinematic sub-system which may be caused by a past merger event (Anantharamaiah \\& Goss 1996).  The highest resolution measurement of the two-dimensional velocity field available to date is that of Anantharamiah \\& Goss (1996) who observed the H92$\\alpha$ recombination line with a beam of $\\rm1.8^{\\arcsec}\\times1.0^{\\arcsec}$ and a velocity resolution of 54 \\kms. The line emission is detected in the central $\\rm9^{\\arcsec}\\times4^{\\arcsec}$ (approximately 150 pc$\\times$60 pc) region of \\n253 and oriented roughly along the major axis of the galaxy.  The observed velocity field of the H92$\\alpha$ line emission shown in Figure 1a, has an elongated S-shaped pattern with iso-velocity contours running almost parallel to the major-axis of the galaxy. For pure solid body rotation, the iso-velocity contours are expected to run parallel to the minor-axis. \\anantha showed that this velocity field could be fitted to a set of three orthogonally rotating nested rings of ionized gas. Such an interpretation is however only empirical and has no clear physical basis. The central velocity field of \\n253 has been observed over a larger scale ($\\rm\\sim30^{\\arcsec}\\times 10^{\\arcsec}$) but with coarser angular resolution ($\\rm\\sim3^{\\arcsec}-6^{\\arcsec}$) in molecular lines; CO (Canzian et al. 1988), HCN (Paglione et al. 1995) and CS (Peng et al 1996).  As mentioned earlier the molecular line emission is oriented along the NIR bar which is tilted by 18\\deg with respect to the major axis whereas the H92$\\alpha$ line emission is mainly along the major axis.  The velocity field of the molecular gas is also distinctly different from that of the ionized gas.  Peng et al. (1996) have suggested that the morphology and kinematics of the dense molecular gas can be explained by gas moving in a bar potential.  In this paper, we have attempted to model the inner velocity field of NGC 253 to determine whether the velocity pattern observed by \\anantha can also be explained by motion of gas in a bar potential. We also present in this paper, new observations of the velocity field in the CO line with an angular resolution of $5.6''\\times 2.6''$.  Since gas is a collisional system (unlike stars which can be collisionless), it tends to settle on closed, non-intersecting orbits in the bar potential.  Using a bar model, we have detemined a set of closed orbits from which a velocity field is derived and compared with observation in the H92$\\alpha$ and the CO lines.  While the bar model succesfully accounts for the velocity field observed in the H92$\\alpha$ line in the inner ($9''\\times 4''$) region, it does not account of the observed CO velocities on a larger scale ($40''\\times 15''$).  The only earlier attempt to model the nuclear velocity field of NGC~253 using a bar model is that of Peng et al (1996) who attempted to explain the velocity field observed on the larger scale in the CS line. Although, Peng et al (1996) have reported success of their model, by comparing observed position-velocity (PV) diagrams with PV plots of closed x1 and x2 orbits, we show in this paper that a single bar model cannot explain both the H92$\\alpha$ and the CO (or the CS) velocity fields. ",
        "conclusions": "We have attempted to model the observed nuclear velocity field of ionized and molecular gas in NGC 253 using a simple logarithmic bar potential. The parameters for the potential were derived from the optical rotation curve. The velocity field was determined from the \\zx1 and \\x2 orbits in the bar potential and compared with observations of the H92$\\alpha$ (Anantharamaiah \\& Goss 1996) and CO lines.  The results are summarized below.  \\noindent 1) The \\x2 orbits, projected onto the plane of the sky, lie roughly along the major axis of the galaxy. The \\x2 ring appears very narrow and ridge like. We identify the ionized gas observed in the H92$\\alpha$ line as lying along the inner \\x2 orbits. Also, when we compare the integrated CO intensity maps in the literature with the model we find that the molecular gas is also distributed mainly over the \\x2 orbits.  \\noindent 2)   The velocity field within the inner 8\\sec predicted from the bar potential is similar to that observed by \\anantha.  The model predicts velocity gradients both along the major and minor axes. The isovelocity contours run parallel to the major axis as observed. The model \\pv diagram agrees reasonably well with the observations.  \\noindent 3) The model velocity field  on a larger scale ($\\sim$45\\sec) is significantly different from the observed velocity fields in CO  and CS (Peng et al. 1996).  However, the model \\pv diagrams agree reasonably with the observations. We therefore conclude that agreement in \\pv diagrams is a necessary but not sufficient condition to explain the complete velocity field.  \\noindent4) The observed velocity field on a larger scale ( $\\sim$45\\sec) appears to be a twisted version of the model velocity field with the direction of the gradient changing with radius.  We suggest that this perturbation may have been caused by the accretion of an object with mass of $\\sim 10^{6} M_{\\odot}$ about $10^{7}$ years ago. This accretion event may also have triggered the nuclear starburst observed in the galaxy. Enhanced optical images of NGC 253 (Malin 1981, Beck et al 1982) show a distorsion in the outer regions of the galactic disk which may be a signature left by the merger event."
    },
    "0010/astro-ph0010408_arXiv.txt": {
        "abstract": "We present some phenomenological aspects of the pre-big-bang cosmological model inspired by the duality properties of string theory. In particular, assuming the spatial sections of the homogeneous background geometry to be isotropic, we discuss the quantum production of perturbations of the background fields (gravitons, dilatons, moduli fields), as well as the production of particles which do not contribute to the background, which we call ``seeds''. As such we consider the cases of electromagnetic and axionic seeds.  We also discuss their possible observational consequences, for example, we study whether they can provide the origin of primordial galactic magnetic fields, and whether they can generate the initial fluctuations leading to the formation of large-scale structure and the measured cosmic microwave background anisotropies. We finally analyze axion and photon production in four dimensional anisotropic pre-big-bang cosmological models. ",
        "introduction": "The pre-big-bang scenario (PBB) is a cosmological model inspired by the duality properties of string theory~\\cite{pbb1,pbb2,lwc}. It differs from conventional cosmological models based on general relativity by the presence of the dilaton field.  In the PBB model, the initial state of the universe is the string perturbative vacuum, instead of a hot and dense state as predicted by the standard cosmological model. In the string (S) frame -- in which weakly coupled strings move along geodesic surfaces~\\cite{sf} -- the low-energy string effective action in $D$ dimensions, neglecting both finite-size effects and loop corrections, can be written as \\bea \\label{action} S&=&-{1\\over 16 \\pi G_{\\rm D}}\\int d^D x ~~\\sqrt{|g_{\\rm D}|} ~~e^{-\\phi} \\nonumber \\\\ &\\times & \\left[R_{\\rm D}+ \\partial_\\mu\\phi\\partial^\\mu\\phi - {1\\over 12}H_{\\mu\\nu\\alpha} H^{\\mu\\nu\\alpha}+ V(\\phi)\\right ]~+~S_{\\rm M}~, \\eea where $ G_{\\rm D}$ is Newton's constant in $D$ dimensions and $R_{\\rm D}$ is the Ricci scalar of the $D$-dimensional space-time. In Eq.~(\\ref{action}), $H_{\\mu\\nu\\alpha}$ denotes the field strength of the two-index antisymmetric tensor $B_{\\mu\\nu}$ (i.e., $H=dB$), $\\phi$ is the dilaton field and $V(\\phi)$ is the dilaton potential.  During the pre-big-bang era when the effective coupling $e^{\\phi/2}$ is small, it is usually assumed that the dilaton potential ($V\\sim \\exp[-\\exp(-\\phi)]$) can be neglected. On the other hand, during the post-big-bang era the dilaton potential vanishes, leading to a constant massive dilaton sitting at the minimum.  The action $S_{\\rm M}$ describes ``bulk'' string matter satisfying the classical string equations of motion in a given background.  The low-energy string effective action, Eq.~(\\ref{action}), leads to a set of cosmological equations for a homogeneous and spatially flat background which obey the ``scale-factor duality''~\\cite{pbb1}: if the background fields are only time-dependent, and if $\\{a_i(t),\\phi(t); i=1,...,D-1\\}$ denote the scale factors and dilaton of a given homogeneous exact solution with vanishing antisymmetric tensor, then the system $\\{\\tilde{a}_i(t),\\tilde{\\phi}(t); i=1,...,D-1\\}$, obtained through the transformation in $(D-1)$ spatial dimensions: \\be a_i\\rightarrow \\tilde{a}_i=a_i^{-1}~,~\\phi\\rightarrow \\tilde{\\phi}=\\phi-2\\sum_i\\ln a_i~, \\ee is a new exact solution of the cosmological equations~\\cite{pbb1,dt}. In the presence of matter, which can be approximated as a perfect fluid, a scale factor duality remains a symmetry of the solutions if we add the transformation law $p/\\rho \\rightarrow -p/\\rho$~\\cite{pbb1}.  The solution of the string cosmology equations with the perturbative vacuum as initial condition, describes a phase of growing curvature and growing dilaton. It is called the ``dilaton phase''. Once the curvature $H^2$ (where $H=\\dot a/a$; a dot stands for differentiation with respect to cosmic time $t$) reaches the string scale $\\lambda_{\\rm s}^{-1}$, the background enters the ``string phase'' during which  higher derivative terms in the $\\alpha'$ expansion become important.  During this phase, the curvature scale is supposed to remain constant while the string coupling increases.  At the beginning of the string phase, the string coupling can be arbitrarily small; this is one of the parameters of the model.  At the end of the string phase, the dilaton approaches the strong coupling regime (the coupling constant is $\\sim 1$), leading to the transition to the radiation-dominated era characterized by  decelerated evolution and frozen dilaton field. The duration of the string phase is the second parameter of this cosmological model inspired by string theory.  A complete description of the high curvature strong coupling regime and its transition to a radiation-dominated post-big-bang universe with frozen dilaton is still lacking. Some ``toy models'' and approximations have however been studied~\\cite{vb,cc}.  By combining the duality transformation with time-reversal symmetry ($t\\rightarrow -t$), another symmetry of the model, we can always associate to any given decelerated expanding post-big-bang solution with decreasing curvature: \\be \\dot a > 0~~,~~\\ddot a < 0~~, ~~\\dot H < 0~, \\ee an accelerated expanding pre-big-bang solution with growing curvature: \\be \\dot a > 0~~, ~~\\ddot a > 0~~, ~~\\dot H > 0~. \\ee Transforming this accelerating solution into the Einstein (E) frame --- in which the dilaton coupling is absorbed into the metric by a conformal transformation, i.e., the graviton and dilaton kinetic terms are diagonalized --- through the conformal $(D-1)$-dimensional rescaling~\\cite{pbb2}: \\be a\\rightarrow a ~e^{-\\phi/(D-2)} ~~~,~~~ dt\\rightarrow dt ~e^{-\\phi/(D-2)}~, \\ee the pre-big-bang solution transforms into a Friedmann universe with accelerated contraction and growing curvature \\be \\dot a < 0~~, ~~\\ddot a < 0~~, ~~\\dot H < 0~. \\ee Let us consider as an example~\\cite{pbb1}, a spatially flat isotropic background in $D=4$ space-time dimensions with vanishing antisymmetric tensor and dilaton potential and with non-zero bulk string matter. The scale factor duality maps the standard radiation-dominated solution with frozen dilaton \\be a\\sim t^{1/2}~~,~~\\phi={\\rm const.}~~,~~\\rho=3 p~~,~~t>0 \\ee into the pre-big-bang solution \\be a\\sim (-t)^{-1/2}~,~\\phi=-3\\ln(-t)~,~\\rho=-3 p~~,~~t<0 \\ee which describes an accelerated expansion of the pole- or super-inflation type~\\cite{pole} with growing curvature and gravitational coupling.  The phenomenological drawbacks (i.e., horizon, flatness problems) of the standard cosmological model can be equally well solved, either by a sufficiently long era of super-inflation in the string frame, or an era of accelerated contraction in the Einstein frame.  Thus, the choice of frame is just a matter of convenience, and the physical description of the PBB scenario is equivalent in the S and E frames~\\cite{pbb2}. In this paper we usually work in the string frame, otherwise we specify it.  The usual relations with the adiabatic decrease of temperature $T$: \\be \\rho \\sim 1/a^{\\rm D} \\sim T^{\\rm D} ~~~,~~~ T\\sim 1/a \\ee in the radiation-dominated background, are mapped~\\cite{pbb2} to: \\be \\rho \\sim 1/a^{{\\rm D}-2} \\sim 1/T^{{\\rm D}-2} ~~~,~~~ T\\sim a \\ee in the dual pre-big-bang background, where the temperature increases with the scale factor $a$. Going from the phase with scale factor $a$ to the dual one with scale factor $1/a$, the temperature $T$ remains unchanged, since we applied not only a duality transformation but also a time reversal, changing the range of cosmic time $t$ from $[-\\infty,0]$ to $[0,\\infty]$.  The PBB scenario attempts to describe the very early universe near the Planck scale, when general relativity breaks down. It should not be considered as an alternative to the conventional standard cosmological model supplied by inflation, but as a completion to it. Moreover, in full string theory the pre-big-bang solution is expected to be smoothly connected to a standard post-big-bang universe whose evolution is described by general relativity.  One area of activity in string cosmology studies precisely this high curvature/high coupling phase which has to be described by string theory, either via corrections to the low curvature/low coupling equations or by some other means.  Another area of research deals with those observational consequences of the PBB model which are most probably independent of this transition and which distinguish the PBB model from other inflationary cosmologies. One then also wants to study how well different observables agree with present observational constraints.  The cosmological implications of string theory, and more precisely, the comparison of the predictions of the PBB scenario with observations, will provide a test for string theory as a fundamental theory and fix some of the parameters of the PBB model.  However, despite the attractive features of the PBB scenario, one should keep in mind that there still remain theoretical and phenomenological aspects which are not yet fully understood. Among the phenomenological aspects of the PBB scenario, vacuum fluctuations and the corresponding particle production are important issues and they will be discussed in this paper. In what follows, we calculate the spectra of the produced particles and indicate some possible observational consequences.  We consider phenomenological aspects of the pre-big-bang solutions derived from the low-energy string effective action. We decompose the spatial sections of the full $D$-dimensional space-time into a three-dimensional external expanding sub-manifold and a $(D-4)$-dimensional internal sub-manifold. The internal spatial dimensions shrink down to the final compactification scale, which is typically given by the fundamental length parameter of string theory.  At conformal time $\\eta=-\\eta_1$, when string corrections become important, we make an instantaneous transition from the pre-big-bang to the post-big-bang era where the internal dimensions as well as the dilaton are supposed to be frozen while the three external dimensions keep expanding like in an ordinary radiation-dominated Friedmann universe. A crucial but reasonable assumption, is that the spectra of the produced particles on scales much larger than $\\eta_1$, are not influenced by the details of the transition from the pre- to the post-big-bang era. This has been verified numerically with some toy models for the pre- to post-big-bang transition~\\cite{cf}.  In what follows, we neglect the effects of an intermediate string phase. More precisely, in our model the universe starts at the dilaton driven era and then enters the radiation-dominated era. It is possible that the intermediate string phase is long and affects a considerable region of the spectrum. Then, the spectra we calculate are still valid for all modes which exit the horizon before the string phase.  Our paper is organized as follows.  In Section~II we discuss the quantum creation of background perturbations (gravitons, dilatons and moduli fields). In the context of the PBB scenario, there is an accelerated shrinking of the event horizon during the super-inflationary phase~\\cite{pbb2}, which ends when a maximal curvature scale is reached and higher-derivative terms can no longer be neglected. After an intermediate string phase, the universe enters the decelerating, radiation-dominated phase~\\cite{pbb1,pbb2,sf}. We shall see that, for generic matching conditions one obtains red dilation and moduli spectra while the graviton spectrum is blue. This is due to the fact that, during the pre-big-bang phase, scalar perturbations (the dilaton and moduli perturbations) grow on super horizon scales. Hence long wave-length modes which leave the horizon early have more time to grow. On the other hand, gravity waves are roughly constant and given by the curvature at horizon exit which is larger for the scales exiting later, leading to a blue graviton spectrum, in contrary to standard inflation where curvature remains essentially constant which implies a flat spectrum.  In Section~III we discuss the quantum creation of fields which do not contribute to the background. We call them ``seeds''.  Seeds are produced by the amplification of quantum fluctuations of fields which are present in string theory but are not part of the homogeneous background.  In particular, we study electromagnetic (EM) seeds and their r{\\^ o}le for primordial galactic magnetic fields, and Kalb-Ramond axion  seeds and their r{\\^ o}le for large-scale structure and the cosmic microwave background (CMB) anisotropies. Both cases are characteristic of the PBB scenario, since in the conventional picture based on general relativity, there is no Kalb-Ramond axion and electromagnetic perturbations cannot be excited due to their conformal coupling to the metric and the absence of a dilaton field.  Up to this point we assume the spatial section of the four-dimensional homogeneous background geometry to be isotropic, in other words there is a single scale factor. This assumption will be dropped in Section~IV, where we analyze perturbations in four-dimensional anisotropic pre-big-bang models. In particular, we study the creation of axions and photons and compare the result with the isotropic case. In Section~V we state our main conclusions.  Notation: Cosmic time is denoted by $t$, and conformal time by $\\eta$; they are related by $t=\\int a d\\eta$, where $a$ is the scale factor. Correspondingly, a dot stands for derivative w.r.t. cosmic time $t$, and a prime stands for derivative w.r.t. conformal time $\\eta$.  The scale factors are $a$ (and $b$ in the anisotropic case). The dilaton field is $\\phi$ or $\\varphi$, the modulus field is ${\\cal B}$, and the axion field is $\\sigma$.  The spectral index of the perturbation spectrum is denoted by $n$.  The gauge-invariant Bardeen potentials are $\\Phi$ and $\\Psi$.  The critical density is $\\rho_{\\rm c}=3M_{\\rm Pl}^2H^2/(8\\pi)$, where $H= a'/a^2=\\dot a/a$.  The transition scale in units of the Planck mass $M_{\\rm Pl}$ is $g_1= H_1/M_{\\rm Pl}$. Since the universe is radiation dominated at time $\\eta_1$, the fraction of the total (critical) energy density in radiation at a given time $\\eta$ is $\\Omega_{\\gamma}(\\eta)=(H_1/H)^2(a_1/a)^4$.  Latin indices $i,j$ take values $1,2,3$, while capital letters $A,B$ are equal to $1,...,D-4$, where $D$ stands for the dimensionality of the space-time. The subscripts $\\rm E$, $\\rm S$, $\\rm A$ stand for Einstein, string and axion frames respectively. ",
        "conclusions": "We have analyzed some phenomenological aspects induced by particle production in the PBB scenario, which is a particular cosmological model inspired by the duality properties of string theory. Assuming that the transition from the pre- to the post-big-bang era will not affect the observational consequences of the PBB model, we compare the theoretical observables with current observational data. In doing so, we provide a test for string theory as a fundamental theory and fix some of the parameters of the PBB model.  We first considered a $D$-dimensional space-time, containing a four-dimensional homogeneous and isotropic external metric and a $(D-4)$-dimensional compactified internal metric, with vanishing axion contribution. We study the evolution of perturbations that may be generated by the parametric amplification of vacuum fluctuations as the universe goes from the pre- to the post-big-bang era. We study scalar and tensor metric perturbations. We find that scalar perturbations have generically red spectra while tensor perturbations have blue spectra. A robust prediction of the PBB model is $n_{\\rm T}=3$, while standard inflationary models require $n_{\\rm T}<0$. Dilaton and moduli field perturbations  in the standard pre-big-bang model may acquire red spectra with a spectral index $n=0$. This is of course very different from a scale-invariant Harrison-Zel'dovich spectrum with $n=1$ and leads to large perturbations on large, cosmologically relevant scales during the radiation era. Therefore the ordinary pre-big-bang model is in conflict with observations if the $n=0$ spectrum is inherited. However, if one includes an exponential dilaton potential, the dilaton and moduli perturbations can induce scale invariant adiabatic perturbations in the radiation era like ordinary inflation which are in perfect accordance with present observations e.g. of CMB anisotropies. The blue spectra of the tensor modes (gravity waves), which are normalized to $g_1^2$ at the high frequency end, decay rapidly towards longer wavelengths and are completely negligible on cosmological scales.  Even if a field does not contribute to the background evolution, quantum fluctuations cannot be neglected. The induced energy density perturbation is then of second order in the field perturbation, but it can lead to appreciable perturbations in space-time geometry. Perturbations of fields which do not contribute to the background are referred to as ``seeds''. We analyze electromagnetic and Kalb-Ramond axion seeds, and study their r\\^ole for the origin of primordial galactic magnetic fields, the large-scale structure and CMB anisotropies. We compute the stochastic fluctuations of the energy-momentum tensor of the seeds and determine their contribution to the multipole expansion of the temperature anisotropy.  We find that electromagnetic perturbations lead to a ``blue'' power spectrum whose amplitude is fixed at the string scale; on larger scales, it decays too fast to produce primordial magnetic fields which may be amplified to the presently observed values.  However, since the contribution of electromagnetic perturbations to the large-scale anisotropy is negligible, the COBE normalization does not impose any constraints to the production of seeds for galactic magnetic fields. Kalb-Ramond axions, which are either massless or  have a mass up to $100 MeV$ can lead to a flat or slightly blue spectrum, in reasonable agreement with current data. The actual value of the axion spectral index depends on the rate of contraction of the internal dimensions during the pre-big-bang era.  We also consider four-dimensional spatially flat anisotropic PBB cosmological models. Computing the energy spectra for massless Kalb-Ramond axions, we find that, when integrated over directions, this model leads to infra-red divergent spectra, as in the case of a four-dimensional spatially flat isotropic model. Similarly, analyzing photon production in this background, we find that the obtained blue spectra do not differ significantly from the isotropic case. Therefore, also the four-dimensional anisotropic pre-big-bang models suffer from the unphysical red axion spectrum and the blue photon spectrum found in the four-dimensional isotropic pre-big-bang models.  The analysis presented here is valid for modes which exit the horizon before the string phase. However, if the intermediate string phase is sufficiently long, the spectra can be affected on sufficiently large scale which may be of interest for cosmology. Some examples of spectra from significantly long string phase have been explored in Refs.~\\cite{axionmass,7,Buon}). \\vspace{0.2cm}\\\\ {\\bf Acknowledgment}\\\\ We have benefited from discussions with Alessandra Buonanno, Ed Copeland, Gabriele Veneziano and Filippo Vernizzi. The work of R.D. and K.E.K. is supported by the Swiss NSF."
    },
    "0010/astro-ph0010122_arXiv.txt": {
        "abstract": "Two \\rosat\\ high resolution images separated by nearly five years have been used to determine the expansion of the X-ray remnant of Tycho's supernova (SN1572). The current expansion rate averaged over the entire remnant is $0.124\\pm0.011 \\,\\% \\,\\rm yr^{-1}$, which, when combined with the known age of the remnant, determines the mean expansion parameter $m$, defined as $R\\propto t^m$, to be $0.54\\pm0.05$. There are significant radial and azimuthal variations of the X-ray expansion rate. The radial expansion in particular shows highly significant evidence for the more rapid expansion of the forward blast wave as compared to the reverse-shocked ejecta, an effect that has not been seen previously.  The expansion parameter varies from $m=0.71\\pm0.06$ at the outermost edge of Tycho's supernova remnant (SNR) to a value of $m=0.34\\pm0.10$ on the inside edge of the bright rim of emission. These values are consistent with the rates expected for a remnant with constant density ejecta evolving into a uniform interstellar medium during the ejecta-dominated phase of evolution.  Based on the size, age, and X-ray expansion rates, I obtain values for the explosion energy and ambient density of $E\\approx 4-5\\times 10^{50}\\,\\rm ergs$ and $n_0 \\approx 0.35-0.45\\, \\rm cm^{-3}$.  As is also the case for Cas A and Kepler's SNR, the X-ray expansion rate of Tycho's SNR appears to be significantly higher than the radio expansion rate.  In the case of Tycho's SNR, however, the difference between radio and X-ray expansion rates is clearly associated with the motion of the forward shock. ",
        "introduction": "Measurement of the current rate of expansion provides essential information on the dynamical state of supernova remnants (SNRs). This is particularly true for the so-called historical remnants, those for which the date of the supernova (SN) explosion is known, since a comparison of the average expansion rate with the current expansion rate yields a measure of the deceleration.  Unfortunately, the rate of expansion, although rapid compared to most astrophysical objects in the cosmos, is still rather long on human timescales, necessitating measurements over the course of years, if not decades, in order to attain accurate results.  \\rosat\\ was the first X-ray observatory that had sufficient angular resolution ($\\sim$4$^{\\prime\\prime}$ half-power radius for the High Resolution Imager) and operated for a long enough time that significant measurements of the expansion rate of young SNRs are possible.  \\par  The time-averaged expansion rate of Tycho's SNR based on the outermost extent of the remnant ($\\sim$8$^\\prime$ in diameter) in either the radio or X-ray band and its well-known age is $\\sim$0\\farcs56 yr$^{-1}$.  The observed proper motion of the optical filaments, which are believed to trace the location of the blast wave, indicate much lower current expansion rates ranging from 0\\farcs18 yr$^{-1}$ to 0\\farcs28 yr$^{-1}$ (Kamper \\& van den Bergh 1978).  Evidently these filaments are locations where significant deceleration of the SN blast wave has occurred.  From the width of the broad H$\\alpha$ emission Smith et al.~(1991) derive shock velocities in the range 1500--2800 km s$^{-1}$ that, when combined with the proper motion measurements, imply a distance of 1.5--3.1 kpc.  This range is in good agreement with other distance estimates to the remnant (Green 1984; however, see Schwarz et al.~1995) and in the following I adopt a value of 2.3 kpc for the distance.  Because the optical filaments cover only a limited portion of Tycho's SNR, the optical data are unable to provide a comprehensive picture of the expansion of the remnant.   \\par  The radio remnant of Tycho's SN has also been observed to be expanding at the current epoch (Strom, Goss, \\& Shaver 1982; Tan \\& Gull 1985; Reynoso et al.~1997, hereafter R97). Although in each successive analysis the radio data have improved, the basic result of these studies have remained in general agreement. The current radio expansion rate, averaged over the outer rim, is 0.113 \\% yr$^{-1}$ or, expressed equivalently, the expansion parameter, defined as $R\\propto t^m$, is $m=0.471\\pm0.028$.  This result is between the free expansion rate, $m\\sim 1$, and the expansion rate expected for a remnant in the Sedov phase of evolution, $m=0.4$. There is significant azimuthal variation of the radio expansion rate, while interior features appear to show the same expansion rate as the rim (R97).   Vancura, Gorenstein, \\& Hughes (1995), using data from two satellite observatories, quoted a current X-ray expansion rate for the SNR based on a 11.5 yr time baseline that was consistent with these other values. Here I present more accurate results on the X-ray expansion of Tycho's SNR using high resolution images accumulated by the \\rosat\\ satellite in two epochs separated by 4.55 yr.  A preliminary report on this work, using a different analysis approach, was given by Hughes (1997). ",
        "conclusions": "Numerous authors (Chevalier 1982; Dwarkadas \\& Chevalier 1998; Truelove \\& McKee 1999) have modeled the expansion rates of young supernova remnants.  Dwarkadas \\& Chevalier (1998) in particular examined how different assumed density profiles for the SN ejecta affect the evolution of the resulting remnant, assumed to be interacting with a uniform density ISM.  They considered three principal cases: a power law profile ($\\rho \\propto r^{-7}$), an exponential profile, and a constant density profile.  The expansion parameter uniquely defines the age and radius of the remnant, conventionally expressed in scale-free variables.  The constants of proportionality between the scaled values and the true physical radius and age depend on the three independent dimensional parameters: explosion energy $E$, ejecta mass $M_{\\rm ej}$, and ambient density $\\rho_0$.  Thus, given the known size $R=2.8 (D/2.3\\,\\rm kpc)\\,\\rm pc$ and age $t=425\\,\\rm yr$ of Tycho's SNR it is possible to determine two of the three dimensional parameters.  Here I assume that the ejecta mass is $1.4\\,M_\\odot$ and solve for the other two quantities.  The power-law profile predicts a maximum expansion parameter of $m=0.57$ which is too low to be consistent with the X-ray expansion rate.  The other two model ejecta profiles can accommodate the high rate observed for the forward shock; however, the inferred values of $E$ and $\\rho_0$ are quite different in the two scenarios. Compared to the uniform density case, the expansion parameter for the exponential profile model falls more rapidly with time, so that for a given value of the expansion parameter the scaled age and radius are smaller for the exponential profile.  The inferred dimensional parameters for the exponential profile are $E_{51} = E/10^{51}\\, {\\rm erg} \\approx (0.1 - 0.2)\\, (M_{\\rm ej}/1.4\\, M_\\odot)\\,(D/2.3\\,{\\rm kpc})^2$ and $n_0 = \\rho_0/\\mu_{\\rm H} \\approx (0.004 - 0.08)\\, (M_{\\rm ej}/1.4\\, M_\\odot)\\,(D/2.3\\,{\\rm kpc})^{-3}\\,\\rm cm^{-3}$ ($\\mu_{\\rm H}$ is the mean mass per hydrogen atom), rather low values for Tycho's SNR.  On the other hand, more appropriate values are obtained using the uniform density ejecta model: $E_{51} \\approx (0.4 - 0.5)\\, (M_{\\rm ej}/1.4\\, M_\\odot)\\,(D/2.3\\,{\\rm kpc})^2$ and $n_0 \\approx (0.3 - 0.6)\\, (M_{\\rm ej}/1.4\\, M_\\odot)\\,(D/2.3\\,{\\rm kpc})^{-3}\\,\\rm cm^{-3}$.  These are fully consistent with the values that Hamilton, Sarazin, \\& Szymkowiak (1986) found in their study of Tycho's X-ray spectrum in which they also assumed a uniform density ejecta profile. Thus it appears that both the expansion of the forward shock and the X-ray emission properties of Tycho's SNR can be well explained with this simple model. What about the slower motion of the reverse-shocked ejecta?  Truelove \\& McKee (1999) have parameterized the evolution of both the forward and reverse shocks in young SNRs for a number of cases, including a uniform density ejecta model. For the range of scaled ages that describe the expansion of the forward shock, the reverse shock expansion parameter is $m_{\\rm RS} = 0.51-0.63$.  This rate is in reasonable agreement with the measured expansion parameter of the interior portions of Tycho's SNR ($m = 0.45\\pm0.04$). (Although equating the motion of the ejecta to the reverse shock is not strictly correct, it serves as a reasonable first approximation.)  An acceptable joint fit ($\\chi^2 = 2.7$ for 1 d.o.f.) to the measured X-ray expansion parameters is obtained for best-fit values of $m\\approx0.64$ (forward) and $m\\approx0.49$ (reverse).  The inferred values of $E$ and $n_0$ in this case are similar to those quoted above.  The forward shock expansion rate implies a shock velocity of $4600\\pm400\\, (D/2.3\\,{\\rm kpc})\\,\\rm km\\,s^{-1}$, which in turn implies a mean post-shock temperature of $kT_S = {3\\over 16}\\, \\mu m_p v_S^2 = 25\\pm4 (D/2.3\\,{\\rm kpc})^2\\,\\rm keV$ for a mean mass per particle of $\\mu =0.61$, which assumes a fully-ionized plasma with 10\\% helium. This temperature is quite a bit higher than the estimate, by Hwang, Hughes, \\& Petre (1998), of the post-shock electron temperature of the blast wave in Tycho's SNR, $kT_e \\approx 4\\,\\rm keV$.  This difference most likely arises from either the non-equilibration of electron and ion temperatures at the shock front or the partition of a significant fraction of the shock energy in Tycho's SNR into relativistic particles as was recently found to be the case in SNR E0102.2$-$7219 (Hughes, Rakowski, \\& Decourchelle 2000).  Discriminating between these possibilities will be the focus of a future article."
    },
    "0010/astro-ph0010314_arXiv.txt": {
        "abstract": "We present the first high-resolution X-ray spectrum of the Seyfert 1.5 galaxy NGC 4151. Observations with the Chandra High Energy Transmission Grating Spectrometer reveal a spectrum dominated by narrow emission lines from a spatially resolved (1.6 kpc), highly ionized nebula. The X-ray narrow-line region is composite, consisting of both photoionized and collisionally ionized components. The X-ray emission lines have similar velocities, widths, and spatial extent to the optical emission lines, showing that they arise in the same region. The clouds in the narrow-line region must contain a large range of ionization states in order to explain both the optical and X-ray photoionized emission. Chandra data give the first direct evidence of X-ray line emission from a hot plasma ($T\\sim10^7$ K) which may provide pressure confinement for the cooler ($T=3\\times10^4$ K) photoionized clouds. ",
        "introduction": "NGC 4151 (z=0.00332) is a well-studied Seyfert 1.5 galaxy. It is known to be extended in soft X-rays from Einstein and ROSAT observations \\citep{ebh83,mwe95}. The nature of the extended X-ray emission was postulated to be either thermal emission from a collisionally ionized plasma or scattered nuclear flux. \\citet{wyh94} modeled the ASCA spectrum by 3 components: an absorbed power law plus reprocessed emission from a warm absorber and an extra scattered component. Variability was observed in the column density of the warm absorber.  Spectral complexity from unresolved emission lines and ionized absorption edges has been seen in the residuals of power-law fits to the spectra of many Seyfert galaxies. Recent observations with Chandra have revealed narrow absorption and emission lines in the spectra of Seyfert 1 galaxies NGC 5548 ~\\citep{kml00} and NGC 3783 ~\\citep{kbn00}. These two sources are dominated by continuum emission from the central source, so absorption lines are more prominent than emission lines. On the other hand, the nucleus of NGC 4151 is highly absorbed, giving a good view of its narrow emission line region.  In this paper we present Chandra High Energy Transmission Grating Spectrometer (HETGS) observations of NGC 4151. We identify and measure a large number of narrow emission lines in the  X-ray spectrum. The high spectral resolution provided by HETGS allows us to study the ionization, temperature, and kinematics of the X-ray emitting regions. We discuss the association of the extended X-ray emission with the narrow-line region, and interpretations of the multiple X-ray emitting components in the spectrum. ",
        "conclusions": "The power of Chandra HETGS to elucidate the nature of the X-ray emission from active galactic nuclei is demonstrated by high resolution spectra and images of NGC 4151. Direct emission from the nucleus is highly absorbed. Strong, narrow X-ray emission lines are seen from the spatially resolved NLR of NGC 4151. The narrow radiative recombination continuum features and the O {\\sc vii} $f/r$ ratio are consistent with photoionization from the hard nuclear X-ray source. There is also narrow line emission from a hot, collisionally ionized component in the extended NLR. This may come from the long-sought intercloud medium or shocked plasma. In addition, we find strong narrow Fe {\\sc i} K$\\alpha$ emission from the extended NLR. The composite nature of the NLR indicates that both photoionization and collisional heating are important. The extent and kinematics of both X-ray photoionized clouds and hot thermal plasma match those of the optical NLR, showing that they are closely connected. We also suggest that the narrow X-ray emission lines seen in some other Seyfert galaxies may come from the NLR, and may be interpreted in the light of these NGC 4151 observations."
    },
    "0010/astro-ph0010191_arXiv.txt": {
        "abstract": "There are now observations of several emission lines from the ``Warm Ionized Medium\" (WIM) (or, equivalently, the ``Diffuse Ionized Gas\") of the local ISM, the \\pa\\ in the Milky Way, and also in several other galaxies. Interesting features of these observations include the great strength of \\nt\\lam6563 ($\\sim$ \\ha\\ in some cases) and the fact that \\st\\lam6717/\\nt\\lam6583 is almost the same ($\\sim$0.6 -- 0.7) in all locations and objects. Other line ratios (e.g., \\ot\\lam5007/\\hb) vary considerably.  This paper presents simple photoionization models that reproduce the observed spectra, providing extra heating beyond that supplied by photoionization is assumed (see Reynolds, Haffner, \\& Tufte 1999). The same extra heating was used for models of all stellar temperatures being combined together, although it could easily depend on \\ts.  With observed gas-phase abundances (not solar), the line ratios in the local arm at $b$ = 0$\\arcdeg$ are fitted with no extra heating and (S/H) = 13 ppm, as opposed to solar ($\\sim$ 20 ppm). Local gas observed at $b$ = $-35\\arcdeg$ requires extra heating of about \\gam\\ = 0.75, where \\gam\\ is the extra heating in units of $10^{-25}$ erg H$^{-1}$ s$^{-1}$. In the \\pa\\, there are similar results: little extra heating is required at \\z\\ = 500 pc, and \\gam\\ = 3.0 is needed at \\z\\ = 1.2 kpc. To fit the observations, the gas-phase composition in the \\pa\\ must be reduced as required by the Galactic abundance gradient observed for \\htr{}s. The requirements for \\ngc\\ (the best observed other galaxy) at \\z\\ = 1 kpc and 2 kpc are similar to the \\pa: little or no extra heating near the plane (1 kpc in this case) and \\gam\\ $\\sim$ 3 at \\z\\ = 2 kpc. In \\ngc\\ there is also an increase of \\lam5007/\\ha\\ with \\z\\ that can only come about if most of the ionizing radiation is supplied by very hot stars (type O4: \\ts\\ $\\sim$ 50000 K). Either their radiation must propagate from the plane to high \\z\\ through very little intervening matter, or else the stars are located at high \\z. The total power requirement of the extra heating is $\\lesssim$ 15\\% of the power to photoionize the WIM without extra heating.  Extra heating enhances \\otw\\lam3727/\\hb. Figure 1 shows that there is a spread in the predicted values, but the ratio can serve as a useful diagnostic of extra heating. The [S~III]\\lam9065,9531 lines (see Figure 2)are not useful in diagnosing extra heating. ",
        "introduction": "It is well known that very diffuse ionized gas is a major component of the interstellar medium (ISM) of the Milky Way (Reynolds 1991, 1993). In some (but not all) other galaxies about 40\\% of the \\ha\\ is diffuse (Rand 1996; Hoopes, Walterbos, \\& Greenawalt 1996; Domg\\\"orgen \\& Dettmar 1997; Greenawalt et al. 1998; Otte \\& Dettmar 1999), as opposed to H~II regions (although it is not completely clear that the definition of WIM is the same for extragalactic objects as for the Galaxy.) The diffuse gas will be referred to as the ``Warm Ionized Medium\" (WIM) in this paper, although it is also called ``the Reynolds layer\" for the Milky Way gas and also the ``diffuse ionized gas.\" The properties of the WIM, especially regarding its heating, are reviewed by Reynolds, Haffner, \\& Tufte (2000).  The source(s) of the ionization of the WIM have been debated, but photoionization by radiation from OB stars seems to be the only viable source for the bulk of the material. Even for the local Milky Way, the power required for the ionizations is very large: 10 -- 15\\% of the energy produced by OB stars, or all of the total power from supernovae. It is not obvious how the ionizing radiation from OB stars can penetrate the diffuse H~I that is observed in all directions from the Sun. However, it seems likely that OB stars ionize enough of the neutral material surrounding them to allow the required fraction of ionizing radiation to escape (Miller \\& Cox 1993; Dove, Shull, \\& Ferrera 2000). Since the Sun does not possess this ionizing power, there are apparently no directions in our sky that allow ionizing radiation produced by the Sun to escape to great heights from the Galactic plane.  The hypothesis that the WIM is photoionized by hot massive stars is supported by the agreement of its spectrum with models of very dilute H~II regions (e.g. Domg\\\"orgen \\& Mathis 1994; Greenawalt, Walterbos, \\& Braun 1997; Martin 1997; Wang, Heckman, \\& Lehnert 1998; Sembach et al. 2000). The main reason for the general agreement is that the observed spectrum is heavily biased to low stages of ionization (\\nt, \\st), with weak lines of the higher stages that are abundant in ordinary H~II regions ([O~III], [S~III]). This bias is a natural feature of nebulae photoionized by comparatively dilute radiation fields. The models predict the observed anticorrelation of \\nt/\\ha\\ and \\st/\\ha\\ with the intensity of \\ha. However, [O~I]\\lam6300 is weak in the only two Galactic WIM regions for which it has been measured (Reynolds 1985, Reynolds et al 1998), both in the galactic plane. This line is strong only at the outer edges of H~II regions, where there is both O$^0$ (with abundance locked to H$^0$ by resonant charge exchange) and warm free electrons. Present models of the local WIM (e.g., Sembach et al. 2000) assume that some of the ionizing radiation escapes from the Galaxy, so the resulting \\lam6300 is not detected. The model reflects this leakage by being matter bounded, with the fraction of neutral H, $X$(H$^0$), $\\sim$0.1 at the outer edge instead of $\\sim$1.  The heating of the WIM is less well understood than the ionization. Reynolds \\& Cox (1992) suggested that some form of heating in addition to that resulting from photoionization might be required. At present there are several new observations that make the case for extra heating very strong (Reynolds, Haffner, \\& Tufte 1999), and suggest that the extra heating, relative to the photoelectric heating, decreases with density. Thus, photoionization heating dominates in dense regions, such as near the midplane or in bright H~II regions, but extra heating may become stronger than photoionization heating in the very diffuse WIM.  The present paper will interpret several sets of observations in a general way. The emphasis will be to compare observations relatively close to the plane of the respective locations with those farther from the plane. Photoionization models will be constructed with simple geometries (i.e., volume averages of uniformly dense nebulae.) Combining the photoionization models from assumed individual stars in order to mimic the actual complicated situation will result in conclusions that are plausibility arguments rather than detailed models of the real situations. Firm conclusions can be reached as regarding heating requirements beyond photoionization. More elaborate modeling seems unjustified in view of the geometrical complexity, such as fractal structure, that is found in the actual ISM.  The new observations that I have chosen to model are discussed in \\S2 and listed in Table 1. The general characteristics of the photoionization models are discussed in \\S3, and results regarding individual regions are in \\S4. Discussion are given in \\S5 and a summary in \\S6. ",
        "conclusions": "}  In spite of significant differences among the spectra of the WIM in the local ISM, the \\pa, and \\ngc, there a considerable similarity in the simple photoionization models fitting them. No extra heating is required, and little tolerated, at low \\z, where \\nt/\\ha\\ = 0.5 -- 0.6. At larger \\z, where \\nt/\\ha\\ $\\sim$1, \\gam\\ $\\sim$ 3 is needed in at least the outer regions of the plasma where N and S are largely singly ionized. There is little evidence from these photoionization models regarding the relative amounts of extra heating in the regions near the exciting star(s), where the photoionization heating dominates the extra heating if it is assumed to be uniform.  At large \\z\\ in the \\pa\\ and \\ngc, very hot stars (\\ts\\ $\\ge$ 45,000 K) are needed to explain the comparatively large \\nt/\\ha, \\st/\\ha, and \\oo/\\ha\\ ratios observed. These high stellar temperatures might represent the hardening of the radiation escaping from denser \\htr{}s closer to the plane, since the softer photons are absorbed first by the H and He. In \\ngc very hot stars, and not hardening of the stellar radiation, are needed to explain the \\ot\\ observed at \\z\\ = 2 kpc. For the low values of $q$ required for the WIM, the gas producing the hardening would absorb the O$^+$ ionizing photons required for the \\lam5007 line. These hot stars could be in the plane of the galaxy, powering a superbubble (Dove et al. 2000) if there is very little gas between them and \\z\\ = 2 kpc.  These combinations of models fail to predict precisely what values of the ionization parameter, or $q$, are required to produce the observed spectra. For any particular object there is a rather wide range of values that are satisfactory. The other geometrical uncertainties, such as using volume integrals, simplistic stellar atmospheres, or requiring the same heating for all stellar temperatures at a given location, make predictions of the dilution of the radiation field quite speculative.  The propagation of the ionizing photons from the hot stars to the WIM may be the most serious objection to the idea the photoionization is the dominant process operating in the WIM, since only a very small amount of neutral H is completely opaque. The fractal nature of the ISM is shown in molecular clouds (e.g., Elmegreen \\& Falgarone 1996; Stutzki et al. 1998) and in H~I (Green 1993). The percolation of ionizing radiation through hierarchical density structures has not yet been investigated quantitatively at an adequate level, but probably ionizing photons from an O star can penetrate large distances through the lowest density regions within the fractal ISM.  While photoionization from stars seems to be very plausible for providing most of the ionization of the WIM, it is also very likely that other sources contribute. Candidates are shocks (Raymond 1992) caused by supernovae, as well as turbulent mixing layers (Slavin, Shull, \\& Begelman 1993) produced by shear instabilities of outflowing very hot gas. The spectra of both of these phenomena are really produced by photoionization, but the photons are emitted from hot plasma instead of stellar photospheres. The contributions of metagalactic radiation upon the WIM at the largest distances from the plane may be important, especially for producing \\ot. While \\ot\\ is surprisingly bright relative to \\ha\\ at large \\z\\ in \\ngc, the total power required for it is not large. The energy requirements of the WIM at modest \\z\\ ,say, $\\sim$ 1 kpc, are the most problematic constraint on the fraction of the WIM that is contributed by these alternative processes.  The energy requirements of the extra heating can be determined from the details of the models. The detailed models show that the photoelectric heating from a \\ts\\ = 50,000 K star is 30\\% of the power required for the ionization. This number is approximately $kT_*/ \\chi(\\rm H)$. This is expected, since $kT_*$ is the mean energy of a black body spectrum above the ionization threshold $\\chi(\\rm H)$. For \\ts\\ = 35,000 K, the extra heating power at \\z\\ = 1200 pc in the \\pa\\ (i.e., \\gam\\ = 3) is 65\\% of the photoelectric heating, integrated over the volume, and 30\\% of the photoelectric heating for \\ts\\ = 50,000 K. The power requirement of the extra heating is more modest in comparison to the total power needed, because we must then include the ionization energy in addition to the heating. At \\z\\ = 1200 pc the extra heating is 10\\% total power requirement of the WIM. The extra heating power, integrated over $z$ from 500 pc to 1200 pc, is $\\sim$ 5\\% of the total power requirement of the WIM. The power requirements for still larger \\z\\ are even smaller, since the intensity of the WIM decreases rapidly with height.  While \\lam3727 is the largest single coolant when extra heating is small, it is only about 20\\% of the total cooling, with \\nt\\ and \\sth\\ not far behind it. The \\lam3727 is, of course, very sensitive to extra heating because of its high excitation energy as compared to the other optical lines.  Its importance relative to the other forbidden lines increases with \\gam\\ because its emissivity increases faster than that of the red lines.  However, its overall share of the cooling remains about constant because of the onset of cooling by H$^0$ from collisional excitation of the Lyman lines. The collisional cooling exceeds \\lam3727 at about \\gam\\ = 1.5, but the precise boundary is set by the abundances and stellar temperatures.  It is well known that [C~II] 158 \\mic\\ is a major coolant in the neutral ISM, as shown by models of photo-dissociation regions (PDRs) on the outer parts of molecular clouds (e.g., Kaufman et al. 1999). The {\\em COBE} satellite showed (Wright et al. 1991) that 158 \\mic\\ has a strength greater than the sum of all other infrared lines. The present models show that [C~II] 158 \\mic\\ is not a major coolant in the WIM, although most of the gas phase C is singly ionized. The reason is that other coolants become dominant, while 158 \\mic\\ is not boosted much by the relatively high temperature in the WIM as compared to PDRs. A simple calculation shows the effect. Let $j_{158}(T)$ be the emissivity (erg cm$^{-3}$ s$^{-1}$) of 158 \\mic\\ and similarly for \\nt\\lam6583. We find a mean $T$ from setting $I_{\\lambda 6583} / I_{{\\rm H}\\alpha} = j_{\\lambda 6583}(T) / j_{{\\rm H}\\alpha}(T)$, where the $j$'s involve the appropriate ionic abundances. The most favorable case for 158 \\mic\\ is using the largest (N/H), 75 ppm, and the lowest $I_{\\lambda6583}/ I_{{\\rm H}\\alpha}$, 0.60, resulting in $T$ = 7555 K. In this case, $j_{158}/j_{6583} = 0.33$, which may seem significant. However, \\lam6583 provides only a small part of the total cooling, according to models that predict \\lam6583/\\ha\\ = 0.6. For these models, the cooling of \\lam6583 is $\\sim$12\\% of the total, so 158 \\mic\\ is only 4\\% of the total. The \\pa\\ and \\ngc\\ models have lower (N/H) and (O/H) and, therefore, have hotter nebulae, making the 158 \\mic\\ even less important.  The nebular temperature is raised by both extra heating and a shift of the contribution of hot stars to the ionization, so extra heating cannot be clearly identified until it dominates the photoionization heating.  Figure 1 shows that this takes place for \\otw\\lam3727 at \\gam $\\sim$ 2. Below this value, the influences of stellar temperatures and abundances of coolants are comparable with those of extra heating. However, He$^+$/H$^+$ is unaffected by extra heating. Very hot stars are needed if \\heh\\ approaches unity, especially for the low-density situation in the WIM. \\ngc\\ is an excellent example of this situation.  It is significant that the combinations of models at low \\z\\ in the Milky Way cannot tolerate much extra heating, while extra heating is required at large \\z.  The heating might arise from turbulence (Minter \\& Spangler 1997; Minter \\& Balzer 1997). For models with uniform density, this extra heating can also be assumed to represent heating that is uniform per unit volume, such as magnetic reconnection (Raymond 1992; Gon\\c{c}alves et al. 1993; Birk, Lesch, \\& Neukirch 1998; Vishniac \\& Lazarian 1999). See Haffner et al. (1999) for discussion and references.  Heating by PAHs is produced by a change of their state of ionization (Bakes \\& Tielens 1994). Variations in the interstellar radiation field seen along the line of sight should affect the spectra of the near-infrared ``Unidentified Infrared Bands\" (UIBs) in the diffuse ISM and also the extra heating if PAHs are responsible for it. Observations of the spatial variations of the UIBs, and correlations between these changes and the extra heating as shown by \\nt/\\ha\\ and \\st/\\ha, will be crucial in determining if PAHs are the main source of extra heating.  Complete uniformity of the UIB spectra, suggesting that the distribution of stages of ionization of PAHs is uniform in various environments, would not support heating by PAHs.  While there are many free parameters in the combinations of models, there are also strong constraints. The composition was taken to be the same at high and low \\z, and subject to observational values and the galactic abundance gradients for the local gas and the \\pa. Up to 5 line ratios were fitted with 2 stellar temperatures and one relative fraction of ionizing photons. There are some combinations of observations that cannot be predicted by photoionization models.  Our range of $q$ considered ($-4.80\\le\\log(q) \\le -2.60$) implies \\nt$\\ge$ 0.4\\ha, with the minimum occurring for cool stars, \\gam\\ = 0, and gas bounded models with large central voids. My models all have \\oo\\ $\\le$ 0.6\\hb, with the maximum occurring for hot stars, the largest \\gam\\ I considered (4.5), the lowest $q$, and a large central void. We never predict an \\ot\\lam5007/\\hb\\ $<$ 0.15 combined with \\heh\\ $\\ge$0.9 (i.e., \\ts\\ $\\sim$50,000 K because of the generally low stage of ionization present, or small $q$).  At the outer edge of the model, forbidden-line cooling (mainly from [S\\,II]) cannot keep the temperature below $\\sim$ 10000 K if there is strong extra heating that is constant per H nucleus. Collisional ionization of H begins to produce a low but important fraction ($>$ 1\\%) of H$^+$, along with corresponding electrons. In this case, large amounts of [S~II]\\lam6717 and [O~I]\\lam6300 are produced ($\\sim$ twice \\ha) without any photoionization at all. Since such strong \\oo\\ has not been observed, we conclude that collisional ionization and collisional excitation induced by the extra heating cannot produce most of the observed \\ha. Clearly, a physical theory is needed to determine the level of H ionization at which various processes provide extra heating, and to estimate the conditions under which it becomes negligible in comparison to photoelectric heating. The present models arbitrarily end the extra heating when the fraction of H$^0$ is $>$ 0.5, but values of 0.1 or 0.9 give essentially the same results. However, the model cannot be continued to completely neutral H, which never occurs with constant extra heating because of collisional ionization of H."
    },
    "0010/astro-ph0010472_arXiv.txt": {
        "abstract": "We review the possible roles of large scale shocks as particle accelerators in clusters of galaxies. Recent observational and theoretical work has suggested that high energy charged particles may constitute a substantial pressure component in clusters. If true that would alter the expected dynamical evolution of clusters and increase the dynamical masses consistent with hydrostatic equilibrium. Moderately strong shocks are probably common in clusters, through the actions of several agents. The most obvious of these agents include winds from galaxies undergoing intense episodes of starbursts, active galaxies and cosmic inflows, such as accretion and cluster mergers. We describe our own work derived from simulations of large scale structure formation, in which we have, for the first time, explicitly included passive components of high energy particles. We find, indeed that shocks associated with these large scale flows can lead to nonthermal particle pressures big enough to influence cluster dynamics. These same simulations allow us also to compute nonthermal emissions from the clusters. Here we present resulting predictions of $\\gamma$-ray fluxes. ",
        "introduction": " ",
        "conclusions": ""
    },
    "0010/astro-ph0010158_arXiv.txt": {
        "abstract": "We assess supernova (SN)--driven pregalactic outflows as a mechanism for distributing the product of stellar nucleosynthesis over large cosmological volumes prior to the reionization epoch. SN ejecta will escape the grasp of halos with virial temperatures $T_\\vir\\gta 10^{4.3}\\,$K (corresponding to masses $M\\gta 10^8\\,h^{-1}\\,\\msun$ at redshift $z=9$ when they collapse from 2--$\\sigma$ fluctuations) if rapid cooling can take place, and a significant fraction of their baryonic mass is converted into stars over a dynamical timescale. We study the evolution of SN--driven bubbles as they blow out from subgalactic halos and propagate into the intergalactic medium (IGM), and show that to lift the halo gas out of the potential well the energy injection must continue at least until blow--away occurs. If the fraction of ionizing photons that escape the dense sites of star formation into the intergalactic space is greater than a few percent, pregalactic outflows will propagate into an IGM which has been pre--photoionized by the same massive stars which later explode as SNe, and the expansion of the metal--enriched bubbles will be halted by the combined action of external pressure, gravity, and radiative losses. The collective explosive output of about ten thousands SNe per $M\\gta 10^8\\,h^{-1}\\,\\msun$ halo at these early epochs could pollute vast regions of intergalactic space to a mean metallicity $\\langle Z\\rangle=\\Omega_Z/\\Omega_b\\gta 0.003$ (comparable to the levels observed in the \\Lya forest at $z\\approx 3$) without much perturbing the IGM hydrodynamically, i.e. producing large variations of the baryons relative to the dark matter. Rayleigh-Taylor instabilities between the dense shell which contains pristine swept--up material and the hot, metal--enriched low--density bubble may contribute to the mixing and diffusion of heavy elements. The volume filling factor of the ejecta is higher than 20\\% if the star formation efficiency is of order 10\\%. Larger filling factors (not required by current observations) may be obtained for larger efficiencies, moderately top--heavy IMFs, halos where a significant fraction of the gas is in a galactic disk and does not couple to the outflow (as matter is ejected perpendicularly to the disk), or from a population of more numerous sources -- which would therefore have to originate from lower amplitude peaks.  When the filling factor of the ejecta becomes significant, enriched material will typically be at a higher adiabat than expected from photoionization. ",
        "introduction": "In currently popular cosmological scenarios -- all variants of the cold dark matter (CDM) cosmogony with different choices for the parameters $\\Omega_M, \\Omega_\\Lambda, \\Omega_b, h, \\sigma_{8}, n$ -- some time beyond a redshift of 15 the gas within halos with virial temperatures $T_\\vir\\gta 10^4\\,$K [or, equivalently, $M\\gta 10^9\\, (1+z)^{-3/2}h^{-1}\\,\\msun$]\\footnote{Throughout this paper we will assume, unless stated otherwise, an Einstein--de Sitter (EdS) universe with $H_0=100\\,h\\,\\kmsmpc$.}\\, cooled rapidly due to the excitation of hydrogen \\Lya by the Maxwellian tail of the electron distribution, and fragmented. Massive stars formed with some initial mass function (IMF), synthesized heavy elements, and exploded as Type II supernovae (SNe) after a few$\\times 10^7\\,$yr, enriching the surrounding medium; these subgalactic stellar systems, aided perhaps by an early population of accreting black holes in their nuclei, generated the ultraviolet radiation and mechanical energy that reheated and reionized the universe. While collisional excitation of molecular hydrogen may have allowed the gas in even smaller systems [virial temperatures of only a few hundred K, corresponding to masses around $10^7\\,(1+z)^{-3/2}h^{-1}\\,\\msun$] to cool and form stars at earlier times (Couchman \\& Rees 1986; Haiman, Rees, \\& Loeb 1996; Tegmark \\etal 1997), H$_2$ molecules are efficiently photo--dissociated by stellar UV radiation, and such `negative feedback'  could have suppressed molecular cooling and further star formation inside very small halos (e.g. Haiman, Abel, \\& Rees 2000; Ciardi, Ferrara, \\& Abel 2000).  Throughout these crucial formative stages, the all--pervading intergalactic medium (IGM) acted as a source for the gas that is accreted, cools, and forms stars within these subgalactic systems, and as a sink for the metal--enriched material, energy, and radiation which they eject. The well--established existence of heavy elements like carbon, nitrogen, and silicon in the \\lya forest clouds at $z=3-3.5$ may be the best evidence for such an early episode of pregalactic star formation. The detection of weak but measurable \\CIV and \\SIV absorption lines in clouds with \\HI column densities as low as $10^{14.5}\\,\\cmm$ implies a minimum universal metallicity relative to solar in the range $[-3.2]$ to $[-2.5]$ at $z=3-3.5$ (Songaila 1997). There is no indication in the data of a turnover in the \\CIV column density distribution down to $N_{\\rm CIV}\\approx 10^{11.7}\\, \\cmm$ ($N_\\nHI\\approx 10^{14.2}\\,\\cmm$, Ellison \\etal 2000); the analysis of individual pixel optical depths may actually imply the presence of weak metal lines below the detection threshold (Cowie \\& Songaila 1998; Ellison \\etal 2000). Widespread enrichment is further supported by the recent observations of \\OVI absorption in low density regions of the IGM (Schaye \\etal 2000).  In this paper we argue that the observed paucity of regions in the IGM which are of truly primordial composition -- or have abundances as low as those of the most metal--poor stars in the Wilky Way halo (Ryan, Norris, \\& Beers 1996) -- may point to an early enrichment epoch by low--mass subgalactic systems, rather than being due to late pollution by massive galaxies. Whilst outflows of metal--rich gas are direcly observed in local starbursts (e.g. Heckman 1999) and $z\\approx 3$ Lyman--break galaxies (Pettini \\etal 2000), most of this gas may not leave these massive galaxies altogether, but remains trapped in their gravitational potential wells  until it cools and rains back onto the galaxies. By contrast, metal--enriched material from SN ejecta is far more easily accelerated to velocities larger than the escape speed -- about $80\\,\\kms$ at the center of a $10^8 h^{-1}\\,\\msun$  Navarro--Frenk--White (1997, hereafter NFW) halo at $z=9$ -- in the shallow potential wells of subgalactic systems (Larson 1974; Dekel \\& Silk 1986). These early protogalaxies -- with masses comparable to those of present--day dwarf ellipticals -- are then expected to release significant amounts of kinetic energy, heat, and heavy elements into the surrounding intergalactic gas. Many authors have addressed the impact of SN--driven winds from small starbursting galaxies at high--$z$ (e.g. Barkana \\& Loeb 2001; Ferrara, Pettini, \\& Shchekinov 2000; Scannapieco \\& Broadhurst 2001; MacLow \\& Ferrara 1999; Murakami \\& Babul 1999; Nath \\& Trentham 1997; Voit 1996; Tegmark, Silk, \\& Evrard 1993) or from massive galaxies at more modest redshifts (Aguirre \\etal 2000; Theuns, Mo, \\& Schaye 2001) on the thermal and chemical state of the IGM. Gnedin \\& Ostriker (1997) and Gnedin (1998) have argued that violent merging between protogalaxies may be an alternative mechanism to transport metals in the low density IGM, while Barkana \\& Loeb (1999) have discussed the possibility that enriched material in halos with virial temperatures $\\lta 10^4\\,$K could be photoevaporated at reionization. Here we give an idealized assessement of pregalactic outflows as a mechanism for distributing the products of stellar nucleosynthesis over large volumes. We will be concerned exclusively with explosive multi--SN events operating on the characteristic timescale of a few $\\times 10^7\\,$ yr, the lifetime of massive stars. In a complementary study, Efstathiou (2000) has recently shown that a large fraction of the baryonic mass of galaxies with virial temperatures $\\approx 10^5\\,$K can also be expelled in `quiescent' mode, i.e. over the relatively long timescale of 1 Gyr. To anticipate the conclusions of this work, we find that the IGM will be polluted over large scales at the end of the `dark ages' if a fraction greater than a few percent of the baryonic mass of subgalactic halos can be converted into stars over a dynamical timescale. In the case of large star formation efficiencies and/or moderately top--heavy IMFs, the temperature of vast regions of the IGM will be driven to a higher adiabat than expected from photoionization, so as to inhibit in these regions the formation of further protogalaxies by raising the Jeans mass. ",
        "conclusions": "In the previous section we have modeled the evolution of SN--driven bubbles as they blow out from subgalactic halos and propagate into the intergalactic medium prior to the reionization epoch. We have seen that SN ejecta will escape the grasp of halos with virial temperatures $T_\\vir\\gta 10^{4.3}\\,$K at $z=9$ (when they collapse from 2--$\\sigma$ fluctuations) if a significant fraction of their baryonic mass can be converted into stars over a dynamical timescale. Depending on the star formation efficiency, IMF, and detailed physics of the expansion, we find that after about half of the then Hubble time these outflows will have produced a warm, $10^5$~K metal--enriched low--density region, with a size between a few and several ten times larger than the virial radius of the source halo, surrounded by a colder, dense shell which contains most of the swept--up mass. At $t=10^8\\,$yr, typical shell bulk velocities range between $12\\,\\kms$ (for $f_\\star=1\\%$) to $40\\,\\kms$ ($f_\\star=10\\%$) and up to $60\\,\\kms$ ($f_\\star=50\\%$). If the fraction of ionizing photons which escape the dense sites of star formation into the intergalactic space is greater than a few percent, SN--driven bubbles will propagate into a pre--photoionized medium, and their expansion will be halted by external pressure, gravity, and radiative losses. When the velocity of the outflow becomes subsonic, the enriched material gets dispersed by the random velocities in the IGM and the bubble looses its identity.  Knowing the mass within the bubble and the metal yield per supernova we can now derive the metallicity volution of the bubble. We have used the compilation of Todini \\& Ferrara (2000) --essentially based on the Case A of Woosley \\& Weaver (1995) -- who give the chemical composition of the ejecta for zero metallicity Type II SNe (at redshift 9 the universe may be too young to host Type Ia SNe) as a function of their mass. At each time step of the simulation, we check for the massive stars that went supernovae; we sum over the mass of the heavy elements produced by them and divide by the mass of the hot gas within the bubble at that time. This is the typical metallicity inside the bubble if metals can be efficiently mixed within the cavity. Generally speaking, the metal--rich ejecta will be separated from the shell--evaporated gas by a contact discontinuity. The main  problem then is to disrupt this surface. When the cooling time in the cavity becomes shorter than its age, a cooled shell will form which rapidly becomes unstable (Gull 1973) to Rayleigh--Taylor (RT) and Kelvin--Helmoltz (KH) instabilities. The `fingers' produced by RT will penetrate deep into the high metallicity gas and may be eroded by KH instabilities due to the passage of rapidly moving hot gas, i.e. a mixing layer, perhaps creating  a rather uniform distribution of metals inside the bubble.  Further mixing, this time between the shell and the bubble gas, may occur after the expansion stalls. At that time, due to the restoring gravitational force due $g$ on the shell from the DM halo, the acceleration vector points from the dense shell to the rarefied bubble, a RT unstable situation. The growth time of the RT instability on spatial scales $\\lambda_s$ is \\begin{equation} \\label{RTgrow} t_{\\rm RT} \\simeq \\left[{{2\\pi g \\over \\lambda_s } {(\\Delta - 1)\\over (\\Delta + 1)}}\\right]^{-1/2} \\simeq \\left({2\\pi g \\over \\lambda_s }\\right)^{-1/2}. \\end{equation} If the density contrast between the shell and the hot cavity gas is $\\Delta \\equiv n_s/n_h \\gg 1$, and ${g} = 3 \\times 10^{-11}$ cm s$^{-2}$ at $r=10$~kpc, $t_{\\rm RT}$ is less than 10 Myr for any reasonable thickness (tens of parsecs) of the shell. Hence mixing of shell material with the interior metal enriched gas may be quite rapid. Ferrara, Pettini, \\& Shchekinov (2000) have recently pointed out that additional processes related to mixing of the gas caused by time--varying gravitational tidal forces produced by halo interactions must be invoked if the presence of metals at lower redshifts is found to be ubiquitous. The time scale of the process is difficult to estimate as it depends on the peculariar velocities of the halos, which are predicted to be low at high redshift. Note that the metal enrichment of the hot gas inside the bubble will increase its cooling rate, which may become comparable with inverse Compton cooling. This is achieved around metallicity $\\approx 0.5 Z_\\odot$ at $z=9$. While our treatment may then underestimate the cooling rate, we have seen in the previous section that cooling has a weak effect on the overall dynamics of the ejecta.  The time evolution of the bubble average metallicity, $Z_b$, is shown in Figure \\ref{fig12} for the five cases studied. Its value is affected by the dynamical evolution of the bubble through the injection by evaporative of mass from the shell into the cavity. At later times, after blow--away, $Z_b$ rapidly approaches a constant value determined by the fact that (i) metal production ceases with the last explosion after $t=35$ Myr, and (ii) the evaporative mass flow rate (\\ie the mass input in the bubble) drops, being extremely sensitive ($\\propto T^{5/2}$) to the (rapidly decreasing) temperature. The metallicity becomes then the ratio of two constants, and is determined by the past dynamical evolution of the outflow. The final values are in the narrow range $Z_b = 0.2-0.4\\,Z_\\odot$, with the only exception of the peculiar case with $M_l=30\\,\\msun$, which has $Z_b=3\\times 10^{-2}\\,Z_\\odot$. In Table 1 we show the expected bubble and shell masses at stalling. The final ratio $M_b/M_s$ is of order $1$\\%. Thus, if the heavy elements were completely mixed into the swept--up shell, then the resultant metallicities would be lower than those derived above by about two orders of magnitude. It is possible, though, that the heavy elements may remain restricted to a region with a modest volume filling factor, and would not perhaps penetrate into the underdense medium between the halos. As already mentioned, the outcome depends on the onset of RT instabilities: non-linear development of these instabilities would lead to more widespread dispersal of the heavy elements through the IGM.  The details  are sensitive to the sound speed within the bubble when it stalls: this speed depends on how much small--scale internal mixing there has been within the bubble, due to conductivity, etc; but it could be substantially higher than the escape velocity. If so, then pressure--driven `fingers' of hot enriched material (originating within the bubble) could propagate out into the IGM, readily reaching distances of order the interhalo spacing if they maintained their identity and did not mix too rapidly. The effectiveness of mixing  would depend on the extent to which conductivity is inhibited by magnetic fields. It is worth noting that the bubble material (consisting substantially of supernova remnants) may well contain a magnetic field. If so, the dispersal of the first heavy elements and of the first (`seed') magnetic fields are intimately connected.  If the `fingers' are magnetised, then they may well propagate a long way before mixing with their surroundings.  Whether the intergalactic heavy elements are fully mixed is, however, an interesting  question that is still open. The \\lya forest observations imply that most lines of sight through each `cloud' or filament encounter some heavy elements. But it is not clear that the mixing is uniform. It is possible that, even at redshifts of 3--5, the IGM is inhomogeneous on very small scales. Even in the regions that have been enriched, the metals could be  restricted  to magnetised `streaks' with a volume filling factor of as little as one percent. These would  corresponding to the `fingers' which we envisage having formed at $z=10$, which would have been sheared and distorted by the subsequent gravitational clustering that led to the fully--developed forest at $z=3-5$.   What is the spatial extent of the ensemble of wind--driven ejecta? In the adopted flat cosmology ($\\Omega_M=1$, $h=0.5$, $\\sigma_{8}=0.63$, $n=1$, $\\Omega_bh^2=0.019$) and according to the Press--Schechter formalism, the comoving abundance of collapsed dark halos with masses $M\\approx 10^8\\,h^{-1}\\, \\msun$ at $z=9$ is about $80\\,h^3\\,$Mpc$^{-3}$, corresponding to a mean proper distance between neighboring halos of $15\\,h^{-1}\\,$kpc and to a total mass density parameter of order 3 percent. In Figure \\ref{fig15} we plot the expected overall filling factor (or porosity) of metal--enriched material for different scenarios, assuming a `synchronized' (over timescales shorter than the then Hubble time) population of starbursting halos. As already mentioned, with a star formation efficiency of $f_\\star=10\\%$ only a small fraction, about 4\\% percent, of the stars seen today would form at these early epochs. Still, the impact of such an era of pregalactic outflows could be quite significant, as the product of stellar nucleosynthesis would be distributed over distance that are comparable to the mean proper distance between neighboring low--mass systems, i.e. filling factors $\\gta 20\\%$. Larger filling factors may be obtained for larger efficiencies, moderately top-heavy IMFs, halos where a significant fraction of the gas is in a galactic disk and does not couple to the outflow (as matter is ejected perpendicularly to the disk), or from a population of more numerous sources -- which would therefore have to originate from lower amplitude peaks -- but are not required by current observations. Ferrara, Pettini, \\& Shchekinov (2000) have recently considered the process of metal enrichment of the IGM, so it is useful to compare our main results with theirs. They determined the final radius of the bubbles by assuming that once blow--away has occurred, the shell enters the snowplough phase and it is finally confined by the IGM pressure. Essentially, their stalling radius is much smaller than derived in this work because they assumed a low cooling/star formation efficiency factor, $f_bf_\\star=0.0025$ (corresponding for a halo of total mass $2\\times 10^8\\,\\msun$ to a star formation rate of $10^{-3}\\,\\msun\\,$yr$^{-1}$ over 30 Myr), \\ie they studied outflows in a `quiescent' star formation mode that is perhaps more suitable to larger systems, rather than to the starbursting subgalactic halos we have focused on in this work.  It is interesting to note that, following an early era of SN--driven outflows, the enriched IGM may be heated on large scales to characteristic temperatures $T_\\igm\\gta 10^{4.6}\\,$K if the star formation efficiency is $\\gta 10\\%$, so as to `choke off' the collapse of further $M\\lta 3\\times 10^8\\,h^{-1}\\msun$ systems by raising the Jeans mass. In this sense the process may be self-regulating. While more detailed calculations -- which include the contribution to the IGM metal and heat content from all levels of the mass hierarchy as a function of cosmic time -- need to be done to fully assess the impact of subgalactic halos on the thermal (and ionization) state of the IGM, we argue here that it is unlikely that an early input of mechanical energy will be the dominant effect in determining the ionization state of the IGM on large scales (cf. Tegmark \\etal 1993). We have shown in \\S\\,3.5 that if the fraction of ionizing photons that escape the dense sites of star formation into the intergalactic space is greater than a few percent, pregalactic outflows will propagate into an IGM which has been pre--photoionized by the same massive stars which later explode as SNe. The relative importance of photoionization versus shock ionization obviously depends on the efficiency with which radiation and mechanical energy actually escape into the intergalactic space. One can easily show, however, that, during the evolution of a  `typical' stellar population more energy is actually lost in ultraviolet radiation than in mechanical form (e.g. Madau 2000). This is because in nuclear burning from zero to solar metallicity ($Z_\\odot=0.02$), the energy radiated per baryon is $0.02\\times 0.007\\times m_\\nH c^2$; about one third of it goes into H--ionizing photons. While the same massive stars that dominate the UV light also explode as SNe, the mass fraction radiated in photons above 1 Ryd is $4\\times 10^{-5}$, ten times higher than the fraction released in mechanical energy. Of course, a pre--photoionized universe at $10^4\\,$K, heated up to a higher temperature by SN--driven winds, will recombine more slowly. The expected thermal history of expanding intergalactic gas at the mean density and with low metallicity is plotted in Figure \\ref{fig16} as a function of redshift for a number of illustrative cases. The code we use includes the relevant cooling and heating processes and follows the non--equilibrium evolution of hydrogen and helium ionic species in a cosmological context. The gas is allowed to interact with the CMB through Compton cooling and either with a time--dependent QSO ionizing background as computed by Haardt \\& Madau (1996) or with a time--independent metagalactic flux of intensity $10^{-22}\\,\\uvunits$ at 1 Ryd (and power--law spectrum with energy slope $\\alpha=1$). The temperature of the medium at $z=9$ -- where we start our integration -- has been either computed self--consistently from photoheating or fixed to be in the range $10^{4.6}-10^5\\,$K expected from SN--driven bubbles with significant filling factors. The various curves show that the temperature of the IGM at $z=3-4$ will retain little memory of an early era of pregalactic outflows."
    },
    "0010/astro-ph0010534_arXiv.txt": {
        "abstract": "In order to study early type galaxies in their simplest environments, we have constructed a well-defined sample of 30 isolated galaxies. The sample contains all RC3 early-type galaxies with no other cataloged galaxy with known redshift lying within a projected radius of 1h$_{\\rm 100}^{-1}$ Mpc and \\p 1000 \\kms (where we use the recessional velocities in the RC3).  We have obtained optical and near-infrared images of 23 of the galaxies and of a comparison sample of 13 early-type galaxies in X-ray detected poor groups of galaxies. We have applied the techniques of unsharp masking, galaxy model division, and color maps to search for morphological features that might provide clues to the evolution of these galaxies.  Evidence for dust features is found in approximately 75\\% of both the isolated and group galaxies (17 of 22 and 9 of 12, respectively). However, shells or tidal features are much more prevalent in our isolated sample than in our group sample (9 of 22 $= 41\\%$ versus 1 of 12 $= 8\\%$, respectively). The isolation and colors of these shell galaxies make it unlikely that tidal interactions or asymmetric star formation are the causes of such features. One model that is not ruled out is that mergers produce the shells. If shells and dust are both merger signatures, the absence of shells in group ellipticals implies that shells: 1) form more easily, 2) are younger, and/or 3) are longer-lived in isolated environments. ",
        "introduction": "In the coarsest description of the Hubble sequence, early type galaxies (ellipticals, S0's) are characterized by featureless light profiles, whereas the disks of spiral galaxies have distinct structures like dust lanes and arms.  Nevertheless, there is evidence for structure in some early-types.  Shells (or ripples) are detected in at least 10\\% of the ellipticals in the sample of Malin \\& Carter (1983).  Large dust lanes or patches are observed in many early-type galaxies (Ebneter et. al. 1988) and are suggested in others by IRAS detections at 60 and 100 \\m m (Thronson \\& Bally 1987; Marston 1988; Thronson et al. 1989; Knapp et al.  1989; Bregman et al. 1998). One problem in determining the incidence and causes of such features is accounting first for the role of environment.  Most early types lie in clusters and rich groups of galaxies --- hot, dense regions that may affect both the formation and evolution of features.  It is therefore essential to identify a control sample in truly isolated environments, where the local galaxy density is lower than in even poor groups of galaxies.  We have compiled a sample of 30 isolated early type galaxies in the nearby universe.  If we define isolation as having no massive companion within $1h^{-1}$ Mpc, then no close passage with another galaxy at a relative velocity of \\ab 300 \\kms (approximately the mean pair-wise velocity of galaxies; Marzke et al. 1995) could have occurred in the last $\\sim 3h^{-1}$ Gyr (unless it resulted in a merger).  To detect features such as dust and shells, it is necessary to apply a range of image processing techniques to optical and near-infrared images, identifying which methods work best for each type of structure.  In particular, we would like to know 1) the fraction of dusty {\\it isolated} early types, 2) the fraction of {\\it isolated} early types with shells, and 3) how the fractions of galaxies with dust and with shells/tidal tails vary with galaxy environment.  The fraction of isolated early types that are dusty and/or disturbed offer clues to the evolution of these galaxies.  For instance, Malin \\& Carter (1983) find that shells or ripples occur roughly five times more frequently in environments outside of rich clusters. Ebneter et al. (1988) argue a possible trend for early-type galaxies outside of clusters to be dustier than their cluster counterparts. These results suggest that, outside of clusters, such features are either easier to form, younger, or longer-lived. However, both the Malin \\& Carter (1983) and Ebneter et al. (1988) samples contained few isolated galaxies. A large study of truly isolated galaxies is required to properly address the issue of environmental differences in the structure of early types.  The frequency of dusty and morphologically disturbed isolated galaxies can also illuminate the cause of structure in early types. The origins of such features are currently unknown.  Shells, for example, could result from a merger of galaxies (Quinn 1984; Hernquist \\& Spergel 1992), weak tidal interactions between galaxies (Thomson \\& Wright 1990; Thomson 1991), or asymmetric star formation (Lowenstein, Fabian, \\& Nulsen 1987). If shells are found in isolated galaxies, the tidal interactions model could likely be ruled out. If the colors of isolated galaxies with shells are similar to those without shells, then the asymmetric star formation model is less compelling. With a well-defined sample of isolated early types, it is possible to test the validity of different shell formation models for the first time in a simple environment.  In this paper, we analyze the sample of 30 isolated early type galaxies and a comparison sample of 13 poor group ellipticals.  We define these samples in \\S 2.  We discuss the optical and near-infrared observations and data reduction in \\S 3 and \\S 4.  We then search these images for structure. To assist in the detection of features, we use several different analysis techniques, including unsharp masking, model division and color mapping. We describe this data analysis in section \\S 5. We give our results and discuss their relevance to the possibility of mergers in section \\S 6. Finally, we present our conclusions in \\S 7. All distances and luminosities in this paper are calculated assuming a H$_{\\rm o}$ $=$ 100 km s$^{-1}$ Mpc$^{-1}$ and q$_{\\rm o}$ $=$ 0.5 cosmology. ",
        "conclusions": "We use multi-color images of a sample of isolated, early type galaxies to study the properties of such galaxies outside the hot, dense environments of rich clusters. We compare this sample to a sample of early types in X-ray detected poor groups. Our survey has the important advantage over previous studies that our strict isolation criteria ensures that no recent galaxy or environment interaction is possible except for a complete merger.  The majority of galaxies in our sample show evidence for dust ($\\sim$ 75\\%; 17 of 22 for the isolated sample and 9 of 12 for the group sample). We apply two different techniques to search for dust features, model division and color maps. Both methods suggest that the incidence of dust is comparable in isolated and poor group ellipticals. However, the color maps are more successful at uncovering dust (approximately one-third of the galaxies that show evidence for dust in the color maps show no evidence for dust in the model division images). The dust detection rates are identical for our B-K$_{\\rm S}$ and B-R images. Thus, although dust features tend to be most prominent in optical/near-infrared color maps, the optical images alone appear adequate for identifying dust features.  We have used the technique of unsharp masking to search for shells and tidal features.  We found these features in $\\sim$ 41\\% (9 of 22) of our isolated galaxies and approximately 8\\% (1 of 12) of our group galaxies. The isolation and colors of the shell galaxies suggest that neither tidal interactions nor asymmetric star formation cause such features. A more likely model is that mergers produce the shells. Although the dust fraction is virtually identical for the isolated and group environments, the shell fractions are significantly different. If shells and dust are both merger signatures, the absence of shells in group ellipticals implies that shells: 1) form more easily, 2) are younger, and/or 3) are longer-lived in more isolated environments.  We thank Chuck Keeton for providing useful information on statistics and the anonymous referee for comments that improved this paper. We would also like to thank Tod Lauer for his helpful suggestions. This research involved the use of the NED database. Partial support for this program was provided from NASA grants NAG 5-3529 and HF-01087.01-96A.   \\vfill\\eject  \\appendix"
    },
    "0010/astro-ph0010640.txt": {
        "abstract": "In classifying the ensemble of powerful extragalactic radio sources, considerable evidence has accumulated that radio galaxies and quasars are orientation-dependent manifestations of the same parent population: massive spheroidal galaxies containing correspondingly massive black holes. One of the key factors in establishing this unification has been the signature of a hidden quasar detected in some radio galaxies in polarized light. The obscuration of our direct view of the active nucleus usually, but not necessarily exclusively, by a thick nuclear disk or torus can act conveniently as a `natural coronograph' that allows a much clearer view of the host of a radio galaxy than of a quasar.  In this study, we exploit the opportunity to eliminate the quasar glare by performing sensitive spectropolarimetry with the Keck~II telescope of a sample of radio galaxies with redshifts around 2.5. This represents the epoch when quasars were many times more common that they are now and is likely to be the period during which their host galaxies were being assembled into what become the most massive galaxies in the Universe today. We show that dust-reflected quasar light generally dominates the restframe ultraviolet continuum of these sources and that a highly clumped scattering medium results in almost grey scattering of the active galactic nucleus photons. The observations, however, do not exclude a substantial star formation rate averaged over a Gyr of evolution. The sub-mm reradiation from the scattering dust is likely to represent only a small fraction ($\\sim10\\%$) of the total far infrared luminosity. An analysis of the emission lines excited in the interstellar medium of the host galaxy by the hard quasar radiation field reveals evidence of a dramatic chemical evolution within the spheroid during this epoch. Secondary nitrogen production in intermediate mass stars produces a characteristic signature in the N{\\sc v}/C{\\sc iv} and N{\\sc v}/He{\\sc ii} line ratios which has been seen previously in the broad line region of quasars at similar redshifts. We find intriguing correlations between the strengths of the \\Lya~ and N{\\sc v} emission lines and the degree of ultraviolet continuum polarization which may represent the dispersal of dust associated with the chemical enrichment of the spheroid. ",
        "introduction": "Radio galaxies are thought to be the hosts of radio quasars, oriented such that the direct view to their nuclei is obscured by an optically very thick nuclear structure (Antonucci 1993\\nocite{antonucci93} and references therein). This fortuitous occultation of the active galactic nucleus (AGN), which would otherwise outshine an \\( L_{*} \\) galaxy by a factor of up to a thousand or so, acts as a convenient 'natural coronograph', allowing the detailed study of many properties of the surrounding stars and interstellar medium (ISM). Our current belief is that the hosts of powerful radio sources in the distant Universe are destined to become the giant ellipticals of today: the most massive galactic systems we know (McLure et al. 1999\\nocite{mclure99}). While some may have commenced their formation at very high redshift, it is clear that the process of assembly is very active at \\( z\\sim 2.5 \\) (Pentericci et al. 1999, 2000\\nocite{pentericci99} \\nocite{pentericci2000}; van Breugel et al. 1998\\nocite{breugel98}). This corresponds to the epoch when luminous quasars appear to have had their maximum space density (see eg. Pei 1995\\nocite{pei95}; Shaver et al. 1996\\nocite{shaver96}; Fan et al. 2000\\nocite{fan2000}). Indeed, a major goal of our programme is to understand any causal relationship between the formation of the massive black hole and that of the galaxy within which it resides.  Although we may not see it directly, the presence of the huge luminosity radiated by the quasar makes itself felt in many ways in the surrounding galaxy and understanding those processes is essential if we are to deduce the basic properties and evolutionary state of the host. The radiation field ionizes a substantial fraction of the gaseous interstellar medium (ISM), producing emission line halos extending sometimes hundreds of kpc (van Ojik et al. 1997a\\nocite{ojik97a}) and both destroys and accelerates dust (de Young 1998\\nocite{deyoung98}). The ultraviolet (UV) and optical components of the quasar continuum and broad emission lines can now, thanks to large telescope polarimetry, be seen scattered from dust and/or electrons far from the nucleus (see eg. di Serego Alighieri et al. 1996a\\nocite{alighieri96a}; Dey et al. 1996\\nocite{dey96}; Knopp and Chambers 1997b\\nocite{knopp97b}; Cimatti et al. 1998b\\nocite{cimatti98b}). Both of these reprocessing mechanisms, fluorescence and scattering, contribute to the alignment between the observed optical and radio structures seen in radio galaxies beyond a redshift of 0.7 or so (Chambers et al. 1987\\nocite{chambers87}; McCarthy et al. 1987\\nocite{mccarthy87}). Although at optical wavelengths the effects of scattering are seen most clearly at higher redshifts, where the dilution by starlight below the 4000~\\AA~ break is small, the process can be seen to operate in low redshift radio galaxies when they are observed with large telescopes (Cohen et al. 1999\\nocite{cohen99}).  Above 4000~\\AA~ in the restframe, the evolved stellar population contributes substantially to the observed continuum and produces an infrared (K-band) Hubble diagram with a remarkably small dispersion (McCarthy 1993\\nocite{mccarthy93}, but see references for the higher redshift sources, eg. Eales et al. 1997\\nocite{eales97}; van Breugel et al. 1998\\nocite{breugel98}). Below 4000~\\AA, we might expect to see directly a young stellar population in a starburst arising as a precursor to the AGN or, alternatively, as a result of its influence on the surrounding medium. Objects with a redshift around \\( z\\sim 2.5 \\) give optical spectrographs access to the restframe spectral band which includes the strong UV emission lines from \\Lya~ through C\\textsc{iii}{]} and beyond and also to that part of the continuum which displays detectable spectral features from OB stars as well as absorption lines from the ISM. In addition, the spectrum covers the region where Galactic-type dust displays the 2200~\\AA~ signature and the steep extinction increase below 1600~\\AA.  Our programme of Keck observations of \\( z\\sim 2.5 \\) radio galaxies is designed to obtain high quality spectrophotometry of the spectral features mentioned above together with polarimetry having sufficient signal-to-noise to detect a continuum polarization as low as one or two percent. The polarimetry is used to separate the directly received light from the scattered radiation while the low-noise line and continuum spectroscopy can be used to study many properties of the host galaxy, especially the physical and dynamical state and the chemical composition of the ISM. Data from the first two sources observed, 4C$+$23.56 and 4C$-$00.54, are presented in Cimatti et al. \\cite*{cimatti98b} (hereafter paper I). Here, we report the extension of the sample to nine sources. In this paper, the analysis is restricted to the integrated properties defined by the continuum extent. An analysis of the spatial variations will come in a later publication (Villar-Mart\\'{\\i}n et al. in prep.). Throughout this paper we assume $H_{0}=50$ km~s$^{-1}$~Mpc$^{-1}$ and \\( q_{0}=0.1 \\). ",
        "conclusions": "By performing spectropolarimetry of the restframe ultraviolet emission from a sample of $z\\sim2.5$\\ radio galaxies, we have been able to isolate and identify the dominant contributors to their observed spectra. These are:  \\begin{itemize}  \\item{Spatially-extended, dust-scattered light from the underlying obscured quasar. Due to the supposed clumpiness of the scattering medium, a natural `luminosity-weighting' process operates which means that the scattered photons originate predominantly from regions where $\\tau\\sim 1$. This results in approximately grey reflection of a few percent of the quasar luminosity. Deviations from greyness around 2200~\\AA~ are expected for standard Galactic dust and are seen in our data if the quasars do not have abnormally strong FeII emission. Based on the spectral shape and the fractional polarization, the scattered light accounts for between $\\sim$90\\% and $\\sim$30\\% of the total continuum. In the most polarized sources, we also measure the broad line emission from the quasar with approximately the expected equivalent and velocity widths.}  \\item{A nebular continuum which, based on the observed strength of the HeII $\\lambda 1640$\\ recombination line, is typically $\\sim$10\\% of the total flux in the UV.}  \\item{The remaining continuum is assumed to arise from a young stellar population although we have only very weak direct observational evidence that this is the case. Constraints from the observed continuum colour severely restrict the age and reddening of such a population in the low polarization objects.}  \\item{Strong, spatially-extended narrow emission lines from the ISM of the host galaxy. From photoionization modelling studies it is shown that the dominant fraction of this line emission is excited by the ionizing radiation from the quasar.}  \\end{itemize}  The clumpy dust-scattering model we have employed shows that a few percent of the hidden quasar light is rendered visible in the restframe UV spectrum of the radio galaxies. For luminous AGN, this is generally sufficient to dominate the observed continuum in this wavelength range. Only young, luminous and relatively unreddened starbursts will reveal themselves clearly against the scattered `white haze'. Given the large uncertainty in the reddening of starbursts and the consequent broad range in the allowed conversion of stellar UV flux to SFR, it is likely that we will have to await more sensitive sub-mm measurements in order to investigate the consistency between the building of the stellar content of the host spheroid and the chemical evolution of the ISM as inferred from the emission lines.  The scattering models allow us to calculate directly the FIR radiation emitted from the same dust within the ionization cones which reflects the AGN. This is very much a lower-bound to the expected FIR flux since it takes account neither of the warm dust emission from the obscuring torus nor the cool dust emission from the shadowed dust outside the cones which could be heated by obscured star-formation as well as reflected quasar light. We show that the contribution from the scattering dust is typically 10\\% of the observed upper limits (and one 850$\\mu$m detection for 4C$+$48.48) on sub-mm emission from radio galaxies.  With the exception of \\Lya\\ and N{\\sc v}, which show substantial relative strength variation within our sample, the strong emission line spectra are very similar and indicate that the ISM is ionized predominantly by the quasar radiation field with an ionization parameter $U\\sim 0.04$. The presence of absorption components suggests that the variations in integrated \\Lya\\ emission strength are the result of self-absorption by extended neutral hydrogen halos. The variation by more than a factor of 5 in our sample of the N{\\sc v}/C{\\sc iv} and N{\\sc v}/He{\\sc ii} emission line ratios cannot be readily explained by variations in ionization state or by the effects of carbon depletion onto dust. However, galactic chemical evolution models for elliptical galaxies, where the nitrogen enrichment arises predominantly from secondary processes over a timescale of $\\sim$1~Gyr, offer a good fit to the observations of these lines and are generally consistent with the measurements of other lines. This scenario is similar to that inferred from the BLR measurements of QSO at similar redshifts which imply a range of supersolar metallicities from a few to perhaps twenty times solar. The sequence for the radio galaxy ISM suggests, however, a range of metallicity extending from about half to a few times solar. In contrast to the QSO BLR results, where there appear to be no metal-poor luminous AGN, the much more extended gas in the ISM of radio galaxies shows that massive black holes can exist in relatively metal poor hosts.  The N{\\sc v}/C{\\sc iv} and N{\\sc v}/He{\\sc ii} line ratios are positively correlated with the continuum polarization $P$ while \\Lya/C{\\sc iv} is anticorrelated with $P$. We believe that this is most likely to be the result of an evolutionary sequence within which the spheroids become increasingly dust-enshrouded as the chemical enrichment of the ISM proceeds on a timescate of around 1~Gyr. A property of this scenario is that ULIRG like FSC~10214$+$4724 are the {\\em end point}\\ rather than the beginning of such a period.  Our relatively low resolution spectra are not very sensitive to the presence of weak, narrow absorption lines and so we cannot place very stringent constraints on the presence of lines arising in the photospheres of OB stars. We do, however, see a number of low ionization resonance and excited fine structure lines from carbon, oxygen and silicon. These lines are typically much weaker in absorption than seen in Lyman-break galaxies like MS1512-cB58 (Pettini et al. 2000 \\nocite{pettini2000}) and, indeed, sometimes show redshifted emission components (see Fig.3). There is a weak tendency for high polarization objects to show these lines in emission and low polarization objects to show them in absorption. This could indicate the different locations of the background stellar and scattered continuum sources.  While we demonstrate that scattered quasar light generally dominates the UV continuum of these powerful radio galaxies, the presence of an evolved stellar population would be expected to reveal itself at wavelengths above the 4000~\\AA~ break. To this end, we are currently observing a partially overlapping sample of HzRG with $2.2\\leq z\\leq 2.6$ in the near infrared with the ESO VLT in order to sample their restframe optical spectra. Measurements of the continuum should reveal the presence of any evolved stellar population while the familiar optical forbidden emission lines will improve our ability to perform reliable abundance analyses of the ionized ISM.  \\acknowledgement{ We thank Luc Binette for all of his help and support in the use of the MAPPINGS code, Adolf Witt for discussions about dust and Francesca Matteucci for her advice about galactic chemical enrichment models. The W.M.Keck Observatory is operated as a scientific partnership between the California Institute  of Technology and the University of California; it  was made possible by  the generous financial support of the W.M.Keck Foundation. RAEF is  affiliated to the Astrophysics Division, Space Science Department, European Space Agency. }  %\\documentclass{report} %\\usepackage{portland,lscape} %\\begin{document} %\\vspace{-1cm} \\landscape{ \\begin{table}[p] \\centering{ \\begin{tabular}{llllllllllll}\\\\ \\hline  Line\t\t\t\t&$\\lambda_{vac}$ (\\AA)&4C$+$03.24&0943$-$242\t&0828$+$193\t&4C$+$23.56a\t&0731$+$438\t&4C$-$00.54\t&4C$+$48.48\t&0211$-$122\t&4C$+$40.36\\\\ \\hline O\\sc{vi}\t\t\t\t&1031.9,1037.6\t&\t\t&5.6\t\t&\t\t&\t\t&\t\t&\t\t&\t\t&\t\t&\t\\\\ C{\\sc ii} , C{\\sc ii}$^{\\star}$\t\t&1036.3,1037.0\t&\t\t&5.2\t\t&\t\t&\t\t&\t\t&\t\t&\t\t&\t\t&\t\\\\ He\\sc{ii}\t\t\t\t&1084.9\t\t&\t\t&2.2\t\t&\t\t&\t\t&\t\t&\t\t&\t\t&\t\t&\t\\\\ \\Lya\t\t\t\t&1215.7\t\t&254.4 $\\pm$7.3\t&213.8$\\pm$1.8\t&875.2$\\pm$11.2\t&77.4 $\\pm$1.2\t&428.1$\\pm$2.2\t&261.6$\\pm$3.4\t&700.0$\\pm$11.8\t&9.0$\\pm$0.3\t&1140.5$\\pm$38.5\\\\ N\\sc{v}\t\t\t\t&1238.8,1242.8\t&4.8 $\\pm$0.8\t&11.2 $\\pm$0.6\t&36.7 $\\pm$5.9\t&11.5$\\pm$1.3\t&7.1$\\pm$0.6\t&12.4$\\pm$1.2\t&14.5$\\pm$4.0\t&14.9$\\pm$1.3\t&29.1$\\pm$5.6\\\\ \\Lya + N{\\sc v} broad\t\t&\t\t&\t\t&\t\t&\t\t&22.5$\\pm$5\t&\t\t&\t\t&\t\t&22$\\pm$6\t&\t\\\\ Si{\\sc ii}$^{\\star}$\t\t\t&1264.7,1265.0\t&\t\t&2.0\t\t&\t\t&\t\t&1.0\t\t&1.0\t\t&6.4\t\t&\t\t&\t\\\\ Si\\sc{ii}$^{\\star}$\t\t\t&1309.3\t\t&\t\t&1.8\t\t&3.0\t\t&\t\t&1.9\t\t&\t\t&\t\t&0.9\t\t&9.6\t\\\\ C{\\sc ii} , C\\sc{ii}$^{\\star}$\t\t&1334.5,1335.7\t&4.6 $\\pm$1.3\t&2.4\t\t&3.3\t\t&\t\t&2.2\t\t&1.1\t\t&3.9\t\t&\t\t&20.9\t\\\\ O{\\sc iv}$]$, Si\\sc{iv}\t\t&{\\it 1398}\t&6.1 $\\pm$2.5\t&8.9\t\t&31.4\t\t&1.8\t\t&9.7\t\t&10.4\t\t&13.4\t\t&3.2\t\t&22.0\t\\\\ Si{\\sc iv} broad\t\t\t&1393.8,1402.8\t&\t\t&\t\t&\t\t&8.1\t\t&\t\t&\t\t&\t\t&6.5\t\t&\t\\\\ N\\sc{iv}$]$\t\t\t&1486.5\t\t&\t\t&2.4 $\\pm$0.3\t&7.3 $\\pm$0.8\t&2.2$\\pm$0.6\t&1.8$\\pm$0.3\t&\t\t&4.5$\\pm$0.8\t&1.0$\\pm$0.2\t&2.9$\\pm$0.6\\\\ Si\\sc{ii}\t\t\t\t&1526.7\t\t&\t\t&\t\t&\t\t&\t\t&\t\t&\t\t&\t\t&\t\t&\t\\\\ Si\\sc{ii}$^{\\star}$\t\t\t&1533.4\t\t&\t\t&1.2\t\t&\t\t&\t\t&0.9\t\t&\t\t&\t\t&\t\t&\t\\\\ C\\sc{iv}\t\t\t\t&1542.9,1548.2\t&12.5 $\\pm$2.2\t&46.5 $\\pm$1.1\t&181.7$\\pm$4.5\t&18.0$\\pm$1.5\t&46.5$\\pm$1.0\t&29.1$\\pm$2.0\t&55.5$\\pm$0.5\t&28.2$\\pm$1.0\t&136.1$\\pm$5.1\\\\ C{\\sc iv} broad\t\t\t&\t\t&\t\t&\t\t&\t\t&13.9$\\pm$1.8\t&\t\t&\t\t&\t\t&9.9$\\pm$1.0\t&\t\\\\ $[$ Ne\\sc{v}$]$\t\t\t&1575.1\t\t&\t\t&\t\t&0.9\t\t&\t\t&\t\t&\t\t&\t\t&\t\t&\t\\\\ $[$ Ne\\sc{iv}$]$\t\t\t&1602.0\t\t&\t\t&\t\t&2.2\t\t&\t\t&\t\t&\t\t&\t\t&\t\t&\t\\\\ He\\sc{ii}\t\t\t\t&1640.4\t\t&5.2 $\\pm$1.3\t&47.1$\\pm$1.3\t&103.8$\\pm$1.4\t&15.0$\\pm$1.0\t&30.4$\\pm$0.5\t&21.5$\\pm$0.5\t&53.2$\\pm$3.5\t&15.7$\\pm$0.3\t&77.7$\\pm$5.6\\\\ O\\sc{iii}$]$\t\t\t&1660.8,1666.1\t&\t\t&7.6\t\t&18.4\t\t&4.1\t\t&6.4\t\t&3.6\t\t&6.0\t\t&1.2\t\t&12.4\t\\\\ N\\sc{iii}$]$\t\t\t&{\\it 1749}\t&\t\t&1.4\t\t&6.1\t\t&\t\t&\t\t&\t\t&\t\t&\t\t&8.4\t\\\\ Si\\sc{ii}\t\t\t\t&1808.0\t\t&\t\t&1.7\t\t&5.9\t\t&\t\t&1.4\t\t&\t\t&\t\t&\t\t&\t\\\\ Si\\sc{ii}$^\\star$\t\t\t&1816.9,1817.5\t&\t\t&\t\t&1.4\t\t&\t\t&\t\t&\t\t&\t\t&\t\t&10.1\t\\\\ Si\\sc{iii}$]$\t\t\t&1882.5\t\t&\t\t&2.9\t\t&6.2\t\t&\t\t&1.7\t\t&\t\t&\t\t&\t\t&\t\\\\ Si\\sc{iii}$]$\t\t\t&1892.0\t\t&\t\t&2.5\t\t&12.0\t\t&\t\t&1.0\t\t&\t\t&\t\t&\t\t&\t\\\\ C\\sc{iii}$]$\t\t\t&1906.7,1908.7\t&\t\t&29.3 $\\pm$3.1\t&55.9$\\pm$6.8\t&10.6$\\pm$1.0\t&21.2$\\pm$0.5\t&9.6$\\pm$0.7\t&32.8$\\pm$1.0\t&7.4$\\pm$0.5\t&73.9$\\pm$11.8\\\\ C{\\sc iii}$]$ broad\t\t\t&\t\t&\t\t&\t\t&\t\t&12.2\t\t&\t\t&\t\t&\t\t&7.1\t\t&\t\\\\ C\\sc{ii}$]$\t\t\t&{\\it 2326}\t&\t\t&\t\t&19.2\t\t&8.0\t\t&5.0\t\t&3.6\t\t&11.6\t\t&\t\t&43.3\t\\\\ $[$Ne\\sc{iv}$]$\t\t\t&2421.8\t\t&\t\t&\t\t&49.0\t\t&13.6\t\t&18.0\t\t&22.4\t\t&32.6\t\t&4.8\t\t&37.1\t\\\\ $[$O\\sc{ii}$]$\t\t\t&2471.0\t\t&\t\t&\t\t&\t\t&\t\t&\t\t&\t\t&\t\t&\t\t&20.8\t\\\\ \\hline  \\end{tabular} }  \\caption{\\label{em_lines} Emission line flux measurements. Fluxes and errors are in units of  $10^{-17} \\, $ergs$\\,$s$^{-1}\\,$cm$^{-2}$. Excited fine structure lines are labeled with a $\\star$. Vacuum wavelengths are given in the second column. For blends of more than two lines, the average wavelength is shown in italics. Quoted errors for strong lines were computed as described in section \\ref{section_emlines}. The typical error for weaker lines is about 10\\%} \\end{table}  }  \\endlandscape  %\\end{document}    %\\documentclass{report} %\\usepackage{portland,lscape} %\\begin{document}  \\landscape{  \\begin{table}[p] \\centering{ \\begin{tabular}{llllllllllll}\\\\ \\hline Line\t\t\t\t&$\\lambda_{vac}$ (\\AA)&4C$+$03.24&0943$-$242\t&0828$+$193\t&4C$+$23.56a\t&0731$+$438\t&4C$-$00.54\t&4C$+$48.48\t&0211$-$122\t&4C$+$40.36\\\\ \\hline Si\\sc{ii}\t\t\t\t&1260.4\t\t&\t\t&\t\t&\t\t&3.7\t\t&\t\t&3.0\t\t&\t\t&7.5\t\t&1.7\t\\\\ Si\\sc{ii}$^{\\star}$\t\t\t&1264.7,1265.0\t&\t\t&$-$13.4\t&\t\t&\t\t&$-$4.9\t\t&$-$4.6\t\t&$-$12.5\t&\t\t&\t\\\\ O\\sc{i}, O\\sc{i}$^{\\star}$, Si\\sc{ii}\t&{\\it 1304}\t&\t\t&\t\t&\t\t&4.2\t\t&3.9\t\t&5.5\t\t&2.1\t\t&\t\t&\t\\\\ Si\\sc{ii}$^{\\star}$\t\t\t&1309.3\t\t&\t\t&$-$10.8\t&$-$6.8\t\t&\t\t&$-$8.6\t\t&\t\t&\t\t&$-$5.7\t\t&$-$9.6\t\\\\ C{\\sc ii} , C\\sc{ii}$^{\\star}$\t\t&1334.5,1335.7\t&$-$31\t\t&$-$12.4\t&$-$6.8\t\t&6.7\t\t&$-$9.9\t\t&$-$6.1\t\t&$-$6.9\t\t&2.2\t\t&$-$20.9\\\\ Si\\sc{ii}\t\t\t\t&1526.7\t\t&\t\t&\t\t&5.3\t\t&3.1\t\t&\t\t&\t\t&4.7\t\t&6.8\t\t&10.2\t\\\\ Si\\sc{ii}$^{\\star}$\t\t\t&1533.4\t\t&\t\t&$-$7.8\t\t&\t\t&\t\t&$-$4.2\t\t&\t\t&2.7\t\t&1.0\t\t&\t\\\\ Si\\sc{ii}\t\t\t\t&1808.0\t\t&\t\t&$-$7.9\t\t&$-$18.7\t&\t\t&$-$8.4\t\t&\t\t&\t\t&\t\t&\t\\\\ Si\\sc{ii}$^{\\star}$\t\t\t&1816.9,1817.5\t&\t\t&\t\t&$-$4.5\t\t&\t\t&\t\t&\t\t&\t\t&\t\t&11.6\t\\\\ \\hline  \\end{tabular} }  \\caption{\\label{low_ion_lines}Observed frame equivalent width ($W_{\\lambda}$) measurements of low ionization zero-volt or excited fine-structure (labeled with a $\\star$)  lines in \\AA. $W_{\\lambda}$ is positive for absorption lines and negative for emission lines.Vacuum wavelengths are given in the second column. For blends of more than two lines, the average wavelength is shown in italics.The equivalent width measurement error is typically about 10 \\%} \\end{table}  }  \\endlandscape   %\\end{document}"
    },
    "0010/astro-ph0010228_arXiv.txt": {
        "abstract": "We present a comparison between two observational and three theoretical mass functions for eight cosmological models suggested by the data from the recently completed BOOMERANG-98 and MAXIMA-1 cosmic microwave background (CMB) anisotropy experiments as well as peculiar velocities (PVs) and type Ia supernovae (SN) observations. The cosmological models have been proposed as the best fit models by several groups. We show that no model is in agreement with the abundances of X-ray clusters at $\\sim 10^{14.7} h^{-1}M_{\\odot}$. On the other hand, we find that the BOOM+MAX+{\\sl COBE}:I, Refined Concordance and $\\Lambda$MDM are in a  good agreement with the abundances of optical clusters. The P11 and especially Concordance models predict a slightly lower abundances than observed at $\\sim 10^{14.6} h^{-1}M_{\\odot}$. The BOOM+MAX+{\\sl COBE}:II and PV+CMB+SN models predict a slightly higher abundances than observed at $\\sim 10^{14.9} h^{-1}M_{\\odot}$. The nonflat MAXIMA-1 is in a fatal conflict with the observational cluster abundances and can be safely ruled out. ",
        "introduction": "Recently, certain cosmological models have received a fairly strong observational boost. Several groups have used the new cosmic microwave background (CMB) data from the BOOMERANG-98 (the 1998 Balloon Observations Of Millimetric Extragalatic Radiation ANd Geophysics; \\cite{deb-etal00}) and MAXIMA-1 (The first overnight flight of the Millimeter Anisotropy eXperiment IMaging Array; \\cite{han-etal00}) anisotropy experiments to constrain cosmological parameters. Other groups combined the  constraints from CMB with cosmological nucleosynthesis data, peculiar velocities (PVs) and type Ia supernovae (SN) observations. The values of cosmological parameters vary from one set to the next, but all of these models are in reasonable agreement with a flat Cold Dark Model (CDM) universe ($\\Omega_{0} + \\Omega_{\\Lambda} = 1$) dominated by the vacuum energy except MAXIMA-1 with matter density, $\\Omega_0=0.68$ and vacuum energy density, $\\Omega_{\\Lambda}=0.23$ (Balbi et al. 2000).  In this Letter, we compare the abundances of clusters of galaxies predicted by some popular cosmological models with observed abundances. The abundance of clusters has been shown to be one of the simplest but most effective cosmological tools for constraining the models of structure formation. It can place strong constraints on the parameters  of cosmological models (\\cite{kai86}, Peebles, Daly, \\& Juskiewicz 1989, \\cite{sim-sh89}), including the mass density in the universe ($\\Omega_{0}$) and the amplitude of the mass density fluctuations ($\\sigma_{8}$) or, equivalently, the bias factor ($b=1/\\sigma_{8}$; \\cite{evr89}; \\cite{fre-etal90}; \\cite{hen-arn91}; \\cite{bah-cen92}; \\cite{lil92}; \\cite{ouk-bla92}; Kofman, Gnedin, \\& Bahcall 1993; White, Efsthatiou, \\& Frenk 1993; \\cite{bon-mye96}; Eke, Cole, \\& Frenk 1996; Mo, Jing, \\& White 1996; \\cite{via-lid96}; \\cite{bor-etal97};  \\cite{hen97}; \\cite{pen98}; \\cite{pos99}; \\cite{ver-etal01}; Pierpaoli, Scott, \\& White 2001).  The abundance of clusters and their evolution are quantified by the mass distribution function. The theoretical derivation of the mass function of gravitationally bound objects has been pioneered by Press \\& Schechter (1974, hereafter PS). Despite various modifications that have been suggested recently (Cavaliere, Colafrancesco, \\& Scaramella 1991; Blanchard, Valls-Gabaud, \\& Mamon 1992; Monaco 1997(a,b); Audit, Teysser, \\& Alimi 1997; Lee \\& Shandarin 1998, hereafter LS; Sheth, Mo \\& Tormen 1999, hereafter SMT), it remains a viable model of the mass function and is widely used.  In this Letter, we make use of three theoretical models suggested for the cosmological mass function: (i) the original PS mass function $n_{PS}$ assuming the spherically symmetric collapse, (ii) the mass function $n_{\\lambda_3}$ that incorporates the anisotropic collapse as it is described by the Zel'dovich approximation (LS) and (iii) the mass function $n_{ST}$ suggested by Sheth \\& Tormen (1999, hereafter ST) and later derived by SMT that takes into account both the anisotropic collapse and some nonlocal effects. Recently Jenkins et al. (2001) suggested fits to mass functions obtained in the ``Hubble Volume'' N-body simulations of some cosmological models. We have checked that using the fits by Jenkins et al. (2001) does not change the conclusions of this Letter. For comparison with observations we use mass functions obtained for cluster virial masses by Girardi et al. (1998) and Reiprich, B\\\"{o}hringer, \\& Schuecker (2000)   Here we report the results for eight cosmological models. Among these, seven have recently been claimed as the best-fit models satisfying the data from CMB anisotropy experiments ({\\sl COBE} Differential Microwave Radiometer, BOOMERANG-98 and  MAXIMA-1) as well as from nucleosynthesis, large-scale structure and type Ia SN observations. These models are labeled P11, BOOM+MAX+{\\sl COBE}:I, BOOM+MAX+{\\sl COBE}:II, PV+CMB+SN, Refined Concordance, MAXIMA-1, and $\\Lambda$MDM. We have included the Concordance model as a reference model since it is often referred to as the standard $\\Lambda$CDM model.  None of the proponents of the best-fit models in our list have mentioned the explicit cluster abundance test. Rather some of them claimed that their models satisfy one of the many $\\sigma_8-\\Omega_0$ relations reported in the literature, others even did not apply this test at all. We have noticed more than a dozen predictions of the $\\sigma_8-\\Omega_0$ relation in the literature, some of which are in conflict with the others. We believe our approach here to present the result of the cluster abundance test is more explicit.  This Letter is organized as follows: in $\\S$ 2 we briefly summarize the theoretical models of the cosmological mass functions, in $\\S$ 3 we briefly describe the observational mass functions, in $\\S$ 4 we outline the cosmological models, and, finally, in $\\S$ 5 we report and discuss the results. ",
        "conclusions": "We have compared the theoretical predictions of cluster abundance by several cosmological models with the observational mass functions determined by Girardi et al.(1998) ({\\sl filled circles} in Fig.1,2) and Reiprich et al. (2000) ({\\sl open squares} in Fig.1,2). In this Letter we make use of three theoretical mass functions $n_{PS}$, $n_{\\lambda_3}$ and $n_{ST}$. It is worth stressing that in the range of masses ($4 \\times 10^{14} h^{-1}M_{\\odot} \\le M \\le 3 \\times 10^{15} h^{-1}M_{\\odot}$), the theoretical models differ one from another roughly less or similar to the error bars of both observational mass functions.  At $M \\le 10^{14.8} h^{-1}M_{\\odot}$ no model can be reconciled with the Reiprich et al. (2000) X-ray mass function. On the other hand almost all models are in much better agreement with the Girardi et al. (1998) optical mass function. Thus, the resolution/explanation of the discrepancies between optical and X-ray mass functions becomes crucial for the well being of all models in question.  As far as the optical mass function is concerned the Refined Concordance, BOOM+MAX+{\\sl COBE}:I, and  $\\Lambda$MDM models show a reasonable agreement with observations. The P11 and especially Concordance models predict a slightly lower abundances than observed at $\\sim 10^{14.6} h^{-1}M_{\\odot}$. On the other hand, the BOOM+MAX+{\\sl COBE}:II and PV+CMB+SN models predict a slightly higher abundances than observed at $\\sim 10^{14.9} h^{-1}M_{\\odot}$. The MAXIMA-1 model seems to be safely ruled out by the data on cluster abundances.  A similar comparison using the sharp k-space filter for evaluation of $\\sigma_M$, which is better justified for the PS mass function (\\cite{bon-etal91}), showed that all three theoretical mass function are systematically higher than that for the top-hat filter. The sharp k-space filter approach improves the agreement with observations for the P11 and  Concordance models and makes it worse for the BOOM+MAX+{\\sl COBE}:II and PV+CMB+SN models. The conclusions for other models did not change much."
    },
    "0010/astro-ph0010273_arXiv.txt": {
        "abstract": "{\\sf  In the present work, we study the link between rotation and lithium abundance in giant stars of luminosity class III, on the basis of a large sample of 309 single stars of spectral type F, G and K. We have found a trend for a link between the discontinuity in rotation at the spectral type G0III and the behavior of lithium abundances around the same spectral type. The present work also shows that giant stars presenting the  highest lithium contents, typically stars earlier than G0III, are those with the highest rotation rates, pointing for a dependence of lithium content on rotation, as observed for other luminosity classes. Giant stars later than G0III present, as a rule, the lowest rotation rates and lithium  contents. A large spread of about five magnitudes in lithium abundance is observed for the  slow rotators. Finally, single giant stars  with masses $ 1.5 < M/M_{\\odot}\\leq 2.5 $ show a clearest trend for a correlation between rotational velocity and lithium abundance. } ",
        "introduction": "One of the best known properties of the late-type evolved stars is that their rotational velocity and lithium content decrease with age. Nevertheless, the root cause of this property, as well as the relationship between rotation and lithium abundance, are not yet completely established. Which physical processes control the behavior of rotation and lithium once stars evolve along the giant branch? In particular, how does rotation affect lithium dilution? How does magnetic braking affect rotation and lithium content in evolved stars? Solid answers to these questions have been hampered by the difficulties in the measurement of rotation rates for cool stars as well as by the paucity of lithium abundance measurements for large and complete sample of evolved stars. However, over the last 10 years it has become possible to measure rotational velocities of evolved stars with a precision better than 1.0 km\\,s$^{-1}$ (e.g.: Gray 1989; De Medeiros \\&  Mayor 1989, 1999). In addition, large observational surveys of the lithium line  6\\,707.81 \\AA, with high precision, were carried out (e.g.: Brown {\\it et al.} 1989; Wallerstein {\\it et al.} 1994; Balachandran 1990; Randich {\\it et al.} 1999; L\\'ebre {\\it et al.} 1998). As a result, some very interesting new features on the behavior of rotation and lithium content in evolved stars are emerging.   A dramatic drop in the rotation of subgiant and giant stars occurs, respectively, near the spectral types F8IV and G0III (Gray \\& Nagar 1985; Gray 1989; De Medeiros \\& Mayor 1989, 1991). For the more luminous classes, namely bright giant and Ib supergiant stars, De Medeiros \\& Mayor (1989, 1991) observed a sudden decrease in their rotation near the spectral types F9II and F9Ib, respectively. On the other hand, the giant stars, to the left of the rotational discontinuity, namely in the F spectral region, present a wide range of rotational velocity values, from a few km\\,s$^{-1}$ to about  one hundred times the solar rotation rate. To the right of this discontinuity, namely along the G and K spectral regions, the large majority of stars show low rotation. Along this spectral region only binary systems presenting orbital periods shorter than about 250 days and circular or nearly circular orbits and a dozen of apparently single  giant stars present enhanced rotation. Should one expect some kind of link between the rotational discontinuity and the distribution of lithium abundances? A gradual decrease of surface lithium abundance is expected once stars evolve along the giant branch. At the end of the main-sequence, theory predicts that lithium is confined to the outermost regions of the star in a thin convective layer. Once the star evolves up the giant branch, the convective envelope expands towards the stellar interior and the convective mixture of the outer material rich in lithium with deeper and Li-free material leads to the depletion of this fragile element (e.g.: Iben 1967a,b). Lithium is destroyed when the convective envelope with Li-rich material reaches stellar inner regions with temperatures  higher than about $2.5\\times 10^{6}$ K. Following the initial study by Bonsack (1959), different studies have attempted to analyse  the observational behavior of the lithium abundance along the giant branch. The abundance of lithium in F and early-G giants has been investigated by Wallerstein (1966), Alschuler (1975) and Wallerstein {\\it et al.} (1994). These works have shown a steady decline in lithium abundances from F5III to F8III, a wide spread around the spectral type G0III and low values up to G5III. Wallerstein {\\it et al.} (1994) have analysed the behavior of lithium abundances of F2 -- G5 giants looking for the main features of lithium in the Hertzsprung gap. These authors have found that rapidly rotating giants, namely stars with $v\\sin i>50$ km\\,s$^{-1}$, located in the color interval from  $(B-V)$ = 0.40 to $(B-V) = 0.70$ present lithium abundances close to the presumed primordial value $\\log n(Li) = 3.0$. For  slow rotators, $v\\sin i<50$ km\\,s$^{-1}$, a drop in lithium abundance appears in the color interval  $(B-V) = 0.45$ to  $(B-V) = 0.60$, i.e. stars with  $(B-V)>0.45$ show reduced lithium abundances in spite of their early spectral types. Luck (1977), Lambert {\\it et al.} (1988) and Luck and Lambert (1982) have determined lithium abundances for normal G, K and M-type field giants. From these works, strong evidences have emerged that the content of lithium in this spectral region is primarily controlled by the stellar mass. On the basis of a very large survey of about 644 G and K-type giants Brown {\\it et al.} (1989) have found that a small percentage of late-type giant stars have lithium content far in excess of the standard predictions, a few of them approaching the primordial value. The lithium content distribution of the remaining stars in such survey shows that giants only very rarely are in agreement with standard first dredge-up predictions, in the sense that their lithium content falls below the theoretical predictions. De Medeiros {\\it et al.} (1996a) have shown that, except for a few  chromospherically active giants, Li-rich evolved stars show normal rotational velocity with respect to the typical lithium-normal evolved stars of the same spectral type. Recently, a study of lithium abundance for population I subgiants carried out by Lebre at al. (1999) have shown a sudden decrease in the lithium abundance around Teff =  5\\,600 K, corresponding to a spectral type near G2IV. Such a feature occurs slightly later than the rotational discontinuity at F8IV. Furthermore, some dependence of lithium abundance on rotation, in the sense that fastest rotators show enhanced lithium content, was observed in different classes of luminosity. This dependence, for example, was found for population I subgiant (Randich {\\it et al.} 1999, De Medeiros {\\it et al.}  1997, do Nascimento {\\it et al.} 2000) and for unevolved late-type stars in young clusters (Garcia Lopez {\\it et al.} 1994, Randich {\\it et al.} 1998).  In spite of the important results obtained from the works discussed above, up to date there have been no studies on the link between rotation and lithium content along the giant branch. In the present work we investigate specifically the relationship between rotational velocity and lithium abundance for giant stars on the basis of a large and homogeneous data sample for which we have now precise rotational velocities obtained with the CORAVEL spectrometer. ",
        "conclusions": "The distribution of lithium abundance for single giant stars shows a trend for a discontinuity near $(B-V)$ = 0.70, correponding to the spectral type G0III. Such a discontinuity follows the one observed in rotational velocity. In addition, this work shows a clear dependence of lithium abundance on rotation, in the sense that the highest rotators are also the stars presenting the highest lithium content. One observes that giant stars, presenting the highest lithium content and highest rotation are typically stars earlier than the spectral type G0III. Stars located to the right of the drop in rotation, namely stars later than G0III, present as a rule, the lowest rotation rate and lithium content. In fact, these features show, for giant stars, the same dependence of lithium on rotation observed in other luminosity classes. Giant stars with masses $ 1.5 < M/M_{\\odot}~\\leq  2.5 $  show the more solid trend for a correlation between rotational velocity and lithium abundance. For other mass intervals the least-square solutions show a trend for poor correlation, indicating that the dependence of lithium content on rotation is mass dependent. An additional important feature is the large spread in lithium abundance of the slow rotators. Stars with a $v\\sin i$ lower than about 4.0 km\\,s$^{-1}$ show a wide range of lithium abundance values with $\\log n(Li)$~ ranging from about $-1.5$ to the cosmic value, namely five orders of magnitude. Neverthless, because the sample of giants analyzed in the present work is somewhat limited, in particular for late F and early G type stars, additional measurements of lithium abundance are clearly required for stars located in this spectral region, for a more solid analysis of this apparent discontinuity in lithium."
    },
    "0010/astro-ph0010045_arXiv.txt": {
        "abstract": "Sightlines through high- and intermediate-velocity clouds allow measurements of ionic gas phase abundances, $A$, at very low values of \\HI\\ column density, N(\\HI). Present observations cover over 4 orders of magnitude in N(\\HI). Remarkably, for several ions we find that the $A$ vs N(\\HI) relation is the same at high and low column density and that the abundances have a relatively low dispersion (factors of 2--3) at any particular N(\\HI). Halo gas tends to have slightly higher values of $A$ than disk gas at the same N(\\HI), suggesting that part of the dispersion may be attributed to the environment. We note that the dispersion is largest for \\NaI; using \\NaI\\ as a predictor of N(\\HI) can lead to large errors. \\par Important implications of the low dispersions regarding the physical nature of the ISM are: ($a$) because of clumping, over sufficiently long pathlengths N(\\HI) is a reasonable measure of the {\\it local} density of {\\it most} of the H atoms along the sight line; ($b$) the destruction of grains does not mainly take place in catastrophic events such as strong shocks, but is a continuous function of the mean density; ($c$) the cycling of the ions becoming attached to grains and being detached must be rapid, and the two rates must be roughly equal under a wide variety of conditions; ($d$) in gas that has a low average density the attachment should occur within denser concentrations. ",
        "introduction": "The study of elemental abundances in the interstellar medium has a long history, both in the optical and with space-based UV spectrographs, such as {\\it Copernicus} and the Hubble Space Telescope ({\\it HST}). These studies show that most elements have apparent abundances relative to hydrogen substantially below those in the Sun. This is interpreted as due to depletion of ions onto dust grains. \\par Savage \\& Sembach (1996a) summarized high-quality abundance studies based on {\\it HST} data. They concluded that the depletion of an element depends on the nature of the interstellar medium along the sightline. In cold, dense disk gas, S has near-solar abundance, while Si, Mg, Mn, Cr, Fe and Ni are progressively more depleted. In warm-disk gas, the gas-phase abundances appear higher than in the cool gas, while in halo gas they are even larger. \\par Jenkins, Savage \\& Spitzer (1986) used {\\it Copernicus} to measure accurate column densities for \\MgII, \\PII, \\ClI, \\ClII, \\MnII\\ and \\FeII, for stars with distances, $d$, between 0.1 and 2\\,kpc. They measured N(\\HI) from the damping wings of Ly$\\alpha$ and found a relation between the ionic abundance relative to \\HI, $A$, and the average density, \\navg\\ ($\\equiv$ N(\\HI)/$d$). An explanation was provided by Spitzer (1985), who suggested that the observed depletion in a line of sight depends on the fraction of warm and cold (dense) gas encountered, where each component has a specific gas phase abundance of each ion. \\par Crinklaw, Federman, \\& Joseph\\ (1994) observed \\CaII\\ and \\TiII\\ toward 12 stars with distances between 150 and 650\\,pc. The slope of the relation between log[\\navg] and log[$A$(ion)] is similar for \\CaII\\ and \\TiII, but it differs from those for \\FeII, \\MnII, \\MgII\\ and \\PII. They proposed that \\CaII\\ and \\TiII\\ mostly occur in the warm intercloud medium, and that the apparent depletion is very sensitive to the inclusion of high-density regions (with high depletion) in the sightline. \\par These studies were limited to sightlines with N(\\HI) in the range \\dex{20} to 7\\tdex{21}\\,\\cmm2. To extend this range, we used the high- and intermediate-velocity gas (HVCs and IVCs, see Kuntz \\& Danly 1996, Wakker \\& van Woerden 1997, Wakker 2000). In 21-cm emission these stand out in velocity from gas in the Milky Way disk, so that weak components (down to \\dex{18}\\,\\cmm2) can be measured. At low velocities, parcels of gas with such low column densities can not be separated from other high column density concentrations in the line of sight. \\par Wakker (2000) analyzed all the published absorption-line data pertaining to the high- and intermediate-velocity gas. Improved \\HI\\ column densities (and thus improved ion abundances) were determined for about 250 sightlines. \\CaII\\ was measured for 97 components, \\TiII\\ for 6, \\FeII\\ for 13, \\MnII\\ for 5, \\MgII\\ for 10, and \\NaI\\ for 34 (low-velocity \\NaI\\ was taken from Ferlet et al.\\ 1985). Most IVCs have intrinsically near-solar abundance (Wakker 2000). However, HVC complex~C has $A$$\\sim$0.1 solar (Wakker et al.\\ 1999), and the Magellanic Stream has $A$$\\sim$0.25 solar (Gibson et al.\\ 2000). Including these HVC complexes could lower some of the apparent abundances (in 8 cases for \\CaII, 4 for \\FeII\\ and 10 for \\MgII), but no clear effect is visible. [An anomalously low value for \\MgII, in the VHVC probed by Mrk\\,205, was excluded as this cloud may have low intrinsic abundance, and N(\\HI) is uncertain.] \\par No distances are known for many of the HVCs/IVCs, and they fill only a small fraction of the pathlength. Thus, we looked at the relation between $A$ and N(\\HI), rather than \\navg. Unexpectedly, we found tight anticorrelations of N(\\HI) with the abundances of \\MgII, \\CaII, \\TiII, \\MnII\\ and \\FeII. A minimal correlation is expected because both quantities are integrated along lines of sight with varying physical conditions. Variations include ($a$) the ionization of H, ($b$) widely differing histories of grain destruction by interstellar shocks, and ($c$) considerable differences of the interstellar radiation field impinging on the gas. \\CaII\\ and \\NaI\\ are not even the dominant stage of ionization in neutral regions, so the radiation field is of critical importance for their ionization fraction. In \\S2 we show the correlations between column density and abundances, while in \\S3 we discuss some implications. ",
        "conclusions": "The general trend that the gas phase abundances of refractory ions decreases with N(\\HI) is, of course, well known; the ions are removed from the gas phase by sticking to grains. {\\it However, the low dispersions of the depletions at a particular value of N(\\HI), continuing to low values to N(\\HI), are remarkable.} The depletion patterns provide a diagnostic of the density fluctuation within the ISM, since the depletions require encounters between the grains and ions. {\\it The low dispersions of $A$ at a given N(\\HI) suggest additional properties of the ISM:} \\par 1. Large values of N(\\HI) arise from one or a few regions of relatively large local density rather than from a superposition of many regions of lower density. In this way, there is unexpected physical meaning to the integrated column density: a good correlation with the mean {\\it local} density of the typical H atom along the sightline. If large values of N(\\HI) were built up from many regions of low density along the line of sight, the averaged $A$ would correspond to the lower densities. The association of large N(\\HI) with a large local density is strengthened by the fact that the anticorrelation of gas phase abundances with \\navg\\ is better than with N(\\HI) (Savage \\& Bohlin 1979). \\par 2. The low dispersions in log[$A$] suggest that grains are not destroyed in strong shocks, as some models have suggested. If such destruction occurred, a few points with large gas phase abundances would be expected at large N(\\HI). Rather, the detachment of a refractory element from grains seems to be a continuous function of the mean density, and is common enough to produce a low dispersion about the mean. Grain destruction may occur by means of many small shocks with a broad distribution of velocities. \\par 3. Both the rates of ions sticking to grains and becoming detached from them (probably in atomic form) must be rapid enough to achieve rather good statistical equilibrium. These balancing rates lead to a particular value of $A$ for a given density; otherwise, the past history of the \\HI\\ along the line of sight would strongly influence $A$. \\par 4. In view of (3) above, we think that the best possibility is that attachment takes place within regions of far greater than average density distributed along each line of sight. Encounters between grains and ions are very slow if the mean density of the diffuse ISM is assumed (see below). Possibly magnetohydrodynamic waves create non-thermal motions of the ions relative to the grains, since the charge per mass of the grains is vastly less that for the ions. As a result, ions encounter more dust grains and the net attachment rate increases. \\par Under the assumption that relative motions were thermal, Weingartner \\& Draine (1999) calculated average gas phase abundances of a highly depleted species (\\TiII) for large N(\\HI). In their model, dense (\\navg$\\sim$\\dex3\\,\\cmm3) molecular clouds contain half the mass of the ISM, 30\\% is in cool, neutral clouds (\\navg=30\\,\\cmm3) and 20\\% is in warm neutral clouds (\\navg=0.4\\,\\cmm3). The attachment of the Ti onto grains occurs in the molecular and cool clouds. This model shows that the  observed $A$(\\TiII) for large N(\\HI) can only be achieved by circulation into very dense regions. To achieve rapid attachment within the diffuse ISM, such circulation seems necessary, operating at both low and high column densities. An additional condition for rapid attachment of ions is that the gas be cold as well as dense, in order to provide efficient focusing by the Coulomb force onto negatively charged grains."
    },
    "0010/astro-ph0010335_arXiv.txt": {
        "abstract": "We present a quantitative study of the morphology of 41 Extremely Red Objects (EROs). The analysis is based on deep otical and near--infrared images from the Hubble Space Telescope public archive, and performed by fitting to each galaxy image a PSF--convolved bi--dimensional model brightness distribution. Relying both on the visual inspection of the data and on the results of the fitting procedure, we are able to determine the fraction of irregular and/or interacting EROs, and to identify those that more closely resemble local ellipticals. To the former class, whose members are probably high--redshift dusty starburst, belongs about 15\\% of the whole sample, whereas the elliptical--like objects are between 50 and 80\\% of the total. A few galaxies, although characterized by a compact morphology, are best fitted by an exponential distribution, more typical of local spirals. Our data also suggest that irregular EROs are found predominantly in the field, and that -- on average -- they tend to be characterized by the reddest colors. Finally, we plot the rest--frame Kormendy Relation ($\\mu_e$ vs. $R_e$) for a sample of 6 EROs with spectroscopic redshifts ($z \\sim 1.3$), and estimate its evolution with respect to the local relation. ",
        "introduction": "Among the variety of objects discovered so far at high--redshift, a special class is represented by the so called Extremely Red Objects (EROs hereafter), characterized by moderately faint near--IR magnitudes ($K \\sim 18-20$), and extremely red optical--infrared colors (e.g, $R-K > 5$, see for example Elston et al. \\cite{elston}, McCarthy et al. \\cite{macca}, Hu \\& Ridgway \\cite{hu}). The observed colors and luminosities place this class of objects at $z \\geq 1$, an hypothesis confirmed in a few cases by a direct spectroscopic measurement of the redshift (Graham \\& Dey \\cite{grdey}, Spinrad et al. \\cite{spin}, Stanford et al \\cite{stan} -- S97 hereafter, Liu et al. \\cite{liu}). A twofold interpretation of such observational properties is possible: EROs can be either high--redshift starburst galaxies reddened by a large amount of dust, or passively evolving high--$z$ ellipticals characterized by old stellar populations ($\\geq 1$ Gyr).  The importance of assessing the ERO nature and determining their space density is clear: the epoch of formation of massive elliptical galaxies is a crucial test for the standard hierarchical models for structure formation (e.g.: White \\& Rees \\cite{wr}, Kauffmann et al. \\cite{kauff}), which predict such objects to have formed relatively late from the merging of smaller--size objects (presumably disk galaxies). A large density of high redshift evolved ellipticals would imply severe revision to the hierarchical theories. The other relevant question is the global star formation history: calculations based on the observed rest frame UV flux (e.g. Madau et al. \\cite{madau}, Connolly et al. \\cite{conno}) might be significantly underestimated, if a large fraction of the overall star formation at high redshift takes place in highly obscured starburst galaxies (e.g. Steidel et al. \\cite{steid99}; Barger et al. \\cite{barg00}).  One way to disentangle this ambiguity is provided, in some cases, by near--infrared spectroscopy, in particular if the ERO spectrum exhibits features revealing star--formation activity, such as the redshifted H$_\\alpha$ line; this is the case, for example, for the galaxy HR10 (Graham \\& Dey \\cite{grdey}, Dey et al. \\cite{dey99}). More recently, deep near--infrared spectroscopy allowed to classify two more galaxies as likely starburst -- although their spectra lack of spectral features -- from the amount of reddening required to explain their overall spectral energy distribution  (Cimatti et al. \\cite{cimanew}). A different kind of test is provided by observations in the submm waveband, which traces the thermal emission by dust in the starbursts; this method was successful in the case of HR10 (Cimatti et al. \\cite{cimatti}, Dey et al. \\cite{dey99}) whose detection allowed its non--ambiguous classification as a dust reddened starburst, a result furtherly confirmed by the observation of its CO emission (Andreani et al. \\cite{andre}). Other objects, first detected in the submm, have afterwards turned out to be EROs (Smail et al. \\cite{smail}; Gear et al. \\cite{gear}).  When images of sufficient spatial resolution are available, however, the most direct way to distinguish between the two classes is their morphology: elliptical galaxies are compact, regularly--shaped objects, whereas we expect starburst galaxy to look much more irregular (in particular, if the starburst is triggered by a merger, or if a large amount of dust irregularly distributed is present in the galaxy). HR10, imaged by HST, is consistently characterized by a clearly disturbed morphology (see Dey et al. \\cite{dey99}).  For what concerns the total number density of EROs and their link with passively evolving ellipticals, the works by Cowie et al. \\cite{cowie1} and Hu \\& Ridgway \\cite{hu} suggest that at most a fraction of the present day ellipticals ($\\sim$ 10\\%) could have its progenitors among EROs, but at present the question is far from being settled. Thompson et al. \\cite{thomp} and Barger et al. \\cite{barger}, for example, raise this estimate by a factor 4 $\\sim$ 5; Ben\\'{\\i}tez et al. \\cite{benitez} claim that the density of luminous galaxies is comparable with the local value up to $z\\sim 2$; Eisenhardt et al. \\cite{ei2000}, finally, argue that the fraction of red galaxies at $z>1$ might be significantly higher than previously thought, and consistent with a pure luminosity evolution scenario. Finally, Daddi et al. \\cite{daddi} recently showed that EROs are strongly clustered and that such a clustering can explain the origin of the previous discrepant results on the surface density of $z>1$ elliptical candidates as due to strong field-to-field variations. It has also been noted that even a small amount of star formation would drive a high--redshift elliptical galaxies towards bluer colors, so that it would be missed by a sample selection based on photometric properties only (for example, see Schade et al. \\cite{schade}). The problem of identifying high-redshift evolved galaxies, therefore, is not restricted to the ERO population alone; in this perspective, color--based selection criteria appear insufficient. Again, a different diagnostic tool (Franceschini et al. \\cite{france}, Schade et al. \\cite{schade}) is provided by a quantitative analysis of morphological characteristics. A local elliptical galaxy is an evolved system from the point of view of both its stellar population and its internal dynamics; we may presume that, for some objects at high-$z$, a residual small star formation activity (and in general the overall stellar content of the galaxy) could affect the global colors but leave the shape of the brightness distribution more or less unchanged, so that such galaxies could be easily identified on a morphological basis. Of course this kind of approach requires imaging at high angular resolution, such as can only be obtained by space observatories (namely, by the Hubble Space Telescope -- HST hereafter).  As a first effort to investigate the morphology of EROs, we present here a quantitative analysis carried out on deep HST archive images both in the optical red and in the near infrared. Our aim is to identify elliptical galaxies using their morphological characteristics, and establish their fractional abundance with respect to the overall ERO population, assuming that their surface brightness distributions at $z > 1$ are similar to the ones observed in the local universe. In particular, at the resolution provided by HST, a first classification can be performed visually between compact and irregular objects; among the former ones, different distributions can then be distinguished by fitting different models to the data (for example, exponential and de Vaucouleurs profiles, typically associated to disk galaxies and ellipticals respectively). To this purpose, we have implemented a code for the analysis of the surface brightness distributions of such objects as observed by HST, tested its accuracy on a large number of simulated galaxies, and applied it to a sample of 41 EROs.  This paper is organized as follows: we start describing our sample and the data available for every galaxy, turning afterwards to discuss in detail the techniques developed for the final steps of the data reduction, and for the data analysis; the discussion of the resulting parameters and a morphological classification of the sample are carried out in Sect. \\ref{sect_res} and \\ref{sect_disc}; the conclusions follow in Sect. \\ref{sect_conc}. ",
        "conclusions": "}  We have implemented a code to analyze the surface brightness distribution of faint galaxies, optimized to work on deep  HST images, and we have tested its performances on a large set of simulated galaxies. We have then used this code to fit model brightness distributions to a sample of 41 EROs, imaged by HST in the optical or near--infrared wide--band filters, in order to identify a set of high--$z$ elliptical candidates on a morphological basis. The main results of this work can be summarized as follows.  a) We have determined the fraction of irregular objects among the EROs of our sample, amounting to about 15\\%. These galaxies are the favourite candidates to be the hosts of dusty starbursts.  b) Considering the whole sample, the galaxies characterized by a brightness distribution typical of local ellipticals are at least 50\\%. This estimate is based on a simple distinction between irregular/exponentials and de Vaucouleurs profiles; a more accurate analysis, allowing for different values of the shape index $n$, yields a higher fraction of ellipticals (70-80\\%).  c) We find that a rather small fraction of EROs (15 $\\sim$ 30 \\%) is made up of compact objects whose brightness distributions do not resemble the ones of local elliptical galaxies, but are better described by a pure exponential law.  d) Our data suggest that irregular EROs are found predominantly in the field, and that they are characterized -- on average -- by the reddest colors.  e) We have determined the rest--frame Kormendy relation for a subsample of 6 cluster ellipticals, at redshift $\\sim 1.3$. The relation turns out to be brighter than the local one, at fixed size, by 1.4 mag in the $B$ band.  In the near future, we plan to extend this work to larger, complete samples of high redshift galaxies, in particular devoting our efforts to obtain a reliable picture of the whole population of high--redshift ellipticals."
    },
    "0010/astro-ph0010103_arXiv.txt": {
        "abstract": "We present an analysis of constraints on the companion mass for the Galactic microquasar GRS 1915+105.  Using the known inclination angle and stability of the jet axis, we can rule out massive ($M_s \\ga 3 M_{\\sun}$) companions with orbital axes not aligned with the jet axis.  For an aligned orbital axis, we constrain the ratio of the stellar radius to the binary semi-major axis to be $R_s/a <0.342$, based on the lack of X-ray eclipses.  We then show that these constraints together with the X-ray luminosity, approximate black hole mass, and observations of the X-ray absorbing column density towards GRS 1915+105 rule out accretion via stellar wind from a massive companion, implying that accretion occurs via Roche-lobe overflow.  In that case, the constraint on $R_s/a$ rules out any companion with $M_s > 19.4 M_{\\sun}$ for a black hole mass of $M_{BH} \\simeq 30 M_{\\sun}$.  The lack of significant X-ray reprocessing in the infrared from GRS 1915+105 also provides constraints on the mass and temperature range of a Roche-lobe-filling mass donor. ",
        "introduction": "As the first Galactic source of superluminal jets, GRS 1915+105 has undergone intensive study during recent years.  Despite this level of investigation, one of the system's key parameters -- the approximate mass of the mass-donating star in the binary -- remains a subject of controversy.  No binary modulation in the system, such as eclipses, have been observed at any wavelength, preventing insights based on orbital properties and evolution.  Mirabel et al., (1997) observed near-infrared emission lines of \\ion{He}{1} and Br$\\gamma$, which they interpreted as evidence for a massive Oe- or Be-type companion star.  However, Castro-Tirado et al. (1996) and Eikenberry et al. (1998) reported the observation of a weak, time-variable \\ion{He}{2} emission line (later confirmed by Marti et al., 2000). Since normal Oe and Be stars do not typically doubly-ionize He, they suggested that the emission lines arise in the accretion disk surrounding the compact object, leaving the companion star type unconstrained.  Furthermore, Eikenberry et al. (1998) observed the emission lines changing in strength by factors of several on timescales of $\\sim 5-10$ minutes during jet-producing activity.  They interpreted the constancy of the line profiles during these rapid changes as further evidence of the accretion disk origin of the lines.  More recently, Marti et al. (2000) have presented VLT spectra of GRS 1915+105, in which they confirm the weak \\ion{He}{2} emission line and identify another emission line due to NaI ($2.21 \\mu$m). They interpret this feature as evidence that the mass-donating star in GRS 1915+105 is a super-massive star, such as a luminous blue variable (LBV) star.  However, they note that this interpretation does not seem able to explain the line variability observed by Eikenberry et al. (1998).  In this paper, we will also address the nature of the mass-donating star.  Based on the known inclination angle of the jet axis, its stability over timescales of years, and the lack of X-ray eclipses in GRS 1915+105, we place constraints on the ratio of the mass-donating star's radius $R_s$ to the semi-major axis of the binary system $a$.  We then show, based on estimates of the black hole mass in GRS 1915+105, its X-ray luminosity, and the observed X-ray absorption column towards GRS 1915+105, that the accretion in the system cannot be provided by the wind of a massive star.  We then show that for a $30 M_{\\sun}$ black hole undergoing accretion via Roche lobe overflow, the mass-donating star is constrained to have $M_s < 19.4 M_{\\sun}$.  Next, we use limits on X-ray reprocessing on the stellar surface to place further constraints on the mass-donating star.  Finally, we discuss the implications of these results for GRS 1915+105 and present our conclusions. ",
        "conclusions": "  \\begin{itemize}  \\item Based on the $70 \\arcdeg$ inclination angle and stability of the jet axis, we find that either the mass ratio of the system is $q \\ll 1$, or the orbital axis is aligned with the jet axis.  \\item Based on the lack of X-ray eclipses, for massive companions ($q \\ga 0.1$), the radius of the star must be $R_s < 0.342 a$, where $a$ is the semi-major axis of the orbit.  \\item Given these constraints and the X-ray luminosity of GRS 1915+105, all reasonable parameters assuming accretion via a stellar wind produce an X-ray aborption in the wind which is incompatible with observations.  This indicates that accretion must occur via Roche-lobe overflow in GRS 1915+105.  \\item For Roche-lobe overflow, the constraint on $R_s/a$ requires the companion star to have $M_s < 19.4 M_{\\sun}$ (assuming $M_{BH} = 30 M_{\\sun}$).  \\item Based on limits for IR flux from reprocessed X-ray in GRS 1915+105 (Eikenberry et al., 2000), any companion star with $3 M_{\\sun} < M_s < 25 M_{\\sun}$ must have a surface temperature $T \\ga 1-2 \\times 10^4$ K.  The ``normal'' main sequence and giant stars which fall into this range are early-B stars.  \\item The IR emission lines do not arise in the envelope of a massive companion, but instead are likely to be produced by the outer accretion disk and/or a disk wind being irradiated by the hot inner accretion disk.  \\end{itemize}"
    },
    "0010/astro-ph0010429_arXiv.txt": {
        "abstract": "We have made BeppoSAX observations of the Seyfert 2/1.9 galaxy ESO103-G35, which contains a nuclear maser source and is known to be heavily absorbed in the X-rays.  Analysis of the X-ray spectra observed by SAX  in October 1996 and 1997 yields a spectral index $\\alpha_E$ = 0.74$\\pm$0.07 (F$_{\\nu} \\propto \\nu^{-\\alpha_E}$) which is typical of Seyfert galaxies and consistent with earlier observations of this source. The strong, soft X-ray absorption has a column density, $N_H$ of 1.79$\\pm0.09 \\times 10^{23}$ cm$^{-2}$, again consistent with earlier results. The best fitting spectrum is that of a power law with a high energy cutoff at 29$\\pm$10 keV, a cold (E=6.3$\\pm$0.1 keV, rest frame), marginally resolved ($\\sigma$ = 0.35$\\pm$0.14 keV, FWHM $\\sim 31 \\pm 12 \\times 10^3$ km s$^{-1}$) Fe K$\\alpha$ line with EW 290$^{+100}_{-80}$ eV (1996) and a mildly ionized Fe K-edge at 7.37$^{+0.15}_{-0.21}$ keV, $\\tau = 0.24^{+0.06}_{-0.09}$. The K$\\alpha$ line and cold absorption are consistent with origin in a accretion disk/torus through which our line-of-sight passes at a radial distance of $\\sim 0.01$ pc. The Fe K-edge is mildly ionized suggesting the presence of ionized gas probably in the inner accretion disk, close to the central source or in a separate warm absorber. The data quality is too low to distinguish between these possibilities but the edge-on geometry implied by the water maser emission favors the former. Comparison with earlier observations of ESO103-G35 shows little/no change in spectral parameters while the flux changes by factors of a few on timescales of a few months. The 2--10 keV flux decreased by a factor of $\\sim 2.7$ between Oct 1996 and Oct 1997 with no detectable change in the count rate $>$20 keV (i.e. the PDS data). Spectral fits to the combined datasets indicate either a significant hardening of the spectrum (\\alpe$\\sim$0.5) or a $\\sim$constant or delayed response reflection component. The high energy cutoff (29$\\pm$10 keV) is lower than the typical $\\sim 300$ keV values seen in Seyfert galaxies. A significant subset of similar sources would affect current models of the AGN contribution to the cosmic X-ray background (CXRB) which generally assume a high energy cutoff of $\\sim 300$ keV. ",
        "introduction": "The sub-parsec masing disk recently found to be orbiting a $\\sim 10^7$ solar mass black hole in the Seyfert 2 galaxy, NGC 4258, provides the most compelling evidence to date for the existence of a massive black hole in the nucleus of a galaxy (Miyoshi \\etal\\ 1995). The disk is edge-on, the X-ray spectrum is heavily absorbed (Makishima \\etal\\ 1995) and an active nucleus is visible in polarized light (Wilkes \\etal\\ 1995). Nearly all AGNs which contain maser sources and have X-ray observations show strong X-ray absorption consistent with an edge-on disk, suggesting that they generally harbor active nuclei in their cores. This is consistent with the generally accepted scenario that many Seyfert 2 galaxies are edge-on Seyfert 1 galaxies (Antonucci \\& Miller 1985). The combined power of the high-resolution radio observations and the X-ray spectrum to study the structure of the disk both along the line-of-sight and in the plane of the sky makes these objects potentially key to our understanding of the inner regions of active galaxies.  The Seyfert 1.9/2 galaxy ESO103-G35 is a strong X-ray source, originally discovered by HEAO 1/A2 (H1834-653, Marshall \\etal\\ 1979, Piccinotti \\etal\\ 1982). It was identified as a Seyfert 1.9 galaxy at a redshift of 0.0133 by Phillips \\etal\\ (1979) though later spectra have shown no evidence for broad lines (Morris \\& Ward 1988) leaving its  classification in some doubt. It has been observed extensively in the X-rays. EXOSAT observations revealed strong, soft X-ray absorption (\\nh $\\sim 2 \\times 10^{23}$ cm${-2}$) which showed a factor of 2 variation over 90 days (Warwick, Pounds \\& Turner 1988), providing some of the first evidence for variable absorption in a Seyfert nucleus. {\\it Ginga} observations revealed an equivalent width $\\sim 330$ eV Fe K$\\alpha$ emission line whose origin was not clear given the presumed (but not confirmed) presence of a flaring source within the {\\it Ginga} beam during a portion of this observation (Warwick \\etal\\ 1993). The continuum spectrum was consistent with that reported by EXOSAT.  ASCA observations confirmed and extended the earlier studies, reporting a cold, resolved Fe K$\\alpha$ emission line and a mildly ionized Fe K-edge (Turner \\etal\\ 1997, Forster, Leighly \\& Kay 1999). While flux variations up to a factor of a few are common on short timescales, no significant spectral or absorption variations have been detected.    Recent radio measurements have revealed water maser emission in the form of a single line at $+100$ km s$^{-1}$ (Braatz, Wilson \\& Henkel 1997) with respect to the AGN redshift. Due to the large negative declination ($\\delta \\sim -65^o$) of this source, higher resolution and sensitivity observations to further study this maser are difficult at present. However the combination of water maser emission and the strong, absorbed X-ray source suggest a strong parallel with NGC4258 and motivated our observations with BeppoSAX to obtain an X-ray spectrum simultaneously over a wide energy band (1--200 keV). The moderate spatial resolution ($\\sim 2'$) of BeppoSAX facilitates identification of any nearby, contaminating sources and so avoids the confusion present in the {\\it Ginga} data. Here we present an analysis of the BeppoSAX X-ray data. ",
        "conclusions": "ESO103-G35 is an X-ray bright, Sy 1.9/2.0 galaxy containing one of strongest water maser sources known implying that it is viewed accurately edge-on. The presence of cold molecular material along the line of sight to the active nucleus supports this picture. The amount of X-ray absorption has been $\\sim$constant over the past 15 years and, at $2\\times 10^{23}$ cm$^{-2}$, it is typical of Seyfert 2 galaxies and consistent with a line-of-sight passing through the edge-on accretion disk/torus also responsible for the maser emission.  We observe a marginally resolved Fe-K$\\alpha$ emission line whose width, if dynamic, implies a radial distance to the absorber of about 0.01 pc, similar to that expected for an obscuring torus. Alternatively, the width may be due to a range of ionization states in the emitting material consistent with the ionized Fe K edge at 7.37$^{+0.15}_{-0.21}$ keV. The Fe K$\\alpha$ line strength (EW=290$^{+100}_{-80}$ eV) and width are consistent with origin in an accretion disk/torus through which our line-of-sight is passing. The presence of both cold and warm (ionized) absorbing (and emitting) material suggests that the inner parts of the torus/disk are mildly ionized while the outer parts are cold.  The flux level in the energy range 0.6$-$10 keV decreased by a factor $\\sim 2.7$ between October 1996 and October 1997 with no evidence for a change in spectral shape or absorption. However the 10-50 keV flux did not change between the two observations suggesting either a hardening of the hard continuum or the presence of a reflection component (undetectable at the higher flux level) which reacts to continuum changes with some delay. Given the paucity of datapoints, we are unable to place meaningful constraints on the distance of the reflecting material from the continuum source.  The 1996 dataset shows a high energy cutoff at $29\\pm 10$ keV. This is significantly lower than the values $\\sim 300$ keV believed typical to date (Gondek \\etal\\ 1996, Matt 1998). The high energy spectrum of AGN is critical for modelling the AGN contribution to the CXRB so that this result emphasizes the need to include a realistic distribution of cutoff energy in these models. \\noindent"
    },
    "0010/hep-th0010252_arXiv.txt": {
        "abstract": "{We argue that the total observable entropy is bounded by the inverse of the cosmological constant.  This holds for all space-times with a positive cosmological constant, including cosmologies dominated by ordinary matter, and recollapsing universes.  The argument involves intermediate steps which may be of interest in their own right.  We note that entropy cannot be observed unless it lies both in the past and in the future of the observer's history.  This truncates space-time to a diamond-shaped subset well-suited to the application of the covariant entropy bound.  We further require, and derive, a novel Bekenstein-like bound on matter entropy in asymptotically de~Sitter spaces.  Our main result lends support to the proposal that universes with positive cosmological constant are described by a fundamental theory with only a finite number of degrees of freedom.}    \\preprint{\\hepth{0010252} \\\\ NSF-ITP-00-113}  \\begin{document} ",
        "introduction": "\\label{sec-intro}  \\subsection{Banks's proposal} \\label{sec-banks}  Banks~\\cite{Ban00} has proposed that the cosmological constant should not be viewed as an effective parameter to be derived in a theoretical framework like QFT or string theory.  Instead, it is determined as the inverse of the number of degrees of freedom, $N$, in the fundamental theory.% \\footnote{In this paper, $N$ always denotes the number of degrees of freedom; it should not be confused with the size of a gauge group, or the level of supersymmetry.  The space-time dimension is taken to be 4 in order to keep equations simple, but generalization to arbitrary dimensions is trivial.  Planck units are used throughout.} It should thus be considered an input parameter at the most fundamental level of physics.  The proposal can be motivated as follows.  In the presence of a positive cosmological constant, $\\Lambda$, the universe tends to evolve to empty de~Sitter space.  de~Sitter space has a finite entropy $S = 3\\pi / \\Lambda$, given by the area of the cosmological horizon. Thus the universe is most economically described by a theory with the corresponding number of degrees of freedom, $N = 3\\pi / \\Lambda$. Conversely, a quantum gravity theory with a finite number of degrees of freedom, $N$, requires for consistency a cosmological constant $\\Lambda = 3\\pi/N$ to provide a geometric entropy cutoff.  The $\\Lambda$-$N$ correspondence does not solve the cosmological constant problem except by fiat.  It is not clear why the fundamental theory should happen to possess the enormous but finite number of degrees of freedom $N \\sim 10^{122}$ that corresponds to the observationally favoured value of the cosmological constant.  But the proposal offers a radical, and potentially fruitful, change of perspective.  Its most profound implication is the following: A quantum gravity theory with an infinite number of degrees of freedom, such as M theory, cannot describe space-times with a positive cosmological constant.% \\footnote{For a quantum theory, $N$ is defined to be the logarithm of the dimension of Hilbert space; thus, a theory with finite $N$ has a finite-dimensional Hilbert space.  Superstring theories, and current non-perturbative proposals for M-theory~\\cite{BanFis96,Sus97,Mal97}, have an infinite-dimensional Hilbert space.  (Indeed, even a single harmonic oscillator has an infinite-dimensional Hilbert space.)} This is consistent with the fact that no stable de~Sitter vacua are known in M theory.  If the proposal is correct, this gap would not be due to our limited understanding of the theory, but must be ascribed to an obstruction in principle.  The correspondence thus suggests that one should look for a theory with finite $N$ that is self-consistent and {\\em complete\\/}; i.e., it will not do to impose a naive cut-off on an $N=\\infty$ theory.  If such theories exist for arbitrarily large values of $N$, one might expect them to limit to M theory.  However, finite $N$ theories will contain certain qualitative features, such as positive vacuum energy and perhaps supersymmetry breaking,% \\footnote{Ref.~\\cite{Ban00} also explores the possibility of a connection between finite $N$ and supersymmetry (SUSY) breaking, noting that no stable SUSY-violating vacuum states have been firmly identified in M theory (see, however, Refs.~\\cite{KacKum98,KacSil98a,KacSil98b}).  One may therefore speculate that SUSY breaking can only occur in theories with finite $N$, and that both the SUSY and the vacuum energy scales arise from finite $N$.  One then faces the challenge of explaining why the SUSY scale is much larger than $N^{-1}$.  We shall not pursue the connection with supersymmetry in the present paper.} which would be entirely absent in the infinite $N$ limit and could not have been studied there.  How can the proposal be tested?  It asserts that a universe with $\\Lambda>0$ is a system with $N = 3\\pi/\\Lambda$ degrees of freedom. Unfortunately, this cannot be verified at the semi-classical level, as we have no understanding what the true degrees of freedom are. However, a system with $N$ degrees of freedom certainly cannot have entropy greater than $N$.  Thus, the $\\Lambda$-$N$ correspondence predicts that a universe with $\\Lambda >0$ cannot have entropy greater than $N = 3\\pi/\\Lambda$.  We call this prediction the {\\em $N$-bound}. It can be tested.  It is not difficult to see that the $N$-bound is true for vacuum solutions like de~Sitter space (a trivial case).  Moreover, one can argue that it is satisfied for all space-times which are asymptotically de~Sitter at late times, by the generalized second law of thermodynamics.  Indeed, in Ref.~\\cite{Ban00} the $\\Lambda$-$N$ correspondence was conjectured to apply only within this class of space-times.  This includes, for example, the $\\Lambda > 0$ flat Friedman-Robertson-Walker (FRW) solution which appears to describe our universe: it starts out with a big bang and is initially radiation- or matter-dominated; then the matter is diluted by the cosmological expansion; and at some time (as it happens, roughly now) the vacuum energy---which is not diluted---starts to dominate and leads the universe to evolve towards empty de~Sitter space in the far future.  However, solutions with $\\Lambda > 0$ need not necessarily become de~Sitter at late times.  Consider, for example, the time reversal of the cosmological solution just described: it starts out as empty de~Sitter; then more and more matter condenses, which eventually causes the space-time to collapse in a big crunch.  This illustrates, in particular, that one can never be sure to have reached the safety of asymptotic de~Sitter space; there is always the possibility of a huge collapsing shell of matter that cannot be seen yet but will cause an apocalypse in the future.  Another example is a $\\Lambda > 0$ closed FRW universe.  Given a sufficiently large matter density, the cosmological constant will not be strong enough to prevent recollapse. Indeed, $\\Lambda$ might be a negligible contribution to the total energy density at all times.  These $\\Lambda>0$ solutions are perfectly valid from the perspective of semi-classical gravity.  Many of them are physically quite reasonable, and we would find it unconvincing to exclude them {\\em a priori}.  In some cases, a small perturbation can make all the difference between collapse and expansion to asymptotically de Sitter space.  These arguments lead us to advocate a stronger version of Banks's proposal.  We conjecture that the $\\Lambda$-$N$ correspondence holds for all $\\Lambda>0$ universes, including those that do not evolve to de~Sitter in the future.  But if de~Sitter is not the `final state', the second law will be of no help, and it is no longer obvious that the $N$-bound holds.  The $N$-bound is thus a non-trivial prediction of the $\\Lambda$-$N$ correspondence.  Indeed, at first sight, some solutions may appear to have entropy greater than $N$, in contradiction with the correspondence. Nevertheless, it will be argued in this paper that the $N$-bound is valid for all universes with $\\Lambda > 0$.  This statement is far from obvious, and its proof will be seen to require a number of non-trivial intermediate results.  Therefore, our conclusion may be viewed as evidence in favour of the proposed correspondence.  \\subsection{Outline}  Our goal is to prove the following conjecture:  \\paragraph{$N$-bound} {\\em In any universe with a positive cosmological constant $\\Lambda$ (as well as arbitrary additional matter that may well dominate at all times) the observable entropy $S$ is bounded by} \\begin{equation} N = 3 \\pi / \\Lambda. \\end{equation} Here $S$ includes both matter and horizon entropy, but excludes entropy that cannot be observed in a causal experiment.  Note that $N$ is the Bekenstein-Hawking entropy of empty de~Sitter space.  The bound becomes trivial in the limit of vanishing cosmological constant.  As we have argued above, an independent proof of the $N$-bound provides strong support to the proposed $\\Lambda$-$N$ correspondence~\\cite{Ban00}; hence, the correspondence will not be used anywhere in the paper.  In Sec.~\\ref{sec-cd} we ask what constitutes observable entropy.  We argue that one should restrict attention to the {\\em causal diamond\\/} of an observer: the space-time region that can be both influenced and seen by an observer.  Thus, the observable entropy lies in a region bounded by the past and future light cones from the endpoints of the observer's world line.  The covariant entropy bound~\\cite{Bou99b,Bou99c,Bou99d,FlaMar99}, reviewed in Sec.~\\ref{sec-ceb}, can be applied to the cones bounding the observable region.  This turns out to imply only $S\\leq 2N$, however, which does not quite suffice.  In Sec.~\\ref{sec-dbound}, we derive a novel bound on the entropy of matter systems in de Sitter space, the `D-bound', which can be tighter than the covariant bound. In Sec.~\\ref{sec-nbound} we argue that the two bounds can be combined to imply the $N$-bound.  The results are discussed in Sec.~\\ref{sec-summary}.  Banks's discussion~\\cite{Ban00} of the consistency of his proposal involved many of the considerations that enter our derivation of the $N$-bound.  The arguments presented here are strongly influenced by Susskind's emphasis on the operational meaning of physical quantities. The covariant entropy bound~\\cite{Bou99b}, which plays a central role in the present work, generalizes a proposal by Fischler and Susskind~\\cite{FisSus98} and is thought to have its origin in the holographic principle, first formulated by 't~Hooft~\\cite{Tho93} and Susskind~\\cite{Sus95}.  The application of the covariant entropy bound to the past light-cone of an observer was proposed by Banks in Ref.~\\cite{Ban00a}.  The D-bound is related to Bekenstein's bound on the entropy of finite systems in flat space~\\cite{Bek81}.  Its derivation adapts the original arguments of Geroch and Bekenstein, and extends to cosmologically large systems Schiffer's use of the cosmological horizon to obtain Bekenstein's bound~\\cite{Sch92}. Bekenstein's generalized second law of thermodynamics~\\cite{Bek72,Bek73,Bek74} underlies most of the work in this paper.  The semi-classical description of asymptotically de~Sitter space-times was laid out by the work of Gibbons and Hawking~\\cite{GibHaw77a}; see also Ref.~\\cite{FigHoe75}.  Other recent work exploring connections between the holographic principle and the cosmological constant includes Refs.~\\cite{CohKap98,HorMin00,Tho00}. ",
        "conclusions": "Non-perturbative definitions of quantum gravity have been given for certain space-times that are asymptotically flat or AdS~\\cite{BanFis96,Mal97}.  No such description has been found for space-times with a positive cosmological constant.  As no de~Sitter solutions of M-theory are known, one does not even have a microscopic framework.  Banks~\\cite{Ban00} has opened a new perspective on this problem by suggesting that an asymptotically de~Sitter universe is described by a microscopic theory with finite-dimensional Hilbert space.  Quantitatively, the $\\Lambda$-$N$ correspondence relates the cosmological constant of a stable vacuum, $\\Lambda$, to a theory with $N=3\\pi/\\Lambda$ degrees of freedom (i.e., with a Hilbert space of dimension $e^N$).  If this is correct, M-theory (as it is currently understood) will arise only in the limit of vanishing $\\Lambda$ and infinite $N$.  The cosmological constant problem becomes a problem of understanding the particular dimension of Hilbert space chosen for the theory.  In Sec.~\\ref{sec-banks}, considerations of consistency with semi-classical gravity led us to propose the stronger conjecture that the $\\Lambda$-$N$ correspondence applies to all universes with $\\Lambda>0$, whether they are de~Sitter in the future or not.  This conjecture makes the non-trivial, testable prediction that the observable entropy in all such universes is bounded by $N$.  We then argued that this statement, the `$N$-bound', is correct.  This required the combination of the covariant entropy bound with two intermediate results derived in Secs.~\\ref{sec-cd} and~\\ref{sec-dbound}: the D-bound, and the restriction to causal diamonds.  It is hard to see what, other than the $\\Lambda$-$N$ correspondence, would offer a compelling explanation why such disparate elements appear to join seamlessly to imply a simple and general result.  The D-bound has a number of properties that merit further investigation.  In particular, one can show that it is closely related to Bekenstein's bound~\\cite{Bek81}.  Bekenstein's bound, valid for systems in flat space, can be written as $S_{\\rm m} \\leq \\pi r_{\\rm g} R$, where $r_{\\rm g} = 2m$ is the `gravitational radius' of the system and $R$ is the circumscribing radius.  For dilute, spherically symmetric systems in de~Sitter space, the D-bound takes precisely this form as well, despite the significant deviation from flat space.  A full discussion is given elsewhere~\\cite{Bou00b}.  The restriction to causal diamonds arose in Sec.~\\ref{sec-cd} from the requirement to include only operationally meaningful parts of a space-time in a microscopic description.  This principle is independent of the present context of positive $\\Lambda$, and one may expect that causal diamonds will be of wider use.  Banks~\\cite{Ban00a} has sketched a framework for the combination of quantum mechanics and cosmology, in which the variable size of the quantum Hilbert space is related to the maximal area of the observer's past light-cone.  The arguments of Sec.~\\ref{sec-futurecone} suggest a possible modification of this approach that may lead to a more time-symmetric treatment based on the Hilbert space of causal diamonds.  In de~Sitter space, an observer will be immersed in quantum radiation coming from the cosmological horizon.  At the semi-classical level, this radiation is thermal~\\cite{GibHaw77a}.  One would expect that the radiation will occasionally contain large fluctuations that appear to an observer as classical objects.  Taking a global view of the de~Sitter space, one would say that quantum fluctuations originate behind the future horizon, while classical objects enter through the past horizon.  When one restricts to causal diamonds, however, both of these outside regions are eliminated.  Then it is no longer clear how an observer can distinguish between a genuine classical object and a fluctuation in the quantum radiation.  (This view has previously been advocated by Susskind.)  Indeed, all of standard cosmology may be a rare fluctuation in a long-lived de~Sitter space~\\cite{Ban00}.  Most of high energy physics is based on the S-matrix, with the implicit assumptions that an observer of infinite size is located at the infinity of an asymptotically flat space-time---the observer is `outside looking in'.  This point of view will have to be transcended in order to describe experiments in cosmology, where the observer is always of finite size, and is `inside looking out'.  Indeed, a Minkowski infinity typically does not exist in cosmology; but even if it did, real observers would not live there.  On the other hand, the approximation of an observer as a point in the space-time bulk is also unsatisfactory, because an experiment involves the collection and analysis of information.  According to the holographic principle, a non-vanishing amount of information can be obtained only by an observer of non-zero size.  The maximal information involved in an experiment is related not to the size of the universe, but to the size of the experiment.  One may be motivated by these considerations to abandon the distinction between observer and experiment, and also to claim that a general experiment {\\em is} a causal diamond.  The bottom cone is best thought of as arising from the limitation of preparing the apparatus in a causal way; the future cone reflects the limitation of analysing the data causally.  The correspondence between finite $N$ and positive $\\Lambda$ would impose a surprisingly strong restriction on the fundamental theory, if indeed we live in a universe with positive (and true) vacuum energy. We believe that its implications deserve to be further explored."
    },
    "0010/astro-ph0010386_arXiv.txt": {
        "abstract": "We present {\\it XMM-Newton} observations of the bright quasar PKS~0558$-$504. The 0.2--10 keV spectrum is dominated by a large, variable soft X-ray excess. The fastest flux variations imply accretion onto a Kerr black hole. The {\\it XMM-Newton} data suggest the presence of a `big blue bump' in PKS~0558$-$504 extending from the optical band to $\\sim 3$~keV. The soft X-ray spectrum shows no evidence for significant absorption or emission-line features. The most likely explanation for the hot big blue bump is Comptonization by the multi-temperature corona of a thermal accretion disc running at a high accretion rate. ",
        "introduction": "The most prominent feature in the broad-band spectral energy distribution of many active galactic nuclei (AGN) is the `big blue bump' (BBB), which contains a large fraction of the bolometric luminosity. In most broad-line AGN the BBB begins around 1 $\\mu$m, rises to a peak in the ultraviolet and then declines again into the unobservable EUV (e.g. Elvis et al. \\cite{el}). The soft X-ray excesses seen in many AGN may be the high-energy tail of the BBB (e.g. Turner \\& Pounds \\cite{tutwo}; Walter et al. \\cite{wa}; Laor et al. \\cite{la}). The harder 2--10~keV X-ray spectrum can usually be quite well described by a power-law with photon-index $\\Gamma \\approx 1.9$. Most AGN also show an iron K$\\alpha$ fluorescence feature around 6--7 keV and/or a hump of emission above 10 keV (e.g. Pounds et al. \\cite{pou}), widely interpreted in terms of Compton reflection of hard X-rays by optically-thick matter probably in the form of an accretion disc (George \\& Fabian \\cite{ge}).  Three basic classes of model have been proposed to explain the BBB. The first class involves thermal emission from an optically thick accretion disc surrounding the central black hole in which the primary heating is by viscous dissipation within the disc itself (e.g. Shields \\cite{sh}). The disc thermal spectrum may be comptonized by a hot disc corona; a process which changes the observed shape of the continuum, helping to explain soft X-ray emission from discs which are otherwise too cool (e.g. Czerny \\& Elvis \\cite{cz}). A second class of models also involves thermal emission from an accretion disc, but with irradiation by a hard X-ray power-law (perhaps arising from magnetic flares) as the primary energy source (e.g. Ross \\& Fabian \\cite{ro}). In these models the disc reprocesses the hard X-ray emission into thermal emission which is re-radiated at the local disc temperature. Irradiated discs may have flatter radial temperature profiles, and hence different spectra, compared to discs dominated by viscous dissipation. The final class of model involves optically thin, thermal free-free emission from a hot plasma (Barvainis \\cite{ba}), although single temperature models of this kind do not fit the observed continuum very well (Siemiginowska et al. \\cite{si}).  While the spectral form of the optical and UV continuum in AGN are well constrained by existing data, the soft X-ray continuum shape is often quite poorly defined for individual QSOs and therefore places only limited constraints on BBB models. These limitations have been due to the low spectral resolution, signal-to-noise ratio and/or narrow spectral bandpass provided by previous X-ray instruments, thereby making it difficult to quantify absorption/emission features which may contaminate the X-ray spectrum and hence distort the derived continuum shape. The advent of {\\it XMM-Newton} with its broad bandpass, high throughput and good spectral resolution now provides the opportunity to obtain high-quality X-ray spectra and hence make progress in defining the intrinsic shape of the X-ray continuum in AGN.  In this paper we present {\\it XMM-Newton} observations of PKS~0558$-$504, a low-redshift quasar ({\\it z} $= 0.137$; $M_{\\mathrm{V}} \\approx -24.5$) identified during the {\\it HEAO-1} survey. It has quite narrow Balmer lines (FWHM $= 1500$ km s$^{-1}$, Remillard et al. \\cite{re}), which has led to it being routinely described as a Narrow-Line Seyfert 1 (NLS1), although we note that the UV lines, such as Ly$\\alpha$, are broader (FWHM $\\approx 3800$ km s$^{-1}$; Kuraszkiewicz et al. \\cite{ku}). PKS~0558$-$504 is unusual in being one of the few radio-loud NLS1-type galaxies (Schartel et al. \\cite{sch}). We find that the X-ray spectrum of PKS~0558$-$504 is dominated by a strong soft X-ray excess, which may be the high-energy tail of the BBB. The X-ray spectrum is not contaminated by either cold or warm (ionised) absorption features. Soft X-ray narrow emission-lines are also excluded at a high significance level. Thus, we claim to have detected the featureless X-ray continuum emission from a luminous quasar.  \\begin{figure} \\resizebox{\\hsize}{!}{\\includegraphics{XMM04_f1.eps}} \\caption{A Full-Window mode EPIC MOS-1 image of PKS~0558$-$504 taken during {\\it XMM-Newton} orbit 42. A logarithmic intensity scale has been used to show the point spread function. No other strong X-ray sources are seen within the 30 arcminute field of view.} \\label{mosimage} \\end{figure}  \\section[]{XMM-Newton Observations}  We present observations of PKS~0558$-$504 taken with the European Photon Imaging Camera (EPIC) during orbits 42 and 84 (1--2 March 2000 and 24--25 May 2000) of the calibration and performance verification phase of {\\it XMM-Newton}. The EPIC instrument is described in Turner et al. (\\cite{tuone}) and Strueder et al. (\\cite{str}). During both orbits the EPIC MOS cameras were operated in several different observing modes. Here we include only those observations taken with the medium filter, for which the total EPIC MOS exposure times for orbits 42 and 84 are 85.5 ksecs and 53.7 ksecs respectively. Only the full-frame mode EPIC PN data for orbit 84 were used, for which the total exposure time is 10.2 ksecs.  For operational reasons, the EPIC observations in each orbit were divided into several separate exposures of varying lengths. These exposures were reduced using the {\\it XMM-Newton} {\\sc SAS} (Science Analysis Software) package. The event lists output from the standard {\\sc emchain} and {\\sc epchain} scripts were further filtered for each EPIC exposure using the {\\sc SAS} {\\sc xmmselect} task. For the MOS exposures only X-ray events corresponding to patterns 0--12 (similar to {\\it ASCA} event grades 0--4) were selected. Only pattern 0 (single pixel) events were selected for the PN. Known hot or flickering pixels and electronic noise were rejected using the {\\sc SAS} and the low-energy cutoff was set to 200 eV. One of the EPIC MOS-1 images from orbit 42 is shown in Fig.\\ \\ref{mosimage}.   Source spectra and light curves were extracted from each EPIC exposure using a circular region of diameter 1 arcminute centred on the source position. PKS~0558$-$504 is by far the brightest X-ray source in the 30 arcminute EPIC field-of-view. The EPIC cameras are subject to non X-ray background events due to particles which can vary significantly during an exposure. Background spectra and light curves for use in later data analysis were derived in a similar way to the source data using an offset region close to the source. We note that the observed count-rate for PKS~0558$-$504 implies little effect due to pile-up and no significant spectral differences were found between observations taken in different MOS window modes. The {\\sc Xspec v11.0} software package was used to calibrate the background-subtracted EPIC spectra using the most recent camera response matrices derived from ground-based and in-orbit data. Before further analysis the EPIC spectra were binned to give a minimum of 20 counts per bin.   Reflection Grating Spectrograph (RGS; den Herder et al. \\cite{den}) exposures from orbit 84 were also reduced to search for weak/narrow soft X-ray absorption or emission-line features. Three exposures were analysed, two from RGS-1 (each of 25 ksecs) and one from RGS-2 (31 ksecs). The RGS data were first processed using the {\\sc SAS} {\\sc rgsproc } script. The merged RGS event lists were then fed to the {\\sc SAS} {\\sc evselect} and {\\sc xmmselect} tasks, and both source and background spectra for each exposure were extracted interactively. These spectra were then calibrated using response matrices derived from the individual RGS exposures using the {\\sc SAS} {\\sc rmfgen} task.  The Optical Monitor (OM; Mason et al. \\cite{ma}) imaging data for orbits 42 and 84 were reduced using standard {\\sc SAS} tasks and calibrated with the instrument response information for each filter.  \\begin{figure} \\resizebox{\\hsize}{!}{\\rotatebox{-90}{\\includegraphics{XMM04_f2.eps}}} \\caption{The EPIC MOS-1 + MOS-2 0.2--10 keV light curve of PKS~0558$-$504 for orbit 84 in 125 second time bins.} \\label{lcurve} \\end{figure} ",
        "conclusions": "The {\\it XMM-Newton} X-ray spectrum of PKS~0558$-$504 is dominated by a large, variable soft excess. The fastest observed variability implies a minimum efficiency of converting matter to energy of $\\eta = 0.25$, suggesting accretion onto a Kerr black hole. Combining the X-ray data with simultaneous OM optical and UV data reveal a large big blue bump; too broad to be due purely to a thermal accretion disc spectrum. The X-ray spectrum can be fitted by a Comptonization model in which thermal photons, presumably from the disc, are scattered by a hot corona. At least two electron temperatures are required. Overall, PKS~0558$-$504 can be explained by an accretion disc running at a high accretion rate producing copious amounts of soft EUV photons which are then Compton cooling a hot disc corona."
    },
    "0010/astro-ph0010179_arXiv.txt": {
        "abstract": "Observations and theoretical studies have demonstrated that the pulsed spectra of all gamma-ray pulsars terminate at energies below a few hundred GeV. In most cases we expect this cutoff energy $E_o$ to be around 10 GeV. Only with next-generation ground-based $\\gamma$-ray telescopes, which are expected to have non-zero trigger probabilities near 10 GeV, can we expect to detect pulsations. The large $\\gamma$-ray fluxes below $E_o$, together with the associated sharp pulse profiles, compensate for the lack of imaging capability near threshold. For H.E.S.S. we find that the pulsed component of PSR B1706-44 should be detectable near threshold, whereas the unidentified GeV EGRET sources should be detectable if the superexponential cutoff energy is larger than $\\sim 30$ GeV for relatively hard pulsar photon spectra ($\\sim E^{-1.5}$). ",
        "introduction": " ",
        "conclusions": ""
    },
    "0010/astro-ph0010453_arXiv.txt": {
        "abstract": "Inflationary predictions of the cosmic microwave background anisotropy are often based on the slow-roll approximation. We study the precision of these predictions and compare them with the recent data from BOOMERanG and MAXIMA-1. ",
        "introduction": " ",
        "conclusions": ""
    },
    "0010/astro-ph0010209_arXiv.txt": {
        "abstract": "Recent measurements of the autocorrelation function of the Ly$\\alpha$ clouds are analyzed from the point of view of a simple model with strong clustering on the small scales. It is shown that this toy model reproduces fairly well the important linear relation between amplitude of the absorber autocorrelation function and neutral hydrogen column density. In addition, it predicts a correct evolutionary trend of correlation amplitudes. Some possible ramifications of these results are discussed. ",
        "introduction": "The problem of clustering of the Ly$\\alpha$ forest absorbers has been discussed in the course of the last two decades by many authors (Sargent et al.~1980; Dekel 1982; Salmon \\& Hogan 1986; Webb 1987; Ostriker, Bajtlik \\& Duncan 1988; Crotts 1989; Heisler, Hogan \\& White 1989; Liu \\& Jones 1990; Webb \\& Barcons 1991; Barcons \\& Webb 1990, 1991; Fang 1991; Mo et al.~1992; Fardal \\& Shull 1993; Srianand \\& Khare 1994; Chernomordik 1995; Elowitz, Green \\& Impey 1995; Meiksin \\& Bouchet 1995; Carbone \\& Savaglio 1996; Srianand 1996; Ulmer 1996; Fern\\'andez-Soto et al.~1996; Lanzetta, Webb \\& Barcons 1996; Pando \\& Fang 1996; Cen \\& Simcoe 1997; Rauch 1998). The conclusions of the entire effort are still controversial, since the original paradigm that ${\\rm Ly}\\alpha$ clouds do not show no clustering at all (Sargent et al.~1980; Barcons \\& Webb 1990), was somewhat undermined  by findings of weak clustering of intermediate-redshift ${\\rm Ly}\\alpha$ clouds (Webb 1987; Barcons \\& Webb 1991; Webb \\& Barcons 1991; Chernomordik 1995), and seriously questioned by the work on associated C IV absorption (Cowie et al.\\ 1995; Fern\\'andez-Soto et al.~1996; see also Songaila \\& Cowie 1996). In an interesting work, Crotts (1989) has investigated  correlations in the real space across the sky among systems in multiple lines-of-sight, and not only detected small-scale clustering, but also established the increase of clustering amplitudes with increasing column density, the conclusion which we shall quantify below. Difficulties and limitations inherent in any attempt to use the two-point correlation function to deduce the properties of the ${\\rm Ly}\\alpha$ forest are summarized in Rauch et al.~(1992) and Fern\\'{a}ndez-Soto et al.~(1996). Theoretical analysis of Dekel (1982) was very interesting in this respect, since it weakened the dominant paradigm of Sargent et al.~(1980)---which, as we have seen, dictated much of the future development of absorption-line studies---where it looked strongest: in the cosmological part of the argument. As Dekel writes: \"My aim is to point out a cosmological scenario in which galaxies are clustered only weakly at $z > 1.7$, so that the LACs [Ly$\\alpha$ clouds] may cluster just like galaxies. Here both the isothermal and the truncated adiabatic components of the density perturbations play a role in the formation of structure in the universe.\" It is also quite interesting to note that Dekel (1982) was the first to suggest usage of C IV lines to measure the degree of clustering of  absorption systems, idea which was fully realized only 14 years later by Fern\\'{a}ndez-Soto et al.~(1996). It should certainly be mentioned that low spectral resolution of most of existing measurements (e.g.~$250 - 300$ km s$^{-1}$ for the {\\it HST\\/} Key Project; see Jannuzi 1997; Jannuzi et al.\\ 1998) makes investigations on small scales exceedingly difficult. And it is exactly these scales which are, because of their discriminative power, the most interesting from our point of view.  Webb (1987) was the first to point out the presence of weak clustering at small velocity scales, based on the Voigt profile fitting in the high-redshift data, which was confirmed by other investigations (e.g.~Muecket \\& Mueller 1987). As emphasized by Fern\\'andez-Soto et al.~(1996), it is very difficult to directly detect clustering of the Ly$\\alpha$ lines because of the short redshift path length in any individual QSO spectrum, and line blending. If a significant fraction, or most of the observed Ly$\\alpha$ absorption lines are in fact blends of very narrow components, detected amplitudes of the autocorrelation function would be significantly underestimated (see also a discussion in Rauch et al.~1992).  On the basis of these early findings, Ostriker et al.~(1988) first proposed gravitationally induced clustering as one of the possible explanations for excess of pairs of absorbers at small (and in their sample, intermediate\\footnote{which has not been confirmed afterwards; see the discussion in Chernomordik (1995).}) velocity splittings.  However, in a recent important study, Cristiani et al.~(1997), performed the most comprehensive analysis of the clustering of ${\\rm Ly}\\alpha$ clouds in a sample of about 1600 absorbers along 15 lines of sight, and concluded that the small-scale clustering of ${\\rm Ly}\\alpha$ absorbers (a) is real, and (b) can be understood in terms of gravitationally-induced clustering, in the manner of Ostriker et al.~(1988). Parenthetically, the existence of structure in the Ly$\\alpha$ forest was confirmed by independent methods aimed at detection of the deviation of spatial distribution of absorbers from a uniform random one; thus Fang (1991) showed that Ly$\\alpha$ forest deviate from a uniform distribution at 3$\\sigma$ significant level. This is, of course, still weaker from the non-uniformity seen among the known galactic population, but very different from the picture of uniform, diffuse intergalactic population envisaged by Sargent et al.~(1980).  Another recent work of great importance for the development of our ideas on the spatial distribution of Ly$\\alpha$ clouds is that of Ulmer (1996), who investigated a sample of low-$z$ Ly$\\alpha$ lines recently obtained with the {\\it HST\\/} Key Project (Bahcall et al.\\ 1996). Results of that work are particularly significant, since they demonstrate the existence of strong clustering of Ly$\\alpha$ lines at velocity separations at which high-$z$ lines seem completely unclustered, and which represent an intermediate regime between the small-scale ($\\Delta v \\leq 200$ km s$^{-1}$) and large scale clustering. In physical terms, we can hypothesize that the small-scale regime can be plausibly explained as characteristic of the {\\it intragalactic\\/} motions in typical $L \\sim L_\\ast$ galaxies, being on the same order as velocity dispersion in the known galactic subsystems (Binney \\& Tremaine 1987). On the other hand, large-scale clustering is generally believed to trace large-scale structure (i.e.~structures with velocity dispersions similar to that in rich galaxy clusters and higher). Ulmer (1996) has not obtained any information on the small velocity splittings, since his method explicitly rejects velocity splittings with $\\Delta v < 250$ km s$^{-1}$. Still, its implications are important because of the very strong signal found for $250 \\leq \\Delta v \\leq 500$ km s$^{-1}$, which, if extrapolated into $\\Delta v< 250$ km s$^{-1}$ agrees well with results of Cristiani et al.~(1997) and those discussed in further text. Similar excess of absorber pairs for $\\Delta v \\sim 200$ km s$^{-1}$ was found by Srianand \\& Khare (1994) at the $\\sim 4\\sigma$ level. The redshift evolution of clustering is also correctly emphasized in Ulmer (1996), who inferred a substantial increase in the degree of clustering of the Ly$\\alpha$ forest. As we shall see, there is evidence that the same trend is real over most of the history of the universe.  In the rest of this paper, we shall attempt to show that the results of Ulmer (1996) and Cristiani et al.~(1997) for the amplitude of TPCF are consistent with simple model characterized by constant small-scale clustering. Specifically, we shall show that (i) the linear relation between column density of clouds and their autocorrelation amplitude, and (ii) the general evolutionary trend of decrease in clustering with increasing redshift, are successfully explained in such a toy model. Following two sections are, therefore, devoted to these two important issues. Although not specifically endorsing such a simplistic approach, it does make the complex explanations for the observed TPCF properties, involving biasing for the structure formation and gravitationally induced correlations, unnecessary. In contradistinction, models in which the Ly$\\alpha$ forest is locally decoupled from the Hubble flow and physically associated with collapsed structures (e.g.~galaxies) predict, in general, exactly such a behavior. ",
        "conclusions": "This simple picture is what is generally expected from clouds residing in haloes dominated by dark matter, which are plausible physical candidates for our points around which absorbers are clustered with amplitude $\\phi (v)$. In physical terms, the dependence of absorbing column density on distance from the center of an $L_B$ galaxy (Chen et al.~1998)  \\begin{eqnarray} \\label{chenmain} \\log \\left( \\frac{N_{\\rm H\\ I}}{10^{20} \\; {\\rm cm}^{-2}} \\right)=  - 5.33 \\log \\left( \\frac{\\rho}{10 \\; {\\rm kpc}} \\right) + \\nonumber\\\\ +2.19 \\log \\left( \\frac{L_B}{L_{B_\\ast}} \\right) + 1.09,&& \\end{eqnarray} implies that strong absorption will be seen only near the halo center. It is straightforward to conclude that these strong and rare absorption lines have a relatively large probability of having weaker companion lines originating in the same halo, i.e.~within the galaxy velocity dispersion. Therefore, overall clustering strength is expected to increase with column density. These absorption sites can be classical haloes of luminous galaxies or minihaloes (e.g.~Meiksin 1994; Rauch 1998). Hypothesis that at least a fraction of Ly$\\alpha$ clouds is located in extended haloes of luminous galaxies has found very strong support in low-redshift coincident studies (Lanzetta et al.~1996; Chen et al.~1998). In other words, the results of the present small-scale clustering analysis presented support the general subclass of models with dark matter-dominated gravitational confinement. The theoretical basis of such models is given in important works of Mo (1994) and Mo \\& Miralda-Escud\\'e (1996), where the strongly physically motivated two-phase gaseous halo model has been developed in some detail. It is not clear at present whether some variant of Black's (1981) classic self-gravitating confinement can also be cast in form which will satisfy the TPCF constraints, but it does not seem very promising. Contrariwise, the theories of origin of the Ly$\\alpha$ forest which link the absorption to larger systems, i.e.~clusters or superclusters (Oort 1981; Doroshkevich 1984) are in clear disagreement with these correlation measurements, due to much higher velocity dispersion of such structures. The same applies, as correctly noted by Srianand (1996), to the theories involving explosion-type processes (Ozernoy \\& Chernomordik 1978; Chernomordik \\& Ozernoy 1983;  Vishniac \\& Bust 1987). On the other hand, strong anticorrelation between low-$z$ Ly$\\alpha$ equivalent widths (i.e.~H I column densities) and galaxy impact parameter in absorption-selected galaxy sample of Chen et al.\\ (1998), indicates that these objects share intragalactic velocity dispersions (i.e.~the same velocity scales as discussed here). It is difficult, however, to distinguish between models with extended gaseous haloes and huge disks of Maloney (1992). It should be mentioned that York et al.~(1986) have argued that there are large hydrodynamic velocities observed in absorption line systems which are similar to those seen in lines of sight through galaxies with active star formation. Along the same line of thought one should consider the finding of the {\\it HST\\/} Key Project (e.g.~Boksenberg 1995) showing that the line density of Ly$\\alpha$ absorption systems is greater by nearly an order of magnitude in the vicinity of metal-line systems (which are believed to originate in haloes of normal galaxies). In the framework of our toy model, this could be interpreted as an observational justification for using negligible $\\phi (v)$ outside the range spanned by intragalactic velocities ($0-250$ km s$^{-1}$). In general, the conclusion that the redshift dependence of the correlation function amplitudes can discriminate between various models of Ly$\\alpha$ clouds has important and far-reaching consequences.  Parenthetically, the lack of power in the Ly$\\alpha$ absorption line TPCF amplitudes at larger scales in comparison with the TPCF of local galaxies need not, along the general proposition of Dekel (1982), necessarily be understood as indication for physical difference of absorber and galaxy populations. Instead, one may follow the suggestion of Fang (1991) whose results indicate that a biased clustering in the universe simply has not occurred yet at $z \\sim 2$ on large (comoving) scales. We should keep in mind that absence of large-scale clustering in the conventional Ly$\\alpha$ forest samples has been based mainly on surveys of high-$z$ absorption lines, usually with $\\langle z \\rangle \\geq 2$. Only large Ly$\\alpha$ forest surveys at intermediate and small redshift, likely to be available in the near future, will be capable of definitely solving the puzzle.  We conclude that at this level of accuracy of the TPCF measurements, a simple model with large and constant small-scale clustering is able to account for available observational evidence. Empirically well known galactic velocity dispersion seems to be capable of entirely explaining observed small-scale clustering properties of Ly$\\alpha$ clouds, in agreement with the Occam's razor. Much further theoretical work is certainly necessary in order to elaborate the details of the galactic halo model for Ly$\\alpha$ clouds. However, the fact that this model is naturally arising in a compelling theoretical picture such as the halo cloud model is quite remarkable."
    },
    "0010/astro-ph0010515_arXiv.txt": {
        "abstract": "We examine the star clusters in the luminous irregular galaxy NGC 4449. We use a near-infrared spectrum and broad-band images taken with the {\\it Hubble Space Telescope} to place a limit of 8--15 Myrs on the age of the bright central object in NGC 4449. Its luminosity and size suggest that it is comparable to young super star clusters. However, there is a peculiar nucleated-bar structure at the center of this star cluster, and we suggest that this structure is debris from the interaction that has produced the counter-rotating gas systems and extended gas streamers in the galaxy.  From the images we identify 60 other candidate compact star clusters in NGC 4449. Fourteen of these could be background elliptical galaxies or old globular star clusters. Of the star clusters, three, in addition to the central object, are potentially super star clusters, and many others are comparable to the populous clusters found in the LMC. The star clusters span a large range in ages with no obvious peak in cluster formation that might be attributed to the interaction that the galaxy has experienced. ",
        "introduction": "NGC 4449 is a luminous and actively star-forming barred Magellanic-type irregular galaxy. It has an integrated M$_{B}$ of $-$18 and a star formation rate about twice that of the LMC. It has been well known for decades that NGC 4449 is unusual in having neutral hydrogen extending to six times its Holmberg radius (van Woerden, Bosma, \\& Mebold 1975; Bajaja, Huchtmeier, \\& Klein 1994). In addition the HI maps had revealed counter-rotating gas systems: The gas within a 2\\arcmin\\ radius of the center is rotating in the opposite direction to the gas outside that radius. Recently, Hunter \\et\\ (1998) reobserved the extended HI gas around NGC 4449 at higher resolution and resolved the extended gas into enormous streamers that wrap around the galaxy.  The enormous streamers suggest that NGC 4449 has undergone some type of interaction with another galaxy (Hunter \\et\\ 1998, Theis 1999). A perturber could have passed by, with or without giving up mass to NGC 4449, and set up tidal tails and bridges or a perturber could have passed through the disk. Furthermore, the presence of counter-rotating systems is often interpreted in other galaxies as the signature of acquisition of external material as in a merger (Hernquist \\& Barnes 1991; Quinn \\& Binney 1992; Braun \\et\\ 1994), so a small partner could have been absorbed into NGC 4449. In some scenarios the effects of the encounter could remain for timescales on the order of a Gyr, allowing the perturber to move out of the neighborhood or to have been digested by NGC 4449. NGC 4449 is a member of the CVnI ``loose'' cloud of galaxies and the small irregular DDO 125 is only 41 kpc away in the plane of the sky. So, there exist potential interaction partners.  In this paper we examine the nature of the system of star clusters in NGC 4449 and its clues to the galaxy's evolution. The type of clusters, their ages, and distribution could potentially be affected by interaction with another galaxy. For example, if the clusters were all the same age, this might suggest that some short violent event, such as the interaction of galaxies, caused the formation of the clusters.  Additionally, if the galaxy contains a population of the relatively unusual super star clusters, we might argue that these clusters were formed in the interaction. Super star clusters are similar in compactness and luminosity to globular clusters when scaled to a common age. NGC 1569, for example, an Im galaxy that is undergoing a burst of star formation, has made two super star clusters in the recent starburst (see, for example, O'Connell, Gallagher, \\& Hunter 1994), and many other super star clusters have been found in galaxies that have undergone interactions. In addition to learning about NGC 4449, locating additional super star clusters would be interesting in helping to determine what conditions are necessary for the formation of super star clusters, and by inference globular clusters. Understanding the formation of globular clusters would help us better understand our own galaxy's evolution as well as that of NGC 4449.  When one looks at the visible images of NGC 4449, an object in the center of the galaxy stands out because of both its brightness and its size.  Its brightness suggests that it could be one of these super star clusters.  Alternatively, it has informally been referred to by one of us as the nucleus of NGC 4449, but since Im galaxies do not have nuclei, this would be highly unusual. Here we are using the term ``nucleus'' to mean a centrally concentrated substructure of the galaxy that has been an integral part of the galaxy from the galaxy's formation. One of our goals in this study was to determine the nature of the central object. One important piece of information is its age: a nucleus, by our definition, must be old; a star cluster that just happened to form near the center need not be old.  To answer these questions, we used broad-band images obtained with the Hubble Space Telescope ({\\it HST}) to identify star clusters and determine integrated photometry. We compared the integrated photometry to star cluster evolutionary models to estimate ages. To further set a limit on the age of the central object, we obtained near-infrared spectra of the stellar $^{12}$CO(2,0) and $^{12}$CO(3,1) absorption features.  The appearance of these features signals the presence of red supergiant (RSG) populations and hence can put a lower limit on the age of the cluster (Olofsson 1989). From the integrated magnitudes and sizes of the clusters we determined the nature of the star clusters, including the central object, and compared them to other clusters. ",
        "conclusions": "We have examined the nature of the central object in the irregular galaxy NGC 4449 through a near-infrared spectrum of stellar CO features and broad-band {\\it HST} images. The central object in NGC 4449 has many of the characteristics of a super star cluster. The broad-band colors and near-infrared spectrum are consistent with an age of 8--15 Myrs and with the cluster being in the phase dominated by RSGs. While not as compact as many super star clusters, it is still within the realm of globular cluster radii. In addition it is as luminous as many super star clusters. However, there is a peculiar nucleated-bar structure at its center that is not seen in star clusters. We suggest that this structure is related to the interaction that has produced the counter-rotating gas systems and extended gas streamers.  We have used the {\\it HST} images to identify additional compact star clusters. We identified 61 cluster candidates in NGC 4449 but suspect that 14 of these may be background elliptical galaxies. The cluster ages deduced from integrated photometry and compared to cluster evolutinary models span the timescale of the models up to 1 Gyr.  Apparently, NGC 4449 has been an actively star-forming galaxy for quite some time. Therefore, we did not find a ``special time'' when a large number of the star clusters in NGC 4449 formed, and so we cannot connect cluster formation to the date of an interaction. There are three other clusters that may be comparable to super star clusters and many more that may be comparable to the populous clusters found in the LMC. The rest are more modest groupings of stars. The formation of the more massive clusters, but not necessarily the super star clusters, may simply reflect the high level of star formation activity in NGC 4449 rather than its past interaction since NGC 4449 falls on the relationship found by Larsen \\& Richtler (2000) between the integrated properties of galaxies and their young massive star cluster content."
    },
    "0010/astro-ph0010501_arXiv.txt": {
        "abstract": "Faint quasar surveys are necessary complements to bright quasar surveys to remove the degeneracy between redshift and luminosity inherent in any single magnitude-limited sample. I discuss two ongoing surveys for faint quasars at $3.3 < z < 5$ and $5 < z < 6$ using imaging data from the Big Throughput Camera on the CTIO 4-m. I also discuss a sample of faint spectroscopically selected AGN with $z < 4.7$ found serendipitously in the CNOC2 Field Galaxy Redshift Survey.  Faint quasars at $2 < z < 3.5$ from this sample, when combined with literature data on more luminous quasars at the same $z$, show evidence for a much weaker Baldwin Effect in C\\,{\\sc iv}\\,$\\lambda$1549 and Ly$\\alpha$ than previously seen at $z < 1.5$. This may imply that the slopes of the Baldwin Effect for these transitions evolve with redshift, steepening with cosmic time. Finally I discuss the prospects for extending faint quasar surveys to $z \\sim 7$ using near-IR followup of very red objects in the Red-Sequence Cluster Survey. ",
        "introduction": "%  Active Galactic Nuclei (AGN) are very luminous objects whose emission is thought to be powered by the release of gravitational potential energy from matter falling into supermassive black holes. Most AGN physics beyond that simple statement are a matter of dispute at some level, as are many of the physical differences underlying the observed differences in the spectra of AGN.  One reason for at least part of our lack of understanding may be selection effects.  Most high-luminosity AGN --- quasars, with $M_B<-23$ --- have been selected using rest-frame ultraviolet-optical colors. Thus dust extinction has probably masked some portion of the dominant radio-quiet quasar (RQQ) population from detection (Webster et al. 1995). Important steps toward eliminating this bias are beginning to appear from near-IR selected samples (e.g. from 2MASS; see Cutri et al., this volume) and from X-ray selected samples % (Kim \\& Elvis 1999). Radio selection may not sample as distinct a population of AGN as once thought, given the lack of bimodality in $L_{rad}/L_{opt}$ between RQQs and RLQs seen in the FIRST Bright Quasar Survey (White et al. 2000), but much deeper radio surveys would be needed to select the majority of radio-quiet AGN at redshifts $z < 5$.  Another important selection effect is that any single flux-limited sample of AGN (even a near-IR or X-ray selected one) will exclude high-redshift, low-luminosity objects.  This leads to a degeneracy between the redshift and luminosity dependences of any AGN property. Thus, bright quasar surveys need to be accompanied by faint quasar surveys to fill in the $L$$-$$z$ plane and remove the redshift--luminosity degeneracy, and to define the full quasar luminosity function.  At the moment optical color selection of AGN is the method being used to assemble the largest samples of AGN, namely 2.5$\\times$10$^4$ objects from the 2dF Quasar Survey (2QZ; Boyle et al., this volume) and 10$^5$ from the Sloan Digital Sky Survey (SDSS; York et al. 2000). Moderately deep wide-field optical imaging can provide the data needed for complementary faint color-selected quasar surveys. Several surveys are underway which, despite their small size, will form useful comparisons to the 2dF and SDSS surveys (AGN will also turn up in faint galaxy surveys such as discussed by Le Fevre et al., this volume). The eventual goal for optical quasar studies is to reach $M_B=-23$ at all redshifts $z<7$. % ",
        "conclusions": "%  Large surveys of bright quasars need complementary % surveys of faint quasars if redshift and luminosity dependences of various quasar properties are to be disentangled. Together these surveys will enable study of the properties and evolution of quasars % over the entire optically accessible $z<7$ range.  The BFQS and BTC40 surveys are first steps toward large faint quasar surveys, for which 4-m class telescopes can provide imaging and some spectroscopy, but for which 8-m class telescopes are needed % for fainter candidates and % detailed spectral studies. The RCS is an excellent database in which to look for low-luminosity $5.5 < z < 7$ quasars, using snapshot near-IR imaging to separate them from very late M, L and T dwarfs with similarly red $R-z'$ colors.  Comparison of low-luminosity quasars from the CNOC2 survey to high-luminosity quasars at the same redshifts suggests an unexpected variation with redshift of the Baldwin Effect for C\\,{\\sc iv} and Ly$\\alpha$, illustrating the potential for surprises in the new era of wide-field astronomy."
    },
    "0010/astro-ph0010447_arXiv.txt": {
        "abstract": "We report early \\fuse \\ observations of the very hot DA white dwarfs PG1342+444 and REJ0558$-$371. Detection of photospheric absorption lines allows us to estimate the abundances of C, O, Si, P, S and Fe, the first measurements reported for DA stars at such high temperatures. Values of \\teff \\ and log g determined for PG1342+444 from the Lyman line series disagree with the results of the standard Balmer line analysis, an issue that requires further investigation. ",
        "introduction": "Very few true white dwarfs are known to have effective temperatures above $\\approx 70000$K and, therefore, the proposed direct evolutionary link between H-rich CSPN and white dwarfs has hardly been explored. The discovery of several very hot DA white dwarfs with temperatures in excess of 70000K (e.g. REJ1738+665 - 90000K, Barstow \\etal \\ 1994; EGB1 - 100000K, Napiwotzki \\& \\ Sch\\\"onberner 1993; WDHS1 - 100000 to 160000K, Liebert, Bergeron \\& \\ Tweedy 1994), through various observational programmes, provides an opportunity to study this upper range of the DA cooling sequence.  While Balmer line analyses (including the effect of metals) provide important measurements of the DA temperature scale, there remains a serious problem in their use to study the hottest DA stars. At effective temperatures in excess of $\\approx 50000$K, the photospheric hydrogen is almost completely ionized. Hence, the populations of the levels involved in the Balmer line transitions are severely depleted and the observed line strengths extremely weak. As a result, at high temperatures, the values of \\teff \\ become increasingly poorly determined. The use of the Lyman series lines has provided an additional tool to determine \\teff \\ and log g but only a limited number of observations have been available from the \\orf \\ and Hopkins Ultraviolet Telescope (\\hu ) missions (e.g. Dupuis \\etal \\ 1998; Finley \\etal \\ 1997).  The availability of the Far Ultraviolet Spectroscopic Explorer (\\fuse ), covering the $\\approx 900-1200$\\AA \\ wavelength range, presents us with an opportunity of obtaining many more Lyman series spectra for white dwarfs. We present \\fuse \\ observations of the very hot DA white dwarfs PG1342+444 and REJ0558$-$376, comparing the Balmer and Lyman series measurements of \\teff . An important result is the first detection of photospheric  OVI lines in the atmosphere of a DA white dwarf. ",
        "conclusions": "We have presented observations of some of the hottest known white dwarfs made with the \\fuse \\ spectrometer. Detection of photospheric CIV, SiIV, OVI, PV, SVI and FeVI absorption lines allows us to estimate the abundances of these elements, the first such measurements reported for DA WDs at these high temperatures. The observed abundances are similar to those found in the somewhat cooler star G191-B2B (\\teff $\\approx 54000$K). However, PG~1342+444 seems to have a large excess of C when compared to both REJ~0558$-$371 and G191-B2B. The OVI resonance lines are uniquely present in the \\fuse \\ wavelength range, and their detection in the PG~1342+444 and RE~J0558$-$371 spectra is the first such report for any hot white dwarf.  The Balmer and Lyman line studies yield values of \\teff \\ lower than reported for pure H Balmer line studies, as might be expected. The inconsistency between the PG~1342+444 Lyman and Balmer results sn an issue that requires further investigation. Forthcoming observations of the remaining targets in our sample will help examine whether or not the discrepancy originates in the model atmosphere calculations or arises from systematic effects due to interstellar absorption, or in the \\fuse \\ spectrometer calibration."
    },
    "0010/astro-ph0010392_arXiv.txt": {
        "abstract": "We present a 17-point map of the M17SW cloud core in the $1_{10}\\rightarrow1_{01}$ transition of ortho-\\water\\ at 557 GHz obtained with \\swas.  Water emission was detected in 11 of the 17 observed positions.  The line widths of the \\water\\ emission vary between 4 and 9 km s$^{-1}$, and are similar to other emission lines that arise in the M17SW core.  A direct comparison is made between the spatial extent of the \\water\\ emission and the \\thco\\ $J = 5\\rightarrow4$ emission; the good agreement suggests that the \\water\\ emission arises in the same warm, dense gas as the \\thco\\ emission.  A spectrum of the \\wateriso\\ line was also obtained at the center position of the cloud core, but no emission was detected.  We estimate that the average abundance of ortho-\\water\\ relative to \\hh\\ within the M17 dense core is approximately $1\\times10^{-9}$, 30 times smaller than the average for the Orion core.  Toward the \\hii\\ region/molecular cloud interface in M17SW the ortho-\\water\\ abundance may be about 5 times larger than in the dense core. ",
        "introduction": "M17SW is a prototypical giant molecular cloud core at a distance of approximately 2.2 kpc \\citep*{chi80} that was first studied in detail by \\citet{lad76}.  The dense core lies adjacent to a large optical \\hii\\ region and is oriented such that the \\hii\\ region/molecular cloud interface is viewed nearly edge-on. Because of this favorable geometry, this region has been the subject of numerous investigations studying the effect of UV radiation on heating, dissociation, and ionization of the gas within the molecular cloud.  The detection of emission from neutral and ionized fine structure lines of atomic carbon and high-J CO lines \\citep{kee85, har87, gen88, stu88, mei92} well away from the \\hii\\ region/molecular cloud interface suggests that the UV photons penetrate deep into the M17SW cloud core.  The ability of the UV photons to penetrate into the cloud has been attributed to the clumpy structure of the cloud core.  Indirect evidence for a clumpy cloud structure had been previously suggested by \\citet{sne84}, however high resolution observations of CO and CS presented by \\citet{stu90} reveal more directly the complex structure of this cloud core.  Besides studies of the water maser emission at 22 GHz, only one attempt has been made to detect water in M17SW. \\citet{wat80} report a marginal detection of the 183 GHz line of water toward the dense core. With the {\\it Submillimeter Wave Astronomy Satellite} (\\swas) we have observed the lowest energy rotational transition of ortho-\\water\\ and ortho-\\wateriso.  These transitions have upper state energies only 27 K above the ortho-water ground state, making observations with \\swas\\ a powerful means to probe water in the warm, dense molecular gas of M17SW. In this Letter we present a 17-point map that reveals extended \\water\\ emission from this region.  Observations of the \\wateriso\\ transition were made toward the center of the cloud core.  Based on these observations we estimate the relative abundance of ortho-water in M17SW and compare the results with that found for the extended water emission detected in Orion by \\swas\\ \\citep{sne00}.  \\begin{figure*}[t] \\epsscale{1.0} \\plotone{m17_h2o_table1.ps} \\end{figure*}  \\begin{figure*}[b] \\epsscale{1.10} \\plotone{m17_h2o_fig1.ps} \\caption{Spectra of the $1_{10}\\rightarrow1_{01}$ transition of ortho-\\water\\ ({\\it heavy lines}) and the $J=5\\rightarrow4$ transition of \\thco\\ ({\\it light lines}) obtained in M17SW.  The \\thco\\ spectra are divided by a factor of 10 in all cases except at positions (0.0, 0.0) and (-3.2, 0.0) where the spectra are divided by 20.  The sixteen spectra of each transition make up a $4\\times4$ map obtained on a regular grid separated by 3\\farcm2.  The spectra are shown in their correct relative positions on the sky.  Offsets in arcminutes relative to position $\\alpha = 18^{\\rm h} 20^{\\rm m} 22\\fs1$, $\\delta = -16\\arcdeg 12\\arcmin 37\\arcsec$ (J2000) are indicated on each spectrum. \\label{fig1}} \\end{figure*} ",
        "conclusions": "\\swas\\ has made the first detection of thermal water emission from M17 SW.  The emission observed by \\swas\\ is consistent in line width, line velocity, and spatial extent with the emission arising predominately from the warm dense core gas.  The \\swas\\ observations allow us to make the first estimate of the water abundance in this well-studied core.  The average abundance of ortho-\\water\\ relative to \\hh\\ in the cloud core is (1--2)~$\\times10^{-9}$.  Northeast of the core, toward the \\hii\\ region/molecular cloud interface, the relative water abundance is approximately five times larger.  Based on the \\wateriso\\ spectrum, the $3\\sigma$ upper limit on the relative \\water\\ abundance is $3\\times10^{-8}$.  Uncertainties in the derived abundance of water are dominated by the uncertainties in the physical conditions primarily density.  In the effectively thin limit, the abundance of \\water\\ is inversely proportional to the density \\citep[see][]{sne00}.  Therefore, if we had assumed a smaller average density, we would have derived a proportionally higher water abundance.  The study of \\citet{ber96} concludes that the bulk of the column density toward the M17 core arises in the dense gas.  Thus, we believe that uncertainties in the density and density structure cannot conspire to increase the water abundance by more than an order of magnitude.  The average relative abundance of water in the M17SW core is 30 times smaller than the average in Orion \\citep{sne00}.  Although there is substantial variation in the \\water\\ abundance in Orion, the abundance is always significantly larger than in the M17SW.  Only near the interface with the \\hii\\ region in M17SW does the abundance of \\water\\ approach values found for Orion.  The enhanced abundance of water in this interface region could be a result of the evaporation of water-ice--rich mantles from interstellar grains exposed to radiation from the \\hii\\ region. Further discussion of the chemical implications of these \\water\\ abundance results is presented in \\citet{ber00}."
    },
    "0010/astro-ph0010279.txt": {
        "abstract": "Radial velocity observations of the F8 V star $\\upsilon$ Andromedae taken at Lick and at Whipple Observatories have revealed evidence of three periodicities in the line-of-sight velocity of the star. These periodicities have been interpreted as evidence for at least three low mass companions (LMCs) revolving around $\\upsilon$ Andromedae. The mass and orbital parameters inferred for these companions raise questions about the dynamical stability of the system. We report here results from our independent analysis of the published radial velocity data as well as new unpublished data taken at Lick Observatory. Our results confirm the finding of three periods in the data. Our best fits to the data, on the assumption that these periods arise from the gravitational perturbations of companions in keplerian orbits, is also generally in agreement, but with some differences, from the earlier findings. We find that the available data do not constrain well the orbital eccentricity of the middle companion in a three-companion model of the data. We also find that in order for our best-fit model to the Lick data to be dynamically stable over the lifetime of the star ($\\sim\\!2$ billion years), the system must have a mean inclination to the plane of the sky greater than 13 degrees.  The corresponding minimum inclination for the best fit to the Whipple data set is 19 degrees.  These values imply that the maximum mass for the outer companion can be no greater than about 20 Jupiter masses. Our analysis of the stability of the putative systems also places constraints on the relative inclinations of the orbital planes of the companions. We comment on global versus local (i.e., method of steepest descent) means of finding best-fit orbits from radial velocity data sets. ",
        "introduction": "Radial velocity observations of several hundred nearby main-sequence stars have resulted in the detection to date of roughly forty companions with minimum or projected masses (i.e., $m \\sin i$) less than 80 $M_{\\rm J}$, where $M_J$ is the mass of Jupiter \\citep{mar98, mar99, vog00, may97}. Nearly thirty of these companions have $m \\sin i \\ltsim 10$ $M_{\\rm J}$. The evidence to date typically suggests only one companion per star; however, the data from some of the observational studies have suggested that other companions may be present in some cases but will require longer time bases for the observational record before firm conclusions can be established \\citep{cum99}.  Evidence for the presence of multiple low mass companions (LMCs) to a star would be significant for several reasons. A key reason is that it would suggest, at least superficially, a similarity of such a system to our planetary system, and by extension, planetary systems in general.  The first strong evidence for multiple LMCs to a single star, Upsilon Andromedae ($\\upsilon$ And), was reported recently by \\citet{but99} (hereafter referred to as B99). The data presented in that paper are from two independent studies, one conducted at the Lick Observatory and the other conducted at the Whipple Observatory using the Advanced Fiber-Optic Echelle (AFOE) Spectrograph.  Earlier observations of $\\upsilon$ And \\citep{but97} had detected a periodicity in the radial velocity data that indicated the presence of a companion with an orbital period of 4.6 days and a projected mass of $\\sim\\!0.7\\ M_{\\rm J}$. Those authors noted that the data also contained ``evidence for variability in the gamma velocity with timescale of about 2 yr.\" The newer observations (B99) reveal additional periodicities, one in excess of 1200 days (1269 days for the Lick data and 1481 days for the AFOE data), as well as one with $\\sim\\!240$ days.  B99 have modeled these periodicities as arising from the presence of three LMCs in keplerian orbits about $\\upsilon$ And. The eccentricities of the orbits determined by B99 are, in order of increasing orbital period, 0.042, 0.23, and 0.36. This three-companion model for the observations raises interesting challenges in understanding the formation and evolution of that system and its possible relationship to systems such as the Solar System. Particular challenges relate to the dynamical stability of such a system and what constraints it might place on physically realizable companion systems, and to how a system consisting of at least three relatively massive objects could form around a star with the orbital structure that is suggested by the data.  In an effort to explore these challenges in more detail, we re-examine here the published radial velocity data on $\\upsilon$ And, as well as data taken subsequent to announcement of the results and kindly provided to us by G.~Marcy.  Section 2 summarizes the radial velocity data that are used in our analyses.  Models for analyzing the radial velocity data are discussed in Section 3.  The methods that we used for fitting the data and the procedures for assessing the merit of those fits are described in Section 4. The best-fit models, assuming that the periodicities are due to companions (i.e., that the model for describing the data is one comprised of a superposition of Keplerian motion) are presented in Section 5. We examine the dynamical stability of candidate model systems in Section 6. A summary of our results and conclusions regarding the possible nature of the $\\upsilon$ And system are given in Section 7.  % % ",
        "conclusions": "\\subsection{Models}  We have analyzed the radial velocity data given in B99 on $\\upsilon$ And, as well as more recent unpublished data on $\\upsilon$ And taken by the Lick Observatory group. Our motivation for re-examining the B99 data was the prospect of finding feasible (in the $\\chi^2$ sense) models that would be more dynamically stable than those published in B99. Our methodology differed from that of B99 in the following aspects.  First, we explicitly did not analyze what is referred to in B99 as the combined data set. We do so for two reasons.  One is that combining data sets taken with different instruments incurs some risk of introducing systematics into the combined data. More importantly, the existence of two independent data sets serves the valuable purpose of corroborating one set of data by the other. This corroborative aspect is important to maintain when there are few systems capable of measurements with high precision. In the longer term, a continuing set of independent observations could reveal the presence of systematics in one of the systems should they develop. Finally, to the extent that any model of the data describes physical reality, the model should fit all data sets.  Inconsistencies between data sets in this regard should be treated as a warning that the model may be incomplete or incorrect.  We would encourage observers to maintain this important independence in their data sets, while recognizing the inherent strengths (such as better characterization of candidate models) of a larger combined data set.  Second, in finding the minima of the $\\chi^2$ function we relied on a {\\it global} search method -- the genetic algorithm. In cases where there is reason to suspect that there may be multiple periods in the data, the $\\chi^2$ function is likely to be complicated. Such functions are referred to as having a ``rugged landscape\" because they have a large number of intermingled local minima and maxima.  The genetic algorithm will find the global minimum (and other ranking minima) even for functions with an extremely rugged landscape. This is in contrast to the standard, steepest gradient descent, minimization method which is {\\em local} and converges to a local minimum located closest to the starting values.  Thus, the local method does not guarantee finding the global minimum.  Moreover, as starting values of periods can only be provided from data frequency spectrum, the local method requires estimating and interpreting such a spectrum.  Extracting periodicities from the data spectrum may be a laborious process not free of ambiguity, as documented by efforts in B99 and also in our present paper. For example, consider differences in frequency spectra between different data sets (Fig.~\\ref{fig1}). A good illustration of the difficulties in extracting periodicities from the frequency spectrum is provided by the inconsistency of the spectra of residuals of two-companion fits to the AFOE data. (We note that B99 reported checking the uniqueness of their three-companion model with the genetic algorithm, although no details were given.) These problems can be entirely avoided by using the global minimization technique. In the present paper, in addition to using the genetic algorithm, we also employed the local method, both for continuity with B99 and also as a means of fine-tuning the global solution. We recommend that future studies avoid relying exclusively on the local minimization technique.  Third, instead of looking for the ``best-fit\" model we looked for a {\\it set} of viable models. We did not specify formal criteria for model ``viability\", but very roughly, the ``good-fit\" models are characterized by: (a) fitness that is not much smaller than the fitness of the BF model, and (b) residuals with no periodic components.  We have found that two-companion models are not viable, primarily because they leave periodic residuals. We have identified four potentially viable three-companion models.  Our best-fit, three-companion Keplerian model is consistent with that of B99.  There are differences in details of the orbital parameters, but nothing fundamental. The only significant difference between the best-fit model and the second-best-fit model (SEBC) is the small eccentricity of companion C's orbit. Interestingly, the long term dynamical behavior of the SEBC model is qualitatively similar to the best-fit model. The third-best-fit model (SE) has all three companions on orbits with small eccentricities. Finally, we have found that a model (PC145) with companion C orbiting the star with a period of 145 days also offers a good fit to the data. These last two models describe a system which is potentially more stable than that described by the BF model, but the BF model is most consistent with the data.  The new Lick data provides further strong support for the BF model. This is the only model for which the value of $\\chi^2_{red}$ has decreased with addition of the new data, as expected for a physically viable model. Although the value of $\\chi^2_{red}$ for the SEBC model has increased slightly, this model still deserves being monitored against future data. Values of $\\chi^2_{red}$ for SE and PC145 models increased significantly, practically eliminating these models from consideration. We conclude that parameters of the three-companion model of the $\\upsilon$ And radial velocity data are well constrained by the data, with the exception of the eccentricity of the middle companion. Regardless of the eccentricity of the middle companion the system described by such a model presents a challenge from the point of view of dynamical stability.  Because of dynamical stability concerns and also because the present data allows the $\\sim\\!240$\\,d periodicity to be almost sinusoidal, one can contemplate a model consisting of two companions, B and D, and a stellar cycle with a period of $\\sim\\!240$\\,d causing a radial velocity variation of about 50\\,m\\,s$^{-1}$.  There is some evidence that magnetic cycles can indeed induce periodic changes in integrated light velocity. \\citet{dem94} reported a sinusoidal variation of solar integrated light velocity with semiamplitude of 14\\,m\\,s$^{-1}$ and period of 11 years. This variation is positively correlated with the solar disk-averaged magnetograph signal and is over three times as large as the Sun's jitter of 4\\,m\\,s$^{-1}$. However, \\citet{mcm93}, using a different technique, have not found any long-term variability in solar integrated light velocity. Thus, at present such a model is speculation, especially as it is not clear that a stellar magnetic cycle with a sub-yearly period is possible and capable of causing light velocity variations as large as 50\\,m\\,s$^{-1}$ which would be necessary to explain the $\\upsilon$ And observations.  \\subsection{Orbital dynamics constraints}  The radial velocity data do not provide constraints on all the orbital elements of the putative companions: the orbital inclinations to the plane of the sky and the orientation of the lines of nodes in the plane of the sky remain undetermined.  Direct constraints on the viewing geometry of the $\\upsilon$ And system are few. The fact that there is no transit phenomenon \\citep{henry00}, as has recently been found for the star HD209458 \\citep{cha00} , implies that the orbital inclination of the inner companion must be $\\ltsim 82.8$ degrees. Additional constraints on the inclination of the outer companion's orbit come from observations of $\\upsilon$ And by Hipparcos. Using $2\\sigma$ detection limits for Hipparcos accuracy for relative astrometric observations, which is roughly 2 milliarcseconds \\citep{perryman96}, the absence of a reliable detection of a perturbation due to a companion with the orbital periods given by the best-fits gives an upper limit to the true companion masses of 27.7 $M_{\\rm J}$ and 25.12 $M_{\\rm J}$ for the new Lick and AFOE data respectively. Using the minimum masses for the corresponding best-fits, $\\sim 3.76$ $M_{\\rm J}$ and $\\sim 4.92$ $M_{\\rm J}$, the absence of astrometric detection means that the inclination must be $\\gtsim 7.8$ degrees ($\\sin i \\gtsim 0.136$) and $\\gtsim 11.3$ degree ($\\sin i \\gtsim 0.196$) for the new Lick and AFOE data respectively. (Note that the minimum masses given for companion D in B99, Table 3, columns 2 and 3, are in error.  They are too high by a factor of ${(1-e^2)}^{-1/2}$.)  \\citet{maz99} have used a combination of Hipparcos data and the spectroscopically-determined best-fit orbital elements from B99 to deduce a semi-major axis for the orbit of the outer companion and hence an estimate of the true mass.  (This is not inconsistent with the fact that Hipparcos did not detect an astrometric perturbation for $\\upsilon$ And.) The $2\\sigma$ mass range from Mazeh et al.~is $10.1^{+9.5}_{-6.0}$ $M_{\\rm J}$. This approach is useful, but it depends upon the assumed set of spectroscopic elements. The spectroscopic elements we find here for the best-fit model differ somewhat from those used by Mazeh et al.  The robustness of the derived mass against variations in these elements is unknown.  Taken at face value, this mass would constrain the orbital inclination of the outer companion, at the $2\\sigma$ level, to be $\\gtsim 11$ degrees (14.5 degrees) for the new Lick (AFOE) data.  Here we have used orbital dynamics and stability considerations to infer constraints on the undetermined orbital parameters, using the fact that the mutual gravitational interactions in the system depend both upon the true companion masses and on all the orbital elements. {\\it Our studies show that dynamical stability requires $\\sin i \\gtsim 0.23$ for the best-fit model to the New Lick data, and $\\sin i \\gtsim 0.33$ for the best-fit model to the AFOE data.} The $i$ here refers to the present orbital inclinations of companions C and D, assumed equal but not necessarily coplanar. Thus the most massive companion, D, can be at most about 20 Jupiter masses. Our studies also show that the range of allowed relative inclinations of the orbits of the outer two companions C and D decreases with decreasing $\\sin i$ (that is, it is sensitive to the masses of the companions). Even for values of $\\sin i$ close to 1, their relative inclination can be no greater than about 60 degrees for the Lick best-fit model ($\\sim\\!30$ degrees for the AFOE best-fit model). Finally, our analysis provides constraints on the lines-of-nodes of the orbits of companions C and D (Fig.~8).  The above conclusions are qualitatively consistent with those of \\citet{lau99} and \\citet{riviera00} whose studies consisted of a survey of the long term dynamical stability of B99's candidate models for the $\\upsilon$ And system. Laughlin and Adams used statistical arguments based on the long term behavior of the orbital elements (of initially nearly coplanar orbits) to infer likely values of the relative orbital inclination and orbital eccentricities of the outer companions. (Note that these authors used an erroneous definition of relative inclination, namely ``$i_C-i_D$''; the correct definition is given in Eqn.~7.  It is unclear how their conclusions regarding the likely relative inclinations would be affected if this error were corrected.) Our focus here was on using stability considerations to quantitatively constrain possible orbital geometries.  All three dynamical studies suggest that small differences in the parameters of the three-companion models lead to significant differences in the dynamical stability of the system. This extreme sensitivity, while advantageous in constraining the nature of the system if one has a firm handle on the parameters, raises questions as to how general the conclusions can be when the parameters are not well known --- as is the case here.  Our analysis of the orbital evolution shows that for small relative inclinations of the outer two companions C and D, the eccentricity of companion C varies periodically from a maximum of $\\sim\\!0.3$ to a minimum of essentially zero with a period of a few thousand years. This peculiar behavior may harbor clues to the origin of the unusual orbital arrangement in this system. It is noteworthy that our analysis of the radial velocity data led to the conclusion that the eccentricity of companion C is not very well constrained by the observations: models with small eccentricities for companion C provide nearly as good fits to the data (in a $\\chi^2$ sense) as the best-fit solutions."
    },
    "0010/astro-ph0010117_arXiv.txt": {
        "abstract": "Grains exposed to anisotropic radiation fields are subjected to forces due to the asymmetric photon-stimulated ejection of particles.  These forces act in addition to the ``radiation pressure'' due to absorption and scattering. Here we model the forces due to photoelectron emission and the photodesorption of adatoms.  The ``photoelectric'' force depends on the ambient conditions relevant to grain charging.  We find that it is comparable to the radiation pressure when the grain potential is relatively low and the radiation spectrum is relatively hard.  The calculation of the ``photodesorption'' force is highly uncertain, since the surface physics and chemsitry of grain materials are poorly understood at present. For our simple yet plausible model, the photodesorption force dominates the radiation pressure for grains larger than $\\sim 0.1 \\micron$ exposed to starlight from OB stars. We find that the anisotropy of the interstellar radiation field is $\\sim 10\\%$ in the visible and ultraviolet.  We estimate size-dependent drift speeds for grains in the cold and warm neutral media and find that micron-sized grains could potentially be moved across a diffuse cloud during its lifetime. ",
        "introduction": "Anisotropic radiation fields are common in the interstellar medium (ISM). Of course, the radiation is highly anisotropic at points located close to a star or cluster or at the edge of a molecular cloud.  These include the astronomically important photodissociation regions (PDRs), the surface layers of molecular clouds, where incident UV dissociates molecules. The visible/UV radiation field in the diffuse ISM is due primarily to starlight; since stars are unevenly distributed, it too is anisotropic. Dust grains in this radiation field are subjected to torques, if their shapes are somewhat asymmetric, and this may be resposible for the alignment of interstellar grains (Draine \\& Weingartner 1996, 1997), which is inferred from the observed polarization of starlight (Hiltner 1949a, b; Hall 1949; Hall \\& Mikesell 1949).  Grains exposed to anisotropic radiation are subjected to forces as well as to torques, again with potentially important consequences.  If the ``radiative'' forces are large enough to result in substantial gas-grain drift, then this could lead to spatial variation in the dust-to-gas ratio. If the drift depends on grain size, then grain coagulation could result. Here we investigate two potentially significant forces, due to the recoil associated with electrons and adsorbed H atoms, which are ejected from the grain with some non-zero probability following photon absorption.  Lafon (1990) modeled the net force due to anisotropic photoelectron emission for grains exposed to unidirectional radiation fields and found that, for some interstellar conditions, it substantially exceeds the radiation pressure, due to absorption and scattering.  Lafon considered spherical grains and assumed that electrons are emitted only from the illuminated hemisphere. However, interstellar grain radii can be smaller than the photon attenuation length, so that electrons are also emitted from the non-illuminated hemisphere.  In \\S \\ref{sec:fpe}, we revisit the calculation of photoelectric recoil forces on dust grains, employing the prescription of Kerker \\& Wang (1982) for estimating the anisotropy in photoelectron momentum.  In \\S \\ref{sec:fpd}, we consider the force due to the anisotropic photodesorption of adsorbed H atoms.  An adatom starts out in the gas phase, collides with and sticks to the grain surface, and can be removed when an absorbed photon breaks its bond to the surface.  Of course, there are many other grain surface processes, besides adsorption and photodesorption, and these too could play important roles in determining the emission anisotropy. Since interstellar grains have not yet been well characterized, the details of their surface physics are unknown.  We adopt a simple, yet plausible, model and evaluate the ``photodesorption'' force for this one case.  Following the investigations of the photoelectric and photodesorption forces, we discuss, in \\S \\ref{sec:drag}, how surface processes can affect the drag force experienced by drifting grains.  In \\S \\ref{sec:anisotropy}, we estimate the anisotropy of the interstellar radiation field. In \\S \\ref{sec:drift_diffuse}, we estimate drift speeds for grains in the diffuse ISM, tying together the results from the previous sections.  In a separate paper (Weingartner \\& Draine 2001c), we will investigate gas-grain drift in PDRs. ",
        "conclusions": "We have investigated two forces that act on dust grains exposed to anisotropic radiation fields, due to the asymmetric photon-stimulated emission of particles from the grains.  First, we have improved on Lafon's (1990) treatment of the photoelectric force, due to the recoil associated with photoelectrons, by using the realistic prescription of Kerker \\& Wang (1982) for estimating the emission asymmetry and employing the recent photoemission model of Weingartner \\& Draine (2001b).  The photoelectric force depends on the ambient conditions relevant to the grain charging, including the shape of the radiation spectrum and the parameter $G \\sqrt{T}/n_e$.  We find that the photoelectric force is comparable to the radiation pressure in regions where the grain potential is relatively low and the radiation spectrum is relatively hard (Figures \\ref{fig:fpegra}--\\ref{fig:fpetc}).  We find that the photoelectric force is unimportant compared with the radiation pressure in the cold and warm neutral media for grains larger than a few hundred $\\Angstrom$, because of the softness of the interstellar radiation field, as estimated by Mezger et al. (1982) and Mathis et al. (1983).  We have also investigated the force resulting from the asymmetric photodesorption of H atoms adsorbed on the grain surface.  The surface physics and chemistry of likely grain materials are not currently well-known.  We have adopted a very simple, yet plausible, model in order to examine the potential importance of this force.  In particular, we assume that H chemisorbs at all surface sites with sufficient binding energy to supress thermal desorption and site-to-site diffusion.  In our model, the total force associated with gas-phase atoms which collide with the surface includes contributions from three processes:  photodesorption, H$_2$ formation, and elastic reflection of atoms incident on the grain surface.  We call the force on a grain that is not drifting with respect to the gas the ``photodesorption'' force.  When the grain drifts, the total force (due to the above three processes) minus the photodesorption force equals the drag force, which is thus modified from the traditional treatment considering only the reflection of incident atoms (\\S \\ref{sec:drag}).  With our standard choices for parameter values (Table \\ref{tab:pdpars}), we find that the photodesorption force dominates the radiation pressure for grains with $a \\gtsim 0.1 \\micron$ when the radiation field is relatively hard (Figure \\ref{fig:fpd1}).  Laboratory studies of the surfaces of candidate grain materials, including determinations of photodesorption cross sections, are needed in order to clarify the true importance of this force.  Grains exposed to anisotropic radiation will drift relative to the gas due to the combined action of the radiation pressure, photoelectric force, and photodesorption force.  We find that the radiation field in the diffuse ISM is $\\sim 10\\%$ anisotropic in the visible and ultraviolet (\\S \\ref{sec:anisotropy}).  In \\S \\ref{sec:drift_diffuse}, we estimate the maximum likely drift speeds for grains in the cold and warm neutral media.  We find the timescale for grain coagulation resulting from this drift to be $\\gtsim 5 \\times 10^9 \\,$yrs.  Large grains ($a \\sim 1 \\micron$) might drift rapidly enough to be substantially separated from the gas in cold diffuse clouds.  In a separate paper (Weingartner \\& Draine 2001c), we will investigate gas-grain drift in photodissociation regions."
    },
    "0010/astro-ph0010321_arXiv.txt": {
        "abstract": "The integrated spectrum of OH emission from late-type stars is often circularly polarized, by as much as 50\\% in some cases. While the spectra are partially polarized, the individual maser components revealed by VLBI are much more so. Using VLBI observations of late-type stars from the literature, we show that the difference in circular polarization between main lines correlates with a difference in angular extent for a given object.  This is a natural result if turbulent magnetic fields are causing the masers to be polarized via the Cook mechanism, and might serve as a good diagnostic for determining which objects should be investigated in the search for magnetic fields around evolved stars. ",
        "introduction": "In the recent past, maser studies of late-type stars have undergone a resurgence in popularity. U Orionis experienced an unusual outburst in 1972, and several papers about this source were published beginning in the late 1970's (Reid et al. 1979, Cimerman 1979, Fix 1979, Fix et al. 1980, Jewell, Webber and Snyder 1981, Chapman and Cohen 1985). Since then, many authors have used VLBI techniques to determine the distribution, strength, and polarization of OH maser components in various stars (e.g. Chapman, et al. 1994, Szymczak, Cohen and Richards 1998, Sahai, et al. 1999, Szymczak and LeSqueren 1999, Palen and Fix 1999). These observations have been extremely useful in investigations of envelope dynamics and stellar mass loss.  In the early 1970's, Fix began a monitoring program of OH/IR stars at Arecibo observatory. In the course of this program, it became apparent that the OH maser emission was undergoing significant variations in amplitude (as much as 2-3 Jy at the peak) and velocity (of order a few km/s) of spectral features over time scales as short as two years. In 2000, Palen and Fix presented a model which uses a turbulent plasma to vary the coherence of the maser components in order to explain the temporal variation of these masers. This work was based primarily on observations of one of the most well-studied OH/IR stars, U Herculis.  Fortunately, this model makes testable predictions. The first is that objects with a higher degree of polarization (here and throughout, 'polarization' is substituted for 'circular polarization') will be more variable.  This prediction was tested using two decades of Arecibo data by Fix (2000), and he finds a strong correlation between variability and polarization.  Second, this model predicts that in the absence of a magnetic field, masers will be constrained to appear in a region near the central star.  In a smooth non-magnetized radial outflow, the masers are confined to the region of the shell near the line of sight to the central star (Figure 1), because that is the only region where the emitting molecules are strongly velocity coherent along the line of sight.  If, however, the outflow carries a magnetic field, in particular a turbulent magnetic field, then the Cook mechanism (Cook 1975) simulates the required velocity coherence by Zeeman splitting the lines.  This mechanism produces 100\\% polarized emission in each component.  Therefore, we can predict that sources with highly polarized spectra will be composed of maser emission which may arise in regions far from the central star. Masers in sources with less polarized spectra will be confined to a location near the central star.  If the medium is turbulent in both velocity and magnetic field, then the maser emission will be produced by many regions of small angular extent, and nearly 100\\% polarized emission.  In most recent VLBI observations of late-type stars, the masers are partially resolved, and the emission is seen to come from many small, highly polarized components (see, for example, Palen and Fix 2000, Chapman et al. 1994, and Szymczak, Cohen and Richards 1999).  Here we show that in more highly polarized sources these maser regions are found farther from the central star.  In Section \\ref{observations}, we describe the observations taken from the literature, and give some background information about these objects.  In Section \\ref{comppol}, we determine the degree of polarization in the spectra of four late-type stars, and in the individual maser components. In Section \\ref{compext}, we find the extent of the maser emission in these stars, and show that this is correlated with the degree of polarization. In Section \\ref{discussion}, we discuss these results, and make recommendations for future investigation of the problem. ",
        "conclusions": "\\label{discussion}  The plasma turbulence model predicts that the extent of maser emission should correlate with the degree of polarization. We have found this to be true for four late-type stars. The result suggests that the plasma turbulence model of maser emission may be generally applicable.  Correlation does not imply cause and effect, and alternate explanations for this correlation need to be investigated.  A few possible causes can be immediately investigated and ruled out.  In this group of four stars, two show 1665 MHz emission as most polarized and most extended, and two show 1667 MHz emission as most polarized and most extended. There is no preference for one main line to be produced farther from the central star. There is also no correlation in the data between the flux in the line and polarization, or between flux and extent of the emission. This implies that sensitivity of the instruments in the main lines is not complicating the issue.  Finding suitable objects for a more extensive study of this type is difficult. They must be bright in two lines which are close together in frequency. The stars must be near enough that the maser components are resolved by VLBI, and they must be observed simultaneously in both polarizations and both main lines. Thus far there are few suitable candidates in the literature.  This verification of the presence of magnetic fields in these late-type stars lends credibility to the argument that complex, symmetric shapes observed in planetary nebulae may be caused by magnetic shaping. If we can begin to establish a connection between varieties of late-type stars, and the various morphologies of planetary nebulae, we will have come a long way towards an understanding both types of objects.  A similar study of young stellar objects (YSO's) would be very interesting, since a positive result would imply the presence of magnetic fields in their envelopes and accretion disks."
    },
    "0010/astro-ph0010051_arXiv.txt": {
        "abstract": "We have carried out radiative transfer calculations of infalling, dusty envelopes surrounding embedded protostars to understand the observed properties of the recently identified ``Class 0'' sources. To match the far-infrared peaks in the spectral energy distributions of objects such as the prototype Class 0 source VLA 1623, pure collapse models require mass infall rates $\\sim$10$^{-4}\\msun$yr$^{-1}$. The radial intensity distributions predicted by such infall models are inconsistent with observations of VLA 1623 at sub-mm wavelengths, in agreement with the results of Andr$\\acute{\\rm e}$ et al. (1993) who found a density profile of $\\rho \\propto r^{-1/2}$ rather than the expected $\\rho \\propto r^{-3/2}$ gradient. To resolve this conflict, while still invoking infall to produce the outflow source at the center of VLA 1623, we suggest that the observed sub-mm intensity distribution is the sum of two components: an inner infall zone, plus an outer, more nearly constant-density region.  This explanation of the observations requires that roughly half the total mass observed within 2000 AU radius of the source lies in a region external to the infall zone.  The column densities for this external region are comparable to those found in the larger Oph A cloud within which VLA 1623 is embedded.  This decomposition into infall and external regions is not unique, owing to uncertainty in the structure of the molecular gas outside of the infall zone, which in turn implies some uncertainty in estimating the infall rate.  Nevertheless, the environment of Oph A is so dense that any protostellar clouds which fragment out are likely to collapse at very high infall rates, consistent with our spectral energy distribution modeling. The extreme environments of Class 0 sources lead us to suggest an alternative or additional interpretation of these objects: rather than simply concluding with Andr$\\acute{\\rm e}$ et al. that Class 0 objects only represent the earliest phases of protostellar collapse, and ultimately evolve into older ``Class I'' protostars, we suggest that many Class 0 sources could be the protostars of very dense regions, and Class I objects found in lower-density regions may be in comparable evolutionary states. ",
        "introduction": "An important aim of the study of young stellar objects (YSOs) has been to classify them along an evolutionary sequence. According to the widely used classification scheme of Lada \\& Wilking (1984) and Lada (1987), YSOs are divided into three classes based on the slopes of their infrared spectral energy distribution (SED). The reddest objects, dubbed Class I sources, are thought to be protostars surrounded by infalling envelopes. Their SEDs are consistent with dust emission from envelopes with infall rates near those predicted by the standard theory of protostellar formation, which invokes free-fall collapse from an initially thermally-supported, isothermal, singular molecular cloud core in hydrostatic equilibrium (Adams \\& Shu 1986; Adams, Lada \\& Shu 1987). Kenyon, Calvet \\& Hartmann (1993, hereafter KCH), in particular, were able to match the observed SEDs of Class I sources in Taurus-Auriga with rotating infall models of the type described by Terebey, Shu \\& Cassen (1984). This agreement between theory and observation is generally seen as testimony to the success of the standard model of protostellar collapse.  Recently, Andr$\\acute{\\rm e}$, Ward-Thompson \\& Barsony (1993, hereafter AWB) identified YSOs characterized by blackbody-like SEDs that peak at submillimeter wavelengths. The low dust temperatures ($T_{d}$ $\\sim$ 20K) and large dust masses inferred for these sources led AWB to suggest that these objects constitute a new class of objects, called ``Class 0'', which might be even younger than typical Class I objects. Observationally, the defining characteristics of Class 0 sources include a high ratio of submillimeter to bolometric luminosity (L$_{submm}^{\\lambda > 350\\mu m}$/L$_{bol}$ $>$ 5 $\\times$ 10$^{-3}$), invisibility at near-infrared wavelengths, and the presence of molecular outflows (Barsony 1994). Andr$\\acute{\\rm e}$ \\& Montmerle (1994) argued that Class 0 sources have more circumstellar mass than stellar mass. In this picture, Class 0 sources have not yet assembled the bulk of their final stellar mass and are still in the main accretion phase. Consequently, the authors argue that Class 0 sources are the youngest protostars, and subsequently evolve into Class I sources (see Andr\\'e, Ward-Thompson, \\& Barsony 2000 $=$ AWB2).  Following AWB's identification of the prototype Class 0 source VLA 1623 in the rich star-forming core $\\rho$ Ophiuchi A, several other objects of the class have been found in Orion B (Ward-Thompson et al. 1995), Serpens (Hurt \\& Barsony 1996), Cep E (Lefloch et al. 1996), and Perseus (O'Linger et al. 1997) molecular clouds. In a spectroscopic survey of 47 candidate protostars, Mardones et al. (1997) report that Class 0 sources have stronger infall signatures --i.e., larger negative velocity shifts-- than Class I sources, and attribute that result to a difference in the physical conditions in the circumstellar envelopes of the two source types.  In order to understand better the observed characteristics of this `youngest' population of YSOs, we have performed detailed radiative transfer calculations of dust envelope emission in Class 0 sources, following the methods described by KCH and Calvet et al. (1994). We also study the observed radial intensity profiles of the spatially-resolved prototype Class 0 source VLA 1623 at three submillimeter wavelengths. In both cases, we compare the observations to the predictions of the standard model, and explore the possibility that environments exterior to the infall region can affect the spectral energy distribution and the inferred density gradients. ",
        "conclusions": "\\subsection{Environment and high infall rates}  In the previous section we showed that models of a collapsing isolated cloud can fit the SEDs of Class 0 source if a high infall rate is adopted, but that the resulting surface brightness distributions are inconsistent with observations, as originally found by AWB.  We have extended this result to include the effects of external heating, and still find a discrepancy between theory and observation.  The surface brightness distributions can be explained with simple infall models only by including an additional emitting component, which could be material in the near environment of the collapsing cloud.  Infall models generally exhibit surface density distributions $\\Sigma \\propto r^{-1/2}$ (implying $\\rho \\propto r^{-3/2}$ in spherical geometry), whether from spherical or flattened clouds, similarity solutions for collapse or numerical time-dependent calculations, with or without cloud magnetization (e.g., Larson 1969; Penston 1969; Shu 1977; Foster \\& Chevalier 1993; Galli \\& Shu 1993a,b; Hartmann et al. 1994; Hartmann, Calvet, \\& Boss 1996; Li \\& Shu 1996). In isothermal clouds at the point of collapse, many calculations show density distributions which tend to steepen more towards $r^{-2}$ (e.g., Mouschovias 1991, and references therein; Foster \\& Chevalier 1993). These models strongly suggest that the observations of Class 0 sources encompass not only an infall region, but external material not necessarily in collapse.  The question then arises: if the external environment contributes a large fraction of the sub-mm flux, can we in fact say anything about infall rates?  A static cloud, with density distribution $\\rho \\sim r^{-1/2}$ as considered by AWB, could, if physically realizable, fit the observations in the absence of {\\it any} infall. (Note that such a static cloud is inconsistent with the standard model of isothermal clouds, in which a $\\rho \\propto r^{-2}$ density distribution is expected). It seems likely that some fraction of the sub-mm emission arises from an infall region, because some infall must have occurred to produce the central source, and given that pre-stellar cores appear to be much less centrally condensed than Class 0 sources (Ward-Thompson et al. 1994; Andr\\'e et al. 1996; Ward-Thompson, Motte, \\& Andr\\'e 1999; AWB2).  Nevertheless, let us take the extreme limit by supposing that {\\it all} of the sub-mm emission arises from a static cloud. We may then estimate the rate at which this cloud would collapse. For a cloud mass of $\\sim 0.6 \\msun$ concentrated within a radius of 2000 AU, the free-fall timescale is $\\sim 2 \\times 10^{4}$ years. Thus, if this cloud were to collapse, it would exhibit an average infall rate of $\\sim 3 \\times 10^{-5} \\msunyr$. This estimated infall rate is not much smaller than the $\\sim 10^{-4} \\msunyr$ estimated for the pure infalling envelope model based on the SED fitting. Moreover, if the region that fell in to form the outflow source was originally {\\it denser} than the outer static region, as seems probable, the infall rate could be higher than this estimate. Thus, one is lead to the same order of magnitude infall rate from consideration of the spatial concentration of the local gas, independent of the fraction of emission currently arising from an external cloud rather than the infall region.  High infall rates are also suggested by observations of outflows. In a survey of nearby embedded YSOs, Bontemps et al. (1996) found that Class 0 sources have unusually powerful outflows. The Class 0 objects in their sample lie an order of magnitude above the well-known correlation between the outflow momentum flux (integrated over time) and bolometric luminosity that holds for Class I sources. Since models of bipolar outflows (Shu et al. 1994, Ferreira \\& Pelletier 1995) predict that the mass-loss rate is some fraction of the mass-accretion rate, the stronger outflows in Class 0s also call for significantly higher infall rates than in Class Is. Henriksen, Andr\\'e, \\& Bontemps (1997) attempt to derive accretion rates as a function of time for Class 0 sources which are consistent with infall at non-constant rates, again finding very rapid mass accumulation is required to explain outflow properties.  In addition, from a spectroscopic survey of 47 candidate protostars, Mardones et al. (1997) report that Class 0 sources have stronger infall signatures --i.e., larger negative velocity shifts-- than Class I sources. One possible interpretation of their results is that Class 0s have higher density envelopes than Class Is, again implying higher infall rates in Class 0 objects.  Thus, our high infall rates, which imply that Class 0 sources are a short-lived phenomenon, are consistent with previous conclusions in the literature (e.g., AWB). A comparison of source statistics in Taurus-Auriga and Ophiuchus support this conclusion. With an infall rate of $10^{-4} \\mathrm{\\msunyr}$, a 0.5 $\\msun$ star would be accreted in $\\sim$5000 years, a timescale consistent with the dynamical lifetime of VLA 1623.  Andr\\'e \\& Montmerle (1994) and Barsony (1994) estimate a lifetime of $\\sim 10^4$ yr for Class 0 sources in Oph on statistical grounds. In addition, as there are around 200 known pre-main-sequence objects in Taurus, one would expect 5000 yrs$/10^{6}$ yrs $\\times$ 200 or $\\sim$1 Class 0 source; and Taurus-Auriga contains only two ``border-line'' Class 0 sources, L1527 and L1551 IRS 5 (Chen et al. 1995). (AWB2 suggest that two Class 0 sources are present in Taurus, L1527 and the extremely low-luminosity source IRAM 04191+1522 [Andr\\'e, Motte, \\& Bacmann 1999]).  One problem with the picture of high infall rates for Class 0 sources is that much higher accretion rates onto the central protostar should lead to high accretion luminosities. However, Class 0 objects are not very much more luminous than their Class I counterparts. Henriksen, Andr\\'e, \\& Bontemps (1997) and Andr$\\acute{\\rm e}$ (1997) have suggested several factors that may reduce the theoretical accretion luminosity: in Class 0 sources, the central stellar mass might be lower, the stellar radius probably larger, and the amount of accretion energy dissipated in the wind is also likely to be larger. Another natural explanation is that material falls onto a disk, instead of falling directly onto the central star. However, the material must still somehow flow from the disk to the star at some point in the protostellar phase to build up the central star. It may be that Class 0 sources also undergo ``FU Orionis-type'' outbursts once enough material piles up in the disk to make it unstable (Kenyon et al. 1990, 1994).  Note also that an ``external region'' is not necessarily physically unconnected with the Class 0 source; i.e., this material may eventually collapse onto and be incorporated into the central star(s). Whether this is necessarily the case, however, is not clear.  \\subsection{Class 0 vs. Class I: the evolutionary model vs. initial density}  AWB suggested that the Class 0 sources represent the earliest phase of protostellar collapse, based on the large amount of circumstellar material in the near environment (compared with the likely mass of the central protostellar object). Henriksen, Andr\\'e, \\& Bontemps (1997) elucidated a picture which in principle provides a theoretical basis for this evolutionary scheme.  Henriksen et al. pointed out that theoretical models of clouds initially in hydrostatic equilibrium, such as the isothermal Bonnor-Ebert sphere (e.g., Foster \\& Chevalier 1993), tend to have flat inner density distributions surrounded by an outer envelope of roughly $\\rho \\propto r^{-2}$.  In addition, Henriksen et al. emphasized the observational evidence for flattened central regions in pre-stellar cores mentioned in \\S 4.1. When the initially flat inner regions of such a cloud fall in, high mass infall rates are produced (see also Foster \\& Chevalier 1993); material from the outer envelope falls in later at slower rates. This leads to time-dependent infall rates, with the highest rates occuring during the earliest phases of collapse, identified as the Class 0 phase.  This picture, however, does not explain why Class 0 sources tend to be found in dense star forming regions. Moreover, while the general picture of decreasing infall rate with increasing age is quite plausible, it is not clear whether such a Class 0 phase would dominate protostellar infall, because the amount of mass in the flattened inner portions of the initial cloud is generally smaller than the mass in the envelope, implying that most of the protostellar mass must be added during the envelope infall (Class I) phase.  Henriksen et al. (1997) show that, for a flat region of a cloud to contain half of the total cloud mass, the $\\rho \\propto r^{-2}$ envelope can only extend out a factor of 1.6 in radius beyond the flat region --  essentially, the cloud is almost entirely at constant density, with only a small outer envelope region, inconsistent with observations of typical pre-stellar cores. Henriksen et al. suggest that clouds with steeper edges, $\\rho \\propto r^{-3}$, could alleviate this problem while maintaining the required large factor $\\sim 10$ between Class I and Class 0 lifetimes; but they do not show that the very slow infall implied by these models for the Class I phase is consistent with SED models.  Alternatively, one might suppose that the high environmental densities characterizing the birthsites of Class 0 sources {\\it guarantee that envelope masses will be very large on small scales.} In other words, the Class 0 sources might be the Class I sources of dense star-forming regions.  As a specific example, consider two protostellar clouds of the same mass but formed in regions of differing average density. The characteristic infall rates will scale roughly as the cloud mass divided by the free fall time, which in turn scales roughly as the inverse square root of the mean density; thus, $\\md \\propto  M/<\\rho>^{-1/2} \\propto <\\rho>^{1/2} \\propto R^{-3/2}$, where $R$ is the outer radius of the cloud. (Let us fix the central star mass in order to discuss the difference in infall rates.) If we assume that the Class 0 source has an infall rate of $10^{-4} \\msunyr$, and a Class I source has an infall rate of $4 \\times 10^{-6} \\msunyr$, then if the two protostellar clouds originally had the same mass, the factor of 25 in $\\md$ implies that the Class I source would have to be larger in radius by a factor of 8.5.  In other words, the mass within a 2000 AU (12 arcsec at 160 pc) radius in a Class 0 source should be compared to the mass within a radius of 16,000 AU ($\\sim~100$~arcsec) in a Class I source to properly assess the relative evolutionary states of these objects.  The molecular surveys of Myers and collaborators (e.g., Benson \\& Myers 1989; Zhou et al. 1989) have indicated that many cores, such as those in Taurus, have substantial spatial extensions. For instance, even the lowest-mass ammonia cores in the survey of Benson \\& Myers (1989) have $0.8 - 0.6 \\msun$ of material, comparable to that of the Class 0 sources, but distributed within structures of (half-power diameters) $\\sim 10,000$~AU $\\sim 0.05$~pc rather than 2000~AU~$= 0.01$~pc.  In many other cases the ammonia core mass is considerably larger than $1 \\msun$ and spread over larger scales. Whether a substantial portion of the material in a typical Taurus molecular cloud core actually is incorporated into a particular protostellar source is not clear --or whether it makes only one star.  However, the same could be said of Class 0 sources.  Neglect of the possible importance of extended material could bias the comparison between the two types of sources.  Bontemps et al. (1996) attempted a systematic assessment of Class 0 and Class I source masses from measurements of 1.3mm continuum emission. In computing dust masses they adopted a ``typical'' source diameter of 1 arcmin over which to integrate the emission.  In doing so, Bontemps et al. recognized that they might be measuring only the ``inner'' envelope mass of ``isolated'' protostars such as those in Taurus.  Moreover, if our suggestion of the presence of significant ``external'' material in VLA 1623 is correct, then Class 0 mass estimates might be biased by the inclusion of mass that may not end up on the protostar, especially if integrated over a radius of 30 arcsec $\\sim 4800$~AU, rather than the 2000 AU radius adopted in the previous sections. Given the large differences in densities observed in star-forming regions, the systematic biases in mass estimates from adopting uniform source sizes can be of an order of magnitude, which could easily affect assessments of the relative evolutionary states of Class 0 and Class I sources.  Unfortunately, the observational problem of deciding where a protostellar cloud ``stops'' is likely to be quite difficult.  Yet, without consideration of this potential bias, it seems premature to conclude that all Class 0 sources must be in a radically different phase of evolution than Class I sources.  In Taurus, the source statistics of Class I sources suggest a lifetime of $1 - 2 \\times 10^5$~yr (Kenyon et al. 1990, 1994). If one adopts the infall rates of $\\sim 4 \\times 10^{-6} \\msunyr$ typically estimated from SED fitting (ALS; KCH), then one would expect that approximately $0.4 - 0.8 \\msun$ is added to protostars during the Class I phase - which is comparable to the {\\it total} mass accreted onto the average Taurus T Tauri star. It is difficult to see why the Class I period is not an important part of the protostellar phase in Taurus, unless somehow infall rates have been overestimated by an order of magnitude. If, on the other hand, Class I sources {\\it are} the protostars of low-density regions like Taurus, then there is no discrepancy to explain.  The two borderline Class 0 objects in Taurus (L1551 IRS 5 and L1527; Chen et al. 1995; or L1527 and IRAM 04191+1522; Andr\\'e et al. 1999) could be among the youngest protostellar sources in Taurus (Ladd et al. 1998); however, they might also be objects formed from particularly dense subregions within Taurus.  To summarize, we agree with AWB and AWB2 that the Class 0 sources are very young in an absolute sense because their envelopes must collapse so rapidly. However, we suggest that the high infall rates of Class 0 sources can result also from the high densities of their initial pre-stellar cloud cores rather than from always being the very first phases of collapse. Whether Class 0 sources are systematically younger than the Class I sources in a relative sense, i.e. have a much smaller fraction of their total mass accumulated into a central protostar, is far from clear given the observational biases and uncertainties in defining envelope limits. Some Class 0 sources may also be young in this relative sense, but it is difficult to support the conclusion that they are all the ``youngest protostars''.  \\subsection{Dense environments and the standard picture}  Infall rates on the order of $10^{-4}\\msun \\mathrm{yr^{-1}}$ seem to be needed to account for Class 0 sources. This infall rate is much higher than the typical infall rates $\\sim 4 \\times 10^{-6}\\msun \\mathrm{yr^{-1}}$ for Class I sources in Taurus-Auriga (KCH). In the standard singular isothermal sphere model, the infall rate is $\\md \\sim c_s^3 /G$; thus the mean Taurus infall rate requires a sound speed $c_s \\sim 0.25 \\kms$ or a gas temperature of $18$~K assuming a mean molecular weight of 2.3,  only slightly larger than typical temperatures found in the region.  In contrast, to obtain infall rates of $\\sim 10^{-4}\\msun \\mathrm{yr^{-1}}$ in a cloud supported only by thermal gas pressure would require unrealistically high cloud temperatures of about 150K.  Thus, some other source of support must be involved if the cloud is initially in hydrostatic equilibrium.  These considerations emphasize the importance of the density of the protostellar environment to an understanding of Class 0 sources.  The related question is how do such dense regions of material arise? If such regions are in hydrostatic equilibrium, turbulent motions undoubtedly play a role in supporting denser clouds (e.g., Lizano \\& Shu 1989; Fuller \\& Myers 1993). On the other hand, since the sources of turbulence are presumably non-isotropic and inhomogeneously distributed (i.e., winds, magnetic waves excited by moving dense gas clumps, etc.) arranging static situations seems very difficult.  In addition, supersonic magnetic turbulence may decay rapidly (Stone, Ostriker, \\& Gammie 1998). An alternative view is that dense streams of gas in the ISM, however produced, collide and dissipate energy in a shock (e.g., Ballesteros-Paredes, V\\'azquez-Semadeni \\& Scalo 1999), resulting in dense regions which then undergo gravitational collapse.  In such a general picture, there is no obvious reason why densities of protostellar regions need be restricted in any way to levels supportable by thermal gas pressure.  In this way protostars with dense envelopes and rapid infall rates could be produced without reference to the usual constraints of thermal support (ALS).  The Class 0 sources, or at least some of them, could be telling us more about protostellar formation in extreme environments than about the very earliest phases of protostellar collapse.  A related question is, why did the standard model of singular isothermal sphere collapse work so well in Taurus?  We suggest that this is simply due to initial conditions.  In the absence of any complicating factors --magnetic field support, turbulent motions, etc. -- thermal gas pressure provides an absolute lower limit to the forces resisting gravitational collapse, especially as cosmic ray heating seems to maintain cloud temperatures at a minimum value of $\\sim 10$~K in the absence of any other heat sources (Goldsmith 1988).  Absent substantial turbulent support, or high-density flows, Taurus may only represent a limiting case of star formation."
    },
    "0010/astro-ph0010267_arXiv.txt": {
        "abstract": "In this letter we present first results from  spectroscopic observations of Diffuse Ionized Gas (DIG) in the halo of NGC\\,5775  obtained with FORS1$^{1}$ attached to UT1/Antu of the Very Large Telescope (VLT). At our slit position perpendicular to the disk (41\\arcsec \\ SE of the nucleus) the  emission of [\\ion{N}{ii}]$\\lambda$6583, [\\ion{O}{iii}]$\\lambda$5007, and [\\ion{O}{ii}]$\\lambda$3727 is detected out to 9\\,kpc into the halo, allowing possible ionization mechanisms of the DIG to be examined. Photoionization models which assume a dilute radiation field are able to fit the data for the disk, but they cannot account for the line ratios measured in the halo (e.g., [\\ion{O}{i}]/${\\rm H\\alpha}$ or [\\ion{O}{ii}]/${\\rm H\\alpha}$). In particular they fail to predict the observed increase of [\\ion{O}{iii}]$\\lambda$5007/${\\rm H\\alpha}$ with increasing $|z|$. The most striking result concerns the kinematics of the halo gas. Velocities at high galactic latitudes drop from the midplane value to reach  the systemic velocity at $z \\approx$\\ 9\\,kpc. An analysis of VLA archive data of the polarized radio-continuum emission at 4.86\\,GHz and 1.49\\,GHz reveals that magnetic fields in the halo have a strong component perpendicular to the disk and are aligned with the H$\\alpha$ and radio-continuum spurs in the halo. This can result either from a strong wind action or, more likely, from the generation of dipolar magnetic fields. We briefly discuss the interrelation of the magnetic field structure and gas dynamics, in particular the role of magnetic fields in gas outflows, as well as the possible heating of DIG by magnetic reconnection. ",
        "introduction": "There is increasing evidence that halos of normal spiral galaxies contain a complex multiphase interstellar medium (ISM) (for a review see, e.g., Dahlem \\cite{dahlemr}). Besides a hot component as observed by X-ray emission ($>$10$^6$\\,K) or UV absorption ($\\sim$0.5$\\times$10$^5$\\,K), Diffuse Ionized Gas (DIG) constitutes another important phase of typically 10$^4$\\,K (Dettmar \\cite{dettmar92}). This component is easily observed by its $\\ion{H}{ii}$ region like line emission and is found to be widespread in galaxies with actively star forming disks (Rand \\cite{rand96}, Dettmar \\cite{dettmar99}, Rossa \\& Dettmar \\cite{rossa}).  In the last couple of years two important problems have been discussed with regard to the physics of this interstellar gas phase. The observed line ratios are not explained in a straight forward way by pure photoionization models and require  additional heating mechansims of the halo gas (e.g., Golla et al. \\cite{gollaetal}, Rand \\cite{rand97}, T\\\"ullmann\\,\\&\\,Dettmar \\cite{tude}) while first kinematic observations hint at a slower rotation of halo gas if compared to the underlying disk (Rand \\cite{rand98}). Both kinds of observations could help to constrain models of the origin of halo gas which in turn are important for the understanding of galaxy evolution since large scale mass flows are of influence for star formation and chemical evolution.  Only for a few objects, such as NGC\\,891 (Dahlem et al. \\cite{dahlem94}), NGC\\,4631 (Golla\\,\\&\\,Hummel \\cite{gh4631}), and NGC\\,5775, the various components of the halo ISM are observed with sufficient sensitivity and resolution to study possible physical processes at their interfaces such as turbulent mixing or thermal conduction. For such processes yet another ISM component can be of importance, namely the non-thermal component consisting of magnetic fields and  cosmic rays as observed by its radio-continuum radiation. Any large scale gas motion and in particular its $z$-component will also be of importance for dynamo theories which try to explain the large scale distribution of magnetic fields in galaxies. However, various approaches to describe galactic dynamos are still very controversial and more detailed information on galactic gas flows in halos as well as magnetic field structures could help constraining such models.  For all these questions, the case of the Sc edge-on galaxy NGC\\,5775 has received some attention recently, since marked radio-continuum (Duric et al. \\cite{duric}) and $\\ion{H}{i}$ (Irwin \\cite{irwin}) halo components are accompanied by a very extended DIG halo (Dettmar \\cite{dettmar92},  Collins et al. \\cite{coletal}) for which a strong decline of the rotational velocity with height above the plane ($z$) has been found (Rand \\cite{rand00}).  In this letter we report on first spectroscopic results for the DIG halo of NGC\\,5775 with the VLT, confirming the strong $z$-dependence of the halo rotation presented by Rand (\\cite{rand00}), along with a first detailed determination of the magnetic field structure in the halo from VLA radio-continuum observations. ",
        "conclusions": "The most important results can be summarized as follows: \\begin{itemize} \\item emission lines of DIG can be traced out to 9\\,kpc in the halo of NGC\\,5775 \\item the rotational velocity of the DIG drops from the rotational value at the midplane and approaches the systemic velocity at distances $z>$\\,6.0\\,kpc \\item pure photoionization models are unable to reproduce the extreme line ratios of [\\ion{O}{i}]/${\\rm H\\alpha}$, \\ion{He}{i}/${\\rm H\\alpha}$, and [\\ion{O}{iii}]/${\\rm H\\alpha}$ as measured for the halo \\item a high degree of polarization and a strong vertical component of the magnetic field are observed in the halo. We found some, yet weak evidence for the dipolar magnetic field \\item a correlation exists between extraplanar DIG and magnetic fields, evidenced by the alignment of magnetic B-vectors with prominent halo features in the DIG and radio-continuum distribution \\item we suggest that reconnection processes play a substantial role in both heating the DIG structures in the halo and building up regular fields away from the disk plane  \\end{itemize}"
    },
    "0010/hep-ph0010197_arXiv.txt": {
        "abstract": "We study constraints on the natural inflation model from the cosmic microwave background radiation (CMBR).  Inflaton $\\phi$ for the natural inflation has a potential of the form $V=\\Lambda^4[1-\\cos(\\phi/\\sqrt{2}f_\\phi)]$, which is parametrized by two parameters $f_\\phi$ and $\\Lambda$.  Various cosmological quantities, like the primordial curvature perturbation and the CMBR anisotropy, are determined as functions of these two parameters. Using recent observations of the CMBR anisotropy by BOOMERANG and MAXIMA (as well as those from COBE), constraints on the parameters $f_\\phi$ and $\\Lambda$ are derived.  The model with $f_\\phi$ lower than $8.5\\times 10^{18}$ ($5.4\\times 10^{18}$, $4.5\\times 10^{18}$) GeV predicts a power spectrum with index $n_{\\rm S}$ smaller than 0.95 (0.9, 0.85) which suppresses the CMBR anisotropy for smaller angular scale.  With such a small $n_{\\rm S}$, height of the second acoustic peak can become significantly lower than the case of the scale-invariant Harrison-Zeldovich spectrum. ",
        "introduction": " ",
        "conclusions": ""
    },
    "0010/astro-ph0010141_arXiv.txt": {
        "abstract": "We present photometric and spectroscopic measurements of the galaxy populations in clusters from the CNOC1 sample of rich, X-ray luminous clusters at $0.18 < z < 0.55$. A classical measure of the galaxy blue fraction for spectroscopically confirmed cluster members shows a significant Butcher-Oemler effect for the sample, but only when radii larger than 0.5$r_{200}$  are considered. We perform a principal component analysis of galaxy spectra  to divide the total cluster light into contributions from stellar populations of different ages. Composite radial distributions of different stellar populations show strong gradients as a function of cluster-centric radius. The composite population is dominated by evolved populations in the core, and gradually changes at radii greater than the virial radius to one which is similar to coeval field galaxies. We do not see evidence at any radius within the clusters for an excess of star formation over that seen in the coeval field.  Within this redshift range, significant evolution in the fractional population gradient is seen. Both low and high redshift clusters have similar populations in the cluster cores, but higher redshift clusters have steeper gradients and more star forming galaxies at radii  outside of the core region~-- in effect, a restatement of the Butcher-Oemler effect. Luminosity density profiles are consistent with a scenario where this phenomenon is due to a decline over time in the infall rate of field galaxies into clusters. Depending on how long galaxies reside in clusters before their star formation rates are diminished, this suggests a decrease in the infall into clusters of a factor of $\\sim 3$  between $z > 0.8$ and  $z \\sim 0.5$. We also discuss alternative scenarios for the evolution of cluster populations. ",
        "introduction": "The evolving populations in galaxy clusters offer a unique opportunity to directly observe galaxy evolution, and particularly the effects of environment on star forming galaxies. Present-day rich clusters have strikingly different populations from galaxies in poorer environments, suggesting that some mechanism is at work on the cluster population which is absent in the lives of normal field galaxies. The first clues towards understanding the evolution of the cluster population came more than 20 years ago, with the discovery that higher redshift clusters have a larger fraction of blue galaxies than those at the present epoch: the Butcher-Oemler effect (Butcher \\& Oemler 1978, 1984). These purely photometric results were confirmed spectroscopically (e.g., Dressler \\& Gunn 1982, 1992; Lavery \\& Henry 1986; Fabricant, McClintock \\& Bautz 1991) and populations of both star forming and recently post-star formation galaxies in moderate redshift clusters have been identified (i.e., the Balmer-strong H-$\\delta$ or K$+$A galaxies; Dressler \\& Gunn 1982; Couch \\& Sharples 1987; Poggianti \\etal 1999). Many subsequent investigations have focused on the details of how these galaxies are transformed into the population seen in clusters today  (Couch \\& Sharples 1987; Barger \\etal 1996; Poggianti \\etal 1999; Balogh \\etal 1999, B99 hereafter). The emerging picture is that there may be a population of galaxies which were formed very early in the cluster's history, corresponding to the ellipticals often seen in cluster cores (e.g., Bower, Lucey \\& Ellis 1992; Ellis \\etal  1997; Gladders et al. 1998; Kelson \\etal 2000; van Dokkum \\etal 1998). Subsequent generations of infalling field galaxies have had their star formation disrupted, possibly with an associated starburst. As this transformation progresses, these galaxies might be identified with normal-looking spirals, then galaxies with strong Balmer absorption spectra, and finally S0 galaxies which have retained some of their disk structure but have ceased active star formation (Dressler \\etal 1997).  Despite these advances, a number of questions remain about the evolution of cluster galaxies and the Butcher-Oemler phenomenon. Recently, there have been several challenges to the existence of the Butcher-Oemler effect as a universal pattern which equally affects all classes of clusters (Andreon 1998; Andreon \\& Ettori 1999; Smail \\etal 1998; Jones, Smail \\& Couch  2000). The original distant cluster samples which were investigated are a heterogeneous mixture of clusters found by optical signatures or by their association with a strong active galactic nucleus. In many cases, cluster masses or richnesses are still not well determined due to the difficulty of obtaining robust velocity dispersions, and there is a possibility of significant substructure in some objects. More recently, X-ray selected clusters, which in part overlap with optically-selected clusters but on average may be more massive or more relaxed, have been investigated. In several cases (Smail \\etal 1998; B99), the incidence of blue or post-star forming galaxies has been found to be significantly less than that of previous samples. Thus, a worrisome question remains as to how the observed galaxy populations relate to the more global properties of the clusters, such as mass and dynamical state, as well as the cluster sample selection method.  Even if clusters do contain a higher fraction of star forming galaxies at higher redshifts, a second fundamental question is whether this blue population represents a true increase in the average star formation in these galaxies over the properties of galaxies in the coeval field. A significant fraction of galaxies in some spectroscopic surveys have been found to be starbursts or have Balmer lines strong enough to signify a previous starburst encompassing 10-30\\% of the galaxy mass (Lavery \\& Henry 1988; Barger \\etal 1996; Poggianti \\etal 1999). This suggests that temporary episodes of excess star formation are associated with a galaxy's tenure in the cluster. However, Balogh \\etal (1997, B99) found no evidence for excess star formation in a sample of luminous X-ray selected clusters, as the generally smaller numbers of starburst and post-starburst galaxies could be explained by galaxies infalling from the field. Thus, they conclude that the cluster environment works only in the direction of suppressing star formation. The differences in these results may be partially ascribed to differences in the  selection technique for the galaxies observed and identified in both cluster and field; but some of the discrepancy may come from true differences in the cluster samples.  Here we describe an investigation of the relationship between galaxy evolution and cluster structure, based on a well-defined  and homogeneous sample of clusters at $0.2 < z < 0.6$  and coeval field galaxy measurements. Photometry and spectroscopy from the Canadian Network for Observational Cosmology Cluster Redshift Survey (CNOC1; Yee, Ellingson \\& Carlberg  1996) were  used to construct a sample of cluster galaxies and the properties of their stellar population. Field galaxy measurements from the CNOC2 Field Galaxy Survey (Yee \\etal 2000) were used to accurately determine the properties of field galaxies at these redshifts. We present both a classical measure of the cluster blue fraction determined from spectroscopically confirmed cluster members and a spectral principal component analysis (PCA) which delineates the relative fractions of starlight from different stellar populations. In an approach complementary to that of looking at individual galaxy properties in detail, we focus on building smooth composite spatial distributions of the various stellar populations. With these distributions, it is possible to explore the relationship between the evolution of galaxy populations and the growth and evolution of the cluster.  In \\S2 of this paper we present the data and in \\S3 we discuss the blue fraction measurements.  We describe the principal component analysis method and tests of its accuracy from galaxies of known spectral type in \\S4. In \\S5 we present the results based on composite radial distributions of the average  PCA components. We discuss the results and their implications for scenarios of galaxy population evolution in rich clusters in \\S6. Our conclusions are summarized in \\S7. Throughout this paper $q_0$=0.1 and $H_0$=100 km s$^{-1}$ Mpc$^{-1}$ are used. ",
        "conclusions": "We have  investigated the Butcher-Oemler effect in a sample of 15 rich galaxy clusters at $0.18 < z < 0.55$ from the CNOC1 cluster survey. A classical measurement of the cluster blue fraction, based on spectroscopically confirmed cluster members, shows a significant Butcher-Oemler effect, in that high redshift clusters have a larger fraction of star forming galaxies. These galaxies are found primarily outside of the cluster cores, and the Butcher-Oemler effect virtually disappears when a counting radius of only 0.5$r_{200}$ is used.  We have developed an alternative method of measuring stellar populations in galaxies via a principal component analysis (PCA) of galaxy spectra. Basis vectors defined from the low-redshift LCRS spectral database were used to decompose cluster galaxy spectra into three components, corresponding to an old, passively evolving population, a young population matched to the coeval field, and a population of intermediate age, matched to a 0.5 Gyr post-star formation population. These components are shown to be robust measurements of stellar populations by comparison with low redshift galaxy spectra and classical line indices. However, unlike line indices, PCA components directly measure the fraction of light emitted by a given population, allowing for straightforward construction of population fractions and spatial distributions.  Composite population gradients show a smooth transition from the infalling field galaxy population to the older populations seen in the inner regions. The gradients are corrected for the evolution of the field population and cluster richness and show no evidence for an excess of star formation in the cluster; the cluster environment therefore appears to have the monotonic effect of decreasing star formation in infalling galaxies. However, there is evidence for a small excess of intermediate age stars, mostly located in galaxies with mixed stellar populations. This suggests that star formation in these galaxies has been truncated gradually, without an accompanying strong starburst.  Evolution in the cluster galaxy population between $z=0.5$ and 0.2 is manifest in a flattening of the fractional population gradient at later epochs, while the cluster cores remain dominated by an old, red population. Luminosity surface brightness distributions indicate that the old stellar population has a centrally concentrated, unchanging spatial extent as a function of $r/r_{200}$ . Younger galaxy populations have a more extended distribution that also retains its general shape but appears to be decreasing by about a factor of three in amplitude between $z=0.5$ and $z=0.2$. This phenomenon is most consistent with scenarios where the mechanism which truncates star formation in individual galaxies remains constant, but the cluster population evolution is driven by a declining rate of infall into the clusters. Depending on the length of time that star formation in galaxies can survive the cluster environment, this may indicate a change in the infall rates into clusters between $z > 0.8$ and $z \\sim 0.5$; if galaxies reside in the cluster for a significant interval before ceasing star formation, the evolution will correspond to even earlier epochs. We note that the stellar mass fraction of young populations in these clusters is on the order of a only few percent, despite the fact that it requires several Gyr to transform stellar populations from young to old populations. This also suggests that the Butcher-Oemler effect as seen at intermediate redshifts is a lingering echo of more rapid cluster formation which occurred much earlier, in qualitative  agreement with low-density cosmological models. Additional N-body and analytical modeling is necessary to firmly link the observed evolution in population gradients to models of cluster formation and growth.  These results indicate that galaxy populations provide a very sensitive probe of cluster growth, and that  the construction of spatial distributions for different galaxy populations may illuminate both cluster and galaxy evolution. Extending similar studies to both higher and lower redshifts is critical towards building a consistent picture of cluster evolution. However, the best studied samples at intermediate and high redshifts may not be representative of the cluster population, as both X-ray and optically selected samples may introduce systematic biases  in the dynamical state and formation history of the clusters. As the populations seen in individual clusters may be strongly affected by the recent merging history of the cluster, it is important that a wide variety of cluster types be sampled in order to build a truly representative sample. A new generation of high redshift cluster samples based on algorithms which are sensitive to low surface brightness and newly merging cluster structures is necessary to expand and clarify our picture of the formation of galaxy clusters and the effects of environment on galaxy evolution."
    },
    "0010/astro-ph0010188_arXiv.txt": {
        "abstract": "An optical spectropolarimetric study has shown that the detectability of polarised broad H$\\alpha$ in Seyfert 2 galaxies is correlated to the IRAS $f_{60}/f_{25}$ flux ratio where only those Seyfert 2s with ``warm\" IRAS colours show polarised broad line emission. It was suggested that those Seyfert 2s with ``cool\" IRAS colours have highly inclined tori which obscure the broad line scattering screen.  I present here hard X-ray observations inconsistent with this picture showing that the derived column densities of warm and cool Seyfert 2 galaxies are statistically the same. I classify the Bright Galaxy Sample to produce a non-Seyfert comparison. The analysis of the properties of these galaxies with the Seyfert 2s suggest that the IRAS $f_{60}/f_{25}$ flux ratio is consistent with implying the relative strength of galactic to Seyfert emission. I show that this new picture can account for the absence of polarised broad H$\\alpha$ in the cool Seyfert 2s. ",
        "introduction": "The unified model for Seyfert galaxies proposes that all types of Seyfert galaxy are fundamentally the same, however, the presence of an optically thick structure obscures the broad line region (BLR) in many systems. In this paper it is assumed that, in the majority of Seyfert 2s, this structure is a dusty molecular ''torus\" although other galactic structures (e.g.\\ dust lanes/starbursts, see Malkan, Gorjian and Tam, 1998) can perform the same role. In this scenario the classification of a Seyfert 1 or Seyfert 2 galaxy (Seyfert 1--broad permitted lines, Seyfert 2--narrow permitted lines)  depends on the inclination of the torus to the line of sight (Antonucci, 1993). Probably the most convincing evidence for this model comes from optical spectropolarimetry. Using this technique, the scattered emission from the BLR of many Seyfert 2s is revealed in the form of broad lines in the polarised flux (e.g.\\ Antonucci and Miller, 1985, Young et al, 1996, Heisler, Lumsden and Bailey, 1997).  In this unified picture the high energy central source emission (optical to X-ray continuum) is absorbed by the dust within the torus which re-emits this energy at infrared (IR) wavelengths. Independent strong support has been given by hard X-ray (HX, 2 to 10 keV), near-IR and mid-IR observations (e.g.\\ Turner et al, 1997, Risaliti, Maiolino and Salvati, 1999, Alonso-Herrero, Ward, Kotilainen, 1997 and Clavel et al, 2000) showing that Seyfert 2s are generally characterised by strong absorption whilst Seyfert 1 galaxies are relatively unabsorbed.  Heisler, Lumsden and Bailey (1997, hereafter HLB) performed an optical spectropolarimetric study of a well defined and statistically complete IRAS 60$\\mu$m selected Seyfert 2 sample to determine the statistical detectability of polarised broad lines.  The objects were selected at 60$\\mu$m to reduce the possibility of biasing due to torus inclination/extinction effects and all objects were observed to the same signal to noise to ensure similar detection thresholds. In this study a striking relationship between the detectability of polarised broad H$\\alpha$ and the IRAS $f_{60}/f_{25}$ flux ratio was found where only those galaxies with warm IRAS colours ($f_{60}/f_{25}<$4.0) showed a hidden broad line region (HBLR). Both Seyfert 2 galaxy types were found to be well matched in terms of redshift, overall polarisation and detection rate of compact nuclear radio emission. Therefore, without any apparent contradictory evidence, HLB suggested that the IRAS $f_{60}/f_{25}$ ratio provides a measure of the inclination of the torus to the line of sight: in a cool Seyfert 2 the torus is so highly inclined that even the broad line scattering screen is obscured. I present here HX evidence that suggests this picture is incorrect and provide a new view that is consistent with other observations. ",
        "conclusions": "HLB showed a dichotomy in the detectability of polarised broad H$\\alpha$ in Seyfert 2 galaxies where only those objects with warm IRAS colours showed polarised broad line emission. They suggested that those Seyfert 2s with cool IRAS colours had highly inclined tori which obscured the broad line scattering screen. I have presented evidence against this picture and suggested that the IR dichotomy is due to the relative strength of galactic to Seyfert emission. The main results of this study are:  (i) the hard X-ray derived column densities of the warm and cool Seyfert 2s are statistically the same and therefore the IRAS $f_{60}/f_{25}$ flux ratio cannot imply the inclination of the torus in the standard unified model.  (ii) the cool Seyfert 2s have IRAS colours and PAH emission similar to HII galaxies and optical nuclear emission line ratios consistent with a mixture of Seyfert and galactic activity. As the hard X-ray observations suggest that both Seyfert 2 galaxy types have similar nuclear properties, the differences in the IRAS $f_{60}/f_{25}$ flux ratios are most probably due to a weaker Seyfert component and a more dominant galactic component in the cool Seyfert 2s.  (iii) the difference in the detectability of polarised broad H$\\alpha$ is most probably due to a larger component of optical starlight in the cool Seyfert 2s. The reason for the lack of difference in observed polarisation between both Seyfert 2 types is probably due to a larger galactic polarisation component in the cool galaxies caused by their larger galactic extinction."
    },
    "0010/astro-ph0010377_arXiv.txt": {
        "abstract": "We report a strong correlation between the [CII] 158 $\\mum$ cooling line and the mid-infrared flux in the 5-10 $\\mum$ range in a wide variety of star-forming galaxies. The mid-infrared flux is dominated by Aromatic Feature Emission (AFE), which is thought to arise from large polycyclic aromatic hydrocarbon molecules or ``PAHs'' and is generally associated with the smallest interstellar grains.  The [CII] line is the dominant gas coolant in most regions of atomic interstellar gas, and therefore reflects the heating input to the gas.  The ratio of these two quantities, [CII]/AFE, remains nearly constant around 1.5\\% against variations in the ratio of the \\IRAS\\ 60 $\\mu$m band flux to the 100 $\\mu$m band flux, R(60/100).  This is in contrast to the drop in the [CII]/FIR ratio with increasing R(60/100), which signals higher dust temperatures and more intense radiation fields.  We interpret the stable [CII]/AFE ratio as evidence that gas heating is dominated by the PAHs or small grains, which are also AFE carriers, over a wide range of conditions.  The trend of decreasing [CII]/FIR and AFE/FIR with increasing radiation field suggests a decrease in the importance of PAHs or small grains relative to large grains both in gas heating and in dust cooling.  We summarize the observed trends and suggest two plausible scenarios. ",
        "introduction": " ",
        "conclusions": ""
    },
    "0010/astro-ph0010007_arXiv.txt": {
        "abstract": "We present the first results from a major Hubble Space Telescope programme designed to investigate the cosmological evolution of quasar host galaxies from $z \\simeq 2$ to the present day. Here we describe $J$ and $H$-band NICMOS imaging of two quasar samples at redshifts of 0.9 and 1.9 respectively. Each sample contains equal numbers of radio-loud and radio-quiet quasars, selected to lie within the same narrow range of optical absolute magnitude ($-24\\geq M_{V} \\geq -25$).  Filter and target selection were designed to ensure that at each redshift the images sample the same part of the object's rest-frame spectrum, longwards of 4000\\AA~ where starlight from the host galaxy is relatively prominent, but avoiding potential contamination by [O{\\sc iii}]$\\lambda5007$ and H$\\alpha$ emission lines.  At $z\\simeq1$ we have been able to establish host-galaxy luminosities and scalelengths with sufficient accuracy to demonstrate that the hosts of both radio-loud and radio-quiet quasars lie on the same Kormendy relation described by 3CR radio galaxies at comparable redshift (McLure \\& Dunlop 2000).  Taken at face value the gap between the host luminosities of radio-loud and radio-quiet objects appears to have widened from only $\\simeq 0.4$ mag. at $z \\simeq 0.2$ (Dunlop et al. 2001) to $\\simeq 1$ mag. at $z \\simeq 1$, a difference that cannot be due to emission-line contamination given the design of our study. However, within current uncertainties, simple passive stellar evolution is sufficient to link these galaxies with the elliptical hosts of low-redshift quasars of comparable nuclear output, implying that the hosts are virtually fully assembled by $z\\sim1$.  At $z\\simeq2$ the hosts have proved harder to characterise accurately, and for only two of the nine $z \\simeq 2$ quasars observed has it proved possible to properly constrain the scalelength of the host galaxy. However, the data are of sufficient quality to yield host-galaxy luminosities accurate to within a factor $\\simeq 2$. At this redshift the luminosity gap between radio-loud and radio-quiet quasars appears to have widened further to $\\simeq 1.5$ mag. Thus while the hosts of radio-loud quasars remain consistent with a formation epoch of $z > 3$, allowing for passive evolution implies that the hosts of radio-quiet quasars are $\\simeq 2-4$ times less massive at $z \\simeq 2$ than at $z \\simeq 0.2$.  If the relationship between black-hole and spheroid mass is unchanged out to redshift $z \\simeq 2$, then our results rule out any model of quasar evolution which involves a substantial component of luminosity evolution (e.g. Kauffmann \\& Haehnelt 2000). Rather, this study indicates that at $z \\simeq 2$ there is a substantial increase in the number density of active black holes, along with a moderate increase in the fueling efficiency of a typical observed quasar. The fact that this latter effect is not displayed by the radio-loud objects in our sample might be explained by a selection effect arising from the fact that powerful radio sources are only produced by the most massive black holes (Dunlop et al. 2001; McLure \\& Dunlop 2000b). ",
        "introduction": "Recent years have brought great advances in our understanding of the symbiotic relationship between active galactic nuclei (AGN) and the galaxies in which they occur. Some of the most impressive advances concern quasars - the most powerful known AGN -  as improvements in ground-based observing techniques and the advent of the Hubble Space Telescope (HST) have allowed the diffuse `fuzz' of the underlying host galaxy to be reliably separated from the wings of the bright non-stellar nuclear point spread function (PSF) for the first time.  The problems inherent in observing quasar host galaxies from the ground are too well-known to require detailed explanation but, considering the enormous difficulties involved in separating faint, diffuse galaxy light from the PSF of a bright quasar, it is perhaps surprising that so much progress has been made to date using ground-based techniques. Although such studies are effectively limited to $z \\leq 0.3$, at these low redshifts a combination of ground-based and HST programmes is beginning to yield a coherent picture of the properties of quasar hosts in the local universe.  However, the local universe is not the most representative region in which to study the quasar population: arguably the epoch of greatest importance to quasar research occurred at redshifts of $z\\sim2$ to 3, when quasars were $2 - 3$ orders of magnitude more numerous than they are today. Although simulations show that it is very difficult to derive the properties of quasar hosts at such high $z$ from the ground with any degree of confidence (Taylor et al. 1996), numerous attempts have been made with (not surprisingly) confusing results. In this paper we present the results of a study using the Near-Infrared Camera and Multi-Object Spectrometer (NICMOS) on HST to investigate how the luminosities, sizes and morphologies of the hosts of both radio-loud and radio-quiet quasars (RLQs \\& RQQs) have evolved from the `golden era' of quasar activity to the present day.  \\subsection{The host galaxies of low-redshift quasars}  At low redshifts, attention has been focused on determining the luminosities, scalelengths, morphologies and interaction histories of quasar host galaxies, and investigating the extent to which these properties are correlated to the optical and radio luminosity of the quasar (V\\'{e}ron-Cetty \\& Woltjer 1990; Dunlop et al. 1993; McLeod \\& Rieke 1994a,b; Bahcall, Kirhakos \\& Schneider 1995a,b,1996; Hutchings \\& Morris 1995; Disney et al. 1995; Taylor et al. 1996; Bahcall et al. 1997; Hooper, Impey \\& Foltz 1997; Boyce et al. 1998; Carballo et al. 1998; McLeod, Rieke \\& Storrie-Lombardi 1999; Schade, Boyle \\& Letawsky 2000; Hamilton, Casertano \\& Turnshek 2001; M\\'{a}rquez et al. 2001).  For example, it has long been known that low-luminosity AGN display marked preferences in terms of host type, with (radio-quiet) Seyferts favouring spiral hosts whilst (radio-loud) Radio Galaxies are exclusively associated with massive ellipticals. However, recent studies have demonstrated that this distinction breaks down at the higher redshifts and nuclear luminosities typical of radio-loud and radio-quiet quasars, and that powerful nuclear activity is predominantly associated with bulge-dominated galaxies, regardless of radio luminosity.   In our own HST study of low redshift ($0.1\\leq z\\leq0.25$), low luminosity ($-23\\leq M-{V} \\leq -26$) RLQs and RQQs (McLure et al. 1999; Dunlop et al. 2001) we found that for quasars brighter than $M_{R}\\sim -24$ the hosts are invariably massive elliptical galaxies with $L\\geq 2L^{\\star}$, with typical scalelengths of $\\sim 10$~kpc, and display a Kormendy relation identical to that of brightest cluster galaxies. This result is consistent with measurements of the massive dark objects in the nuclei of nearby inactive galaxies, which predict that only the largest spheroidal systems will harbour black holes of the requisite mass to produce luminous quasars (Kormendy \\& Richstone 1995; Magorrian et al. 1998; van der Marel 1999), and also with the discovery by McLeod \\& Rieke (1995) of a lower limit to the mass of a quasar host which is correlated with the luminosity of the quasar. The implication of these studies is that massive elliptical galaxies are the parent population of the quasar phenomenon. However, in a ground-based $K$-band study of luminous ($-25 \\geq M_{V} \\geq -27$) RQQs Percival et al. (2001) found evidence that, on scales much larger than those probed in optical HST images, some of the hosts are dominated by a disc component. Clearly the issue of host morphology has yet to be entirely resolved, but the finding that a quasar host requires at the very least a massive spheroidal component seems secure.  Spectral energy distributions (SEDs) also provide important information on the nature of the host galaxies.  McLure et al. (1999) and Dunlop et al. (2001) found that the $R-K$ colours of their quasar hosts were consistent with normal passively-evolving elliptical galaxies with ages of $\\sim 12$~Gyr, implying that the stellar populations formed at high redshift ($z\\sim3$). Optical off-nuclear spectroscopy of the same objects (Hughes et al. 2000; Nolan et al. 2001) confirms that, longwards of the 4000\\AA~break in their restframe spectra, these hosts are dominated by light from an old, well-established population of stars. Similar results are obtained for radio galaxies (e.g. de Vries et al. 2000).  \\subsection{High-redshift quasar hosts}  The fact that the host galaxies of low-redshift quasars have mature stellar populations and contain a massive bulge component implies that, like local inactive elliptical galaxies, they are arguably consistent with the products of successive merger events.  This raises important questions about the potential redshift dependence of the properties of quasar hosts.  The dramatic cosmological evolution exhibited by the quasar population itself, with a comoving number density which peaks at redshifts of 2-3 before undergoing a rapid decline to its present low value, has been known for many years (e.g. Boyle et al. 1988; Dunlop \\& Peacock 1990; Warren, Hewett \\& Osmer 1994) but the extent and form of any evolution affecting quasar hosts remain as largely unknown quantities. The issue has been given renewed urgency by continuing efforts to characterise the starformation history of the universe, for which the redshift range of $z=2-3$ also appears to correspond to an important epoch (Madau et al. 1996; Hughes et al. 1998; Steidel et al. 1999).  It is therefore important to determine how the luminosities, scalelengths, and the degree of morphological disturbance seen in quasar host galaxies varies with cosmological epoch between the peak of quasar activity at $z \\simeq 2$ and the present day, and whether the hosts of radio-loud and radio-quiet quasars differ in their luminosity or morphological evolution.  However, the difficulties involved in such studies are far more formidable than at low $z$ and this is reflected in the confusion which surrounds the interpretation of ground-based attempts to observe high-redshift quasar hosts.  Several groups have attempted such observations from the ground and have succeeded in detecting extensions around high-redshift quasars (Hintzen, Romanishin \\& Valdes 1991; Heckman et al. 1991; Lehnert et al. 1992; Aretxaga, Boyle \\& Terlevich 1995; R\\\"{o}nnback et al. 1996; Aretxaga, Terlevich \\& Boyle 1998). These studies suggest that the host galaxies of quasars at $z \\simeq 2$ are $\\sim 2.5$ to $3$ mag brighter than those of low-redshift quasars, a result which is consistent with common scenarios for elliptical galaxy evolution. However, the usefulness of any such direct comparison is confused by the fact that these high-redshift quasars are typically 5 magnitudes more luminous than the low-redshift objects studied to date. In addition, the observational situation is somewhat complex since other workers have failed to detect extended emission around high-redshift RQQs (Lowenthal et al. 1995). Hutchings (1995) also concludes that the hosts of RQQs are considerably fainter than those of RLQs at high redshift, a result which could however simply indicate that much of the light detected around high-redshift RLQs is not due to stars, but rather to processes associated with the extreme radio activity, as is found for high-$z$ radio galaxies (Tadhunter et al. 1992), and for low-$z$ RLQs (Stockton \\& MacKenty 1987). This emphasises the importance of avoiding emission lines and sampling the rest-frame spectrum longwards of the 4000\\AA\\ break.  The study described in this paper attempts to avoid the problems outlined above by observing test samples of quasars over a range of redshifts out to $z=2$ but limited to a narrow range of optical absolute magnitude ($-24 \\leq M_{V} \\leq -25$). Careful choice of filters ensures that our images always sample the same emission-line-free region of the object's restframe spectrum, thus avoiding any regions of extended [O{\\sc iii}]$\\lambda5007$ and H$\\alpha$ emission which may be associated with the active nucleus. Thus, at each redshift we are always observing objects of roughly the same intrinsic luminosity, and in the same region (approximately $V$-band) of their rest-frame spectrum.  The layout of this paper is as follows. In section 2 we provide details of how our quasar samples were selected, and how the HST NICMOS observations were designed and implemented. The process of data reduction is described in section 3, including the details of how we chose to tackle the particular problems associated with producing NICMOS images of sufficient quality for a study of this type. In section 4 we explain how host-galaxy parameters were extracted from the data via 2-D modelling, and summarize the basic results of the image analysis. Then in section 5 we explore the implications of combining these new results with the results of our Cycle 6 HST study of quasar hosts at much lower redshift ($z \\simeq 0.2$; McLure et al. 1999; Dunlop et al. 2001), and discuss our findings in the context of existing theoretical predictions. Finally, our results are summarized in section 6.  For ease of comparison with existing results/predictions we assume an Einstein-de Sitter cosmology with $H_{0}=50$~km~s$^{-1}$Mpc$^{-1}$ throughout most of this paper. However, in the penultimate section we focus on the implications of our results within the currently-favoured flat cosmological model with $\\Omega_{m} = 0.3$, $\\Omega_{\\Lambda} = 0.7$ and $H_{0}=65$~km~s$^{-1}$Mpc$^{-1}$.   \\begin{figure*} \\vspace{7.0cm} \\centering \\setlength{\\unitlength}{1mm} \\special{psfile=kukula_fig1.eps angle=0 hoffset=0 voffset=0 hscale=70 vscale=70} \\caption{Left: Absolute $V$ magnitude versus redshift for the RQQs (filled circles) and RLQs (open circles) in the current study. The shaded vertical bars represent the redshift regimes over which the indicated WFPC2 and NICMOS filters remain free from contamination by [O{\\sc iii}]$\\lambda5007$ and H$\\alpha \\lambda6563$ emission lines. Also shown are the quasars from our previous HST imaging study using WFPC2/F675W (McLure et al. 1999; Dunlop et al. 2001). This low-$z$ sample spans almost 3 magnitudes in optical luminosity at a given redshift ($z \\simeq 0.2$) whereas the three samples selected for the current study are confined to only 1 mag in luminosity but span the bulk of the history of the Universe.  Thus these complementary low-$z$ objects provide the baseline against which to measure any cosmological evolution in the high-$z$ samples. Right: Schematic showing typical rest-frame spectra of a quasar (upper panel) and an early-type galaxy (lower panel). The two vertical lines delineate the region of the host-galaxy spectrum which is targeted by the redshift/filter combinations used in the observations described in this paper. Note the avoidance of prominent quasar emission lines and the increase in galaxy flux density longwards of the break feature at 4000\\AA.} \\end{figure*}    \\begin{table*} \\centering \\begin{minipage}{140mm}  \\caption{Summary of the quasar samples observed in the current NICMOS study. RQQs are derived from the UVX survey by Boyle et al. (1990) (SGP) and the survey of Marshall et al. (1984) (BVF), and are confirmed as radio quiet ($P_{5GHz}<10^{24.5}$W Hz$^{-1}$ sr$^{-1}$) by the VLA FIRST \\& NVSS surveys. RLQs were selected from the V\\'{e}ron-Cetty \\& V\\'{e}ron (1993) quasar catalogue, with reported 5~GHz radio luminosities $P_{5GHz}>10^{25}$W Hz$^{-1}$ sr$^{-1}$ and steep radio spectra. $V$-band magnitudes are from various sources, and where necessary have been converted from $B$-band measurements. For the estimation of luminosities at both radio and optical wavelengths we have assumed a quasar spectrum of the form $f(\\nu) \\propto \\nu^{-0.5}$}  \\begin{tabular}{@{}lllllccl@{}} \\hline &       &\\multicolumn{2}{c}{J2000 Position} &     &      &  Observing & Time on \\\\ Name       &  Type & RA ($h~m~s$) & Dec $^{\\circ}~'~''$& $z$ & $V$  &  date (dd/mm/yy)& source (sec) \\\\ \\hline \\multicolumn{8}{c}{$J$-band (F110M) sample ($z\\sim1$)}\\\\ % SGP5:46    &  RQQ  & 00 52 22.76  & $-$27 30 02.8  & 0.955  & 18.88  & 25/08/97  &  $1\\times2048$ \\\\ SGP2:47    &  RQQ  & 00 53 02.24  & $-$29 12 55.5  & 0.830  & 17.17  & 14/08/97  &  $1\\times2048$ \\\\ BVF225     &  RQQ  & 13 04 10.47  & $+$35 36 50.8  & 0.910  & 18.54  & 16/02/98  &  $1\\times2048$ \\\\ BVF247     &  RQQ  & 13 05 05.04  & $+$35 51 20.6  & 0.890  & 18.73  & 15/02/98  &  $1\\times2048$ \\\\ BVF262     &  RQQ  & 13 05 30.91  & $+$35 17 13.5  & 0.970  & 18.57  & 18/02/98  &  $1\\times2048$ \\\\ PKS0440-00 &  RLQ  & 04 42 38.64  & $-$00 17 43.4  & 0.844  & 18.41  & 11/08/97  &  $1\\times2048$ \\\\ PKS0938+18 &  RLQ  & 09 41 23.17  & $+$18 21 06.0  & 0.943  & 18.49  & 18/02/98  &  $1\\times2048$ \\\\ 3C422      &  RLQ  & 20 47 10.39  & $-$02 36 22.5  & 0.942  & 18.69  & 03/11/97  &  $1\\times2048$ \\\\ MC2112+172 &  RLQ  & 21 14 56.68  & $+$17 29 22.7  & 0.878  & 17.89  & 30/10/97  &  $1\\times2048$ \\\\ 4C02.54    &  RLQ  & 22 09 32.82  & $+$02 18 40.9  & 0.976  & 18.19  & 05/11/97  &  $1\\times2048$ \\\\ SA107-626  &  PSF star & 15 40 05.31 & $-$00 17 29.2 & 0    & 12.77  & 30/07/97  &  $8\\times256$  \\\\ SA107-627  &  PSF star & 15 40 07.45 & $-$00 17 23.0 & 0    & 13.47  & 17/02/98  &  $8\\times256$  \\\\ % \\multicolumn{8}{c}{$H$-band (F165M) sample ($z\\sim2$)}\\\\ % SGP2:36    &  RQQ   & 00 51 14.32  & $-$29 05 19.7 & 1.756  & 19.62  & 20/11/97  &  $3\\times2048$ \\\\ SGP2:25    &  RQQ   & 00 52 07.60  & $-$29 17 50.2 & 1.868  & 19.55  & 28/08/97  &  $4\\times2048$ \\\\ SGP2:11    &  RQQ   & 00 52 38.47  & $-$28 51 12.9 & 1.976  & 19.77  & 02/09/97  &  $4\\times2048$ \\\\ SGP3:39    &  RQQ   & 00 55 43.41  & $-$28 24 09.7 & 1.964  & 19.65  & 06/09/97  &  $4\\times2048$ \\\\ SGP4:39    &  RQQ   & 00 59 08.88  & $-$27 51 24.7 & 1.716  & 19.64  & 30/08/97  &  $3\\times2048$ \\\\ 1148+56W1  &  RLQ   & 11 50 44.8   & $+$56 32 56.0 & 1.782  & 19.69  & 30/06/97  &  $4\\times2048$ \\\\ PKS1524-13 &  RLQ   & 15 26 59.44  & $-$13 51 01.3 & 1.687  & 19.69  & 31/07/97  &  $4\\times2048$ \\\\ B2-2156+29 &  RLQ   & 21 58 42.0   & $+$29 59 08.0 & 1.759  & 19.29  & 05/11/97  &  $3\\times2048$ \\\\ PKS2204-20 &  RLQ   & 22 07 33.94  & $-$20 38 34.9 & 1.923  & 19.49  & 04/09/97  &  $4\\times2048$ \\\\ 4C45.51    &  RLQ   & 23 54 22.27  & $+$45 53 05.2 & 1.992  & 19.79  & 14/11/97  &  $4\\times2048$ \\\\ SA107-626  &  PSF star & 15 40 05.31 & $-$00 17 29.2 & 0    & 12.77  & 17/02/98  &  $4\\times256$  \\\\ SA107-627  &  PSF star & 15 40 07.45 & $-$00 17 23.0 & 0    & 13.47  & 30/07/97  &  $4\\times256$  \\\\ \\hline  \\end{tabular} \\end{minipage} \\end{table*} ",
        "conclusions": "Despite the difficulty in determining morphologies and, at $z\\sim2$, scalelengths for the quasar hosts, the information obtained from the NICMOS images allows us to compare the galaxies with other types of active and inactive galaxies at the same redshifts and to investigate the link between quasar and galaxy evolution. We first discuss the implications of our relatively robust results at $z \\simeq 1$, before considering the implications of our $z \\simeq 2$ results in the context of theories of quasar/galaxy evolution.  \\subsection{Comparison of RLQ and RQQ hosts at {\\bf z$\\simeq$1}}  In our study of low-redshift ($z\\sim0.2$) radio-loud and radio-quiet quasars (McLure et al. 1999, Dunlop et al. 2001, Hughes et al. 2000, Nolan et al. 2001), for quasars with $M_{V} \\leq -24$ we found no statistically significant differences between the host galaxies of the two types of quasar. Both RLQs and RQQs seem to lie in large ($r_{1/2}\\sim 10$~kpc, luminous ($L > L^{\\star}$) elliptical galaxies with colours and ages consistent with those of inactive massive elliptical galaxies at the same redshift.  At $z\\sim 1$, although we are unable to determine the morphology of the galaxies with confidence, we find that once again the properties of the radio-loud and radio-quiet hosts are statistically consistent with one another, although there is some evidence of a trend towards a larger luminosity difference than was found at $z \\simeq 0.2$. The mean $J$-band magnitude for the RLQ hosts is $18.8\\pm0.3$ and that of the RQQs is $19.7\\pm0.3$, whilst the mean scalelengths are $12.4\\pm2.4$~kpc and $9.4\\pm3.2$~kpc respectively. A similar tendency towards somewhat less massive hosts for RQQs was also seen in the $z\\simeq0.2$ sample (McLure et al. 1999; Dunlop et al. 2001), although once again the difference was not statistically significant. McLure et al. speculate that this might indicate that the low-$z$ RQQs contain smaller black holes than their radio-loud counterparts, but are accreting gas with greater efficiency, to produce comparable luminosities.  Larger samples will be required at both redshifts in order to determine whether this apparently small but persistent difference is in fact real.   \\subsection{Comparison of quasar hosts and radio galaxies at z$\\simeq$1}  \\begin{figure} \\vspace{7.7cm} \\centering \\setlength{\\unitlength}{1mm} \\special{psfile=kukula_fig5.eps angle=0 hoffset=-0 voffset=-15 hscale=65 vscale=65} \\caption{The apparent $I$-band Kormendy relation described by the $z\\sim1$ quasars (triangles) imaged with NICMOS through the F110M ($\\simeq J$) filter. Conversion to $I$ magnitudes has been carried out assuming $I-J$ colours of 0.8, a typical value for elliptical galaxies at $z\\sim1$ (Fasano et al. 1998). Also shown are the $z\\sim0.8$ 3CR radio galaxies from the sample of Best, Longair \\& R\\\"{o}ttgering (1997; 1998) (crosses). These objects were imaged in $I$-band with WFPC2 on HST and the images reanalysed by McLure \\& Dunlop (2000) with the same algorithm used to model the NICMOS images in the current study. The solid line shows a least squares fit to the combined samples, and has a slope of $3.23\\pm0.42$ (cf 3.5 for the radio galaxies alone).} \\end{figure}  In our previous HST study of quasar hosts at $z\\sim 0.2$, as well as radio-loud and radio-quiet quasars we also included a sample of FRII radio galaxies which was carefully matched to the RLQ sample in terms of extended 5~GHz luminosity and radio spectral index (Taylor et al. 1996). This enabled us to compare the properties of the three main types of powerful active galaxy in the local universe, and to test models which claim to unify RLQs and radio galaxies via viewing angle effects.  We found that the radio galaxies were indistinguishable from the hosts of RLQs and luminous RQQs, having similar sizes, colours and luminosities and following a Kormendy relation identical to that of normal, inactive massive elliptical galaxies (McLure et al. 1999; Dunlop et al. 2001).  Due to scheduling pressures it was not practical to include a comparison sample of radio galaxies in the current NICMOS study. However, by utilising the data in the HST archive we are able to construct a {\\it post hoc} sample of radio galaxies which is roughly matched to the quasars at $z\\sim 0.9$.  Of all the radio galaxy images in the archive the most useful in terms of redshift, radio luminosity and observing waveband are the $I$-band images of 3CR radio galaxies made with WFPC2 by Best, Longair \\& R\\\"{o}ttgering (1997; 1998). These images have been reanalysed by McLure \\& Dunlop (2000) with the same 2-D modelling software used to analyse the current NICMOS images. By assuming an $I-J$ colour of 0.8 for elliptical galaxies at $z\\sim1$ (Fasano et al. 1998) it becomes possible to make a comparison between the quasar hosts imaged in $J$-band with NICMOS and the $I$-band radio galaxy images.  We find that the $I$-band properties of the quasar hosts are very similar to those of the radio galaxies: mean scalelengths (with standard errors) are $12.6\\pm2.3$ (RGs) and $11.0\\pm1.8$ (quasar hosts), whilst the mean absolute $I$ magnitudes are $-24.62\\pm0.13$ (RGs) and $-24.59\\pm0.21$ (quasar hosts).   \\begin{figure*} \\vspace{7.0cm} \\centering \\setlength{\\unitlength}{1mm} \\special{psfile=kukula_fig6.eps angle=0 hoffset=-5 voffset=10 hscale=70 vscale=70} \\caption{Host galaxy versus quasar (nuclear) absolute magnitudes at redshifts of $0.4$, $1$ and $2$. The dashed line shows the value of $L^{\\star}_{V}$ for galaxies in the local universe. The small grey points are the simulated data from the semi-analytic model of Kauffmann \\& Haehnelt (2000), showing the predicted evolution of the relationship between $M_{V}$ of the host and $M_{B}$ of the active nucleus, and assuming the same cosmology used in the current paper. The large symbols show the results of the current NICMOS study, with filled circles representing RQQs and open circles RLQs. Since we currently lack data for objects at $z=0.4$, in the left hand panel we plot the subset of RLQs and RQQs with $-24\\geq M_{V}(total) \\geq -25$ from our WFPC2 study of quasars at $z\\sim0.2$ (McLure et al. 1999; Dunlop et al. 2001). The triangle in the left hand panel shows the locus of the quasar sample of McLeod, Rieke \\& Storrie-Lombardi (1999). Since at each redshift our chosen filter corresponds to rest-frame $V$-band, the absolute $V$-band magnitudes can be calculated directly from the observed $R$, $J$ and $H$ magnitudes without the need to assume any particular spectral shape for the hosts. For the quasars we have converted from $M_{V}$ to $M_{B}$ using a colour index of $B-V=0.4$ (equivalent to a spectrum of the form $f(\\nu) \\propto \\nu^{-0.2}$).  (Figure adapted from Figure~12 of Kauffmann \\& Haehnelt (2000).)  } \\end{figure*}   Given that the redshift distributions of the 10-object 3CR galaxy subsample and the new $z\\sim1$ quasar sample are similar, it is also possible to investigate whether their Kormendy relations are compatible, without the need to make surface-brightness corrections. Once again, we can convert the NICMOS $J$ magnitudes for the quasars to $I$-band values by assuming an $I-J$ colour of 0.8. The two sets of data are shown in Figure~5, where it can be seen that the Kormendy relations for both quasars and radio galaxies are statistically indistinguishable. A least-squares fit to all the data points produces the relationship $\\mu_{1/2} = 3.23_{\\pm0.42} {\\rm log}r_{1/2} + 19.82_{\\pm0.4}$ (for the radio galaxies alone the fit has a slope of 3.5; McLure \\& Dunlop 2000). Thus at these redshifts both the quasar hosts and radio galaxies appear to be derived from the same parent population.  This result is consistent with the unification of RLQs and radio galaxies via viewing-angle effects, a scenario which is already well supported at low redshifts.  The lack of bright nuclear point sources in the radio galaxies means that their morphologies can be determined unambiguously; as expected all have radial surface brightness profiles which are well described by a de Vaucouleurs law. Thus, although our analysis of the quasar hosts cannot distinguish between an elliptical or a disc profile with any degree of confidence, their similarity in all other respects to the 3CR radio galaxies provides strong circumstantial evidence that, as at low redshifts, these quasars inhabit massive elliptical galaxies.  If this is the case it implies that, just as at low redshift, an important prerequisite for the presence of a powerful AGN at $z\\sim 1$ is a host galaxy with a massive spheroidal component. Within the context of hierarchical clustering models for galaxy formation, the similarity (but for passive stellar evolution) between both radio galaxies and quasar hosts at $z\\simeq1$ and their counterparts at $z\\simeq0.2$ is most simply explained by a cosmological model in which the massive elliptical galaxy formation process is essentially complete by $z \\simeq 1$.  \\subsection{The nature of quasar hosts at z$\\simeq$ 2}  The strong cosmological evolution in comoving space density of quasars has been known for nearly 40 years. Much work has gone into empirical fits of the evolution (Schmidt \\& Green 1983; Boyle, Shanks \\& Peterson 1988; Dunlop \\& Peacock 1990; Hewett, Foltz \\& Chaffee 1993; Goldschmidt \\& Miller 1998; Goldschmidt et al. 1999) but this has not led to any physical understanding of its causes.  Some recent theoretical work (Efstathiou \\& Rees 1988; Carlberg 1990; Haehnelt \\& Rees 1993; Percival \\& Miller 1999) has concentrated on explaining the evolution as being driven by evolution in the rate of galaxy merging, following the hypothesis that quasars display a cosmologically short-lived burst, or bursts, of activity following such a merger. Kauffmann \\& Haehnelt (2000) have used a semi-analytic model to link the growth of supermassive black holes with the formation and evolution of galaxy dark matter halos in a cold dark matter (CDM) universe.  They find that the evolution in merger rate of dark halos alone cannot explain the large evolution in quasar space density, in agreement with the analytic work of Percival \\& Miller (1999).  They suppose that accretion timescale also varies with cosmic epoch, and thus they effectively add a component of luminosity evolution onto the basic model (as also proposed by Haehnelt \\& Rees 1993) which can successfully reproduce the approximately luminosity evolution that is observed (Boyle et al. 2000).  Within this framework, quasars of a given luminosity are powered by progressively lower-mass black holes with increasing redshift.  Hence if black-hole mass is indeed related to the mass of the dark halo, as might be implied from the correlations between black-hole mass and galaxy bulge luminosity (Kormendy \\& Richstone 1995; Magorrian et al. 1998; van der Marel 1999) and between black-hole mass and galaxy velocity dispersion (Ferrarese \\& Merritt 2000; Gebhardt et al. 2000), the Kauffmann \\& Haehnelt (2000) model predicts that quasars of a given luminosity should, on average, be found in progressively less massive host galaxies with increasing redshift.  In fact, provided that a relationship between black-hole mass and host galaxy mass exists at all redshifts, then any luminosity-evolution model will predict such an effect, and this prediction can now be tested.  \\begin{figure*} \\vspace{6.5cm} \\centering \\setlength{\\unitlength}{1mm} \\special{psfile=kukula_fig7.ps angle=270 hoffset=-65 voffset=420 hscale=75 vscale=75} \\caption{Mean absolute $V$-band magnitude versus mean redshift for the host galaxies of the RLQs (open circles) and RQQs (filled circles) in the current NICMOS study. Also shown is the subset of 5 RLQs and 7 RQQs from our WFPC2 study of quasars at $z\\sim0.2$ which have total (host $+$ nuclear) luminosities in the same range as the high-redshift samples ($-24\\geq M_{V}\\geq-25$). Error bars show the standard error on the mean. The dotted lines show the luminosity evolution of present day $L^{\\star}$, $2L^{\\star}$ and $4L^{\\star}$ elliptical galaxies, assuming a formation epoch of $z=5$ with a single rapid burst of starformation followed by passive evolution thereafter. Left-hand panel: assuming a cosmology with $H_{0}=50$~km s$^{-1}$ Mpc$^{-1}$, $\\Omega_{m}=1.0$ and $\\Omega_{\\Lambda}=0.0$. Right-hand panel: $H_{0}=65$~km s$^{-1}$ Mpc$^{-1}$, $\\Omega_{m}=0.3$ and $\\Omega_{\\Lambda}=0.7$.} \\end{figure*}    To date, ground-based studies have not strongly supported such a picture, finding evidence for very luminous hosts around quasars at redshifts of $\\sim 2.5$. However, these observations have concentrated on the most luminous quasars, for which there are no counterparts at low redshifts.  Our current HST study thus offers the prospect of the first, proper, unbiased view of how host-galaxy luminosity varies with redshift, in a manner which offers a direct and transparent test of theoretical predictions.  This is not only because we have studied quasars of similar luminosity at $z \\simeq 0.2,1$ \\& $2$, but also because we have observed all the hosts at the same rest wavelength (removing the need for k-corrections), and because host-galaxy luminosities have been derived using an identical modelling technique at all redshifts.  In Figure~6 we show the best-fit host and nuclear luminosities for our quasar samples at $z \\simeq 0.2$, $1$ \\& $2$ superimposed onto the scatterplot predictions for quasars/hosts at $z \\simeq 0.4, 1$ \\& 2 produced by Kauffmann \\& Haehnelt (2000). As can be seen, the agreement between observation and prediction is excellent for the low-redshift sample, but becomes progressively less convincing within the higher redshift bins as the model predictions recede, while our derived host luminosities remain either roughly constant (in the case of the RQQs) or actually increase with redshift (in the case of the RLQs).  We note that the behaviour of the RLQ hosts cannot be explained away simply as the effect of the two unusually luminous RLQs in the $z\\simeq2$ sample. (The $V$ magnitudes quoted in the V\\'{e}ron-Cetty \\& V\\'{e}ron catalogue are often merely extrapolations based on data in other wavebands, so it is perhaps unsurprising that our sample should turn out to contain two objects slightly more luminous than our nominal upper limit of $M_{V}(total)=-25$.) Even when these two objects are excluded, the remaining RLQ hosts at $z\\simeq 2$ show no evidence for a drop in luminosity with redshift, and a tendency towards increasing luminosity is clearly already present at $z\\simeq1$.  To clarify the trend of host luminosity with redshift seen in our sample, we plot mean $M_{V}$ versus redshift in Figure~7, for two alternative cosmologies, overlaid with the predictions for a passively evolving galaxy formed in a short-lived burst at high-redshift ($z = 5$). Passive evolution commencing at high redshift appears to be the most reasonable starformation history to adopt, given the evidence from Nolan et al. (2001) \\& de Vries et al. (2000) that low-$z$ quasar hosts and radio galaxies are dominated by stellar populations of age $\\simeq 12$~Gyr.  These plots demonstrate a number of potentially important points. First, it is clear that in either cosmology the hosts of RLQs brighten with redshift in a manner which is perfectly consistent with pure passive evolution. Basically, the typical host of an RLQ appears to be a passively evolving (present-day) $4L^{\\star}$ elliptical. This result agrees well with what has been found for high-redshift radio galaxies. Specifically, if we compare our $z\\simeq 2$ results directly with $H$-band observations of radio galaxies at comparable redshift, the average $H$-band magnitude for radio galaxies in the redshift band $1.7 - 2.1$ discussed by Lacy, Bunker \\& Ridgway (2000) is $19.2 \\pm 0.3$, while the average $H$-band magnitude for the RLQ hosts in our $z \\simeq 1.9$ sample is $18.9 \\pm 0.7$.  Our results for RLQ hosts are also consistent with the findings of a recent ground-based imaging study of three RLQs at $z \\sim 1.5$ by Falomo et al. (2001).  Taken together, these results therefore provide further support for a picture in which the formation of ellipticals with sufficiently massive black holes to support radio activity (see McLure et al. 1999), is essentially complete before $z \\simeq 2$.  The trend displayed by the hosts of RQQs does appear to be somewhat different however. While the error bars have also grown, it seems that the gap between the luminosities of the hosts of RQQs and RLQs has grown still further to $\\simeq 1.5$ magnitudes, simply because the typical RQQ host in our sample appears to have the same absolute magnitude at all redshifts.  The interpretation of this observation is dependent on choice of cosmology.  In the flat matter-dominated cosmology, the RQQ hosts are inconsistent with having had a constant mass in stars which has evolved passively since $z = 5$.  Evaluation of any change in mass then depends on the choice of star-formation history, but is a factor of 4 under the assumption of passive stellar evolution between $z \\simeq 2$ and the present day. In the $\\Lambda$-dominated cosmology, the RQQ hosts are in fact consistent with no change in mass.  Thus, even accounting for the apparent difference between RLQ and RQQ hosts at $z \\sim 2$, the large variation in typical host mass with redshift predicted by Kauffmann \\& Haehnelt (2000) is not observed.  Recent results from other studies of the hosts of high-redshift RQQs also indicate that they are less luminous than their radio-loud counterparts. In particular, Ridgway et al. (2001) report results from NICMOS imaging of 5 RQQs at $z \\simeq 2 - 3$, in which they find host galaxies with typical luminosities of $\\sim L^{\\star}$. In fact two of their RQQs are rather fainter than those discussed here, and if one confines attention to the three objects in their sample with $M_B \\simeq -24$ their PSF-subtracted host luminosities are $\\simeq 2 L^{\\star}$. Any subsequent correction for PSF over-subtraction would only raise these values further, guaranteeing rather good agreement with the results shown in Figure~7.  Other preliminary NICMOS-based results on the luminosities of RQQ hosts at $z \\simeq 2$ have been presented by Rix et al. (1999), as an interesting by-product of the CfA-Arizona-Gravitional-Lens-Survey (CASTLES).  This study attempts to measure the host luminosities of RQQ in strongly lensed systems, by correcting the observed extended emission for the effects of the gravitational lensing, and it has been claimed that their results mirror exactly the predictions of Kauffmann \\& Haehnelt (2000). Such a conclusion may be premature since it is hard to assess the surface brightness biases inherent in this method, but nonetheless it seems unlikely that the derived RQQ host luminosities could be sufficiently underestimated (by $\\sim2$ mags) to be consistent with the hosts of RLQs and radio galaxies.  It thus seems hard to avoid the conclusion that there exists a real difference of $\\simeq 1$ mag. between the hosts of radio-loud and radio-quiet AGN of comparable nuclear optical luminosity at $z \\simeq 2$.  \\subsection{Implications}  These results raise two fundamental questions. First, what are the implications of the observed rather modest drop in RQQ host mass as a function of redshift? Second, why is a comparable drop in host mass not apparent for the RLQs?  The answer to the first question is that, provided the relationship between black-hole and host-spheroid mass is basically unchanged out to $z \\simeq 2$, our results exclude any model of quasar evolution which involves a substantial component of luminosity evolution. This is because the rather modest drop in RQQ host mass with increasing redshift implies that a typical RQQ black-hole power source at $z \\simeq 2$ is only radiating $\\simeq 2-3$ times more efficiently than at low-redshift. This is obviously completely at odds with pure luminosity evolution models which, to explain the evolution of the quasar optical luminosity function, require black holes to radiate $\\simeq 30$ times more efficiently at $z \\simeq 2$. However, it is also in conflict with more topical and apparently more realistic models, such as the hierarchical models of Kauffmann \\& Haehnelt (2000), which still require a substantial (order-of-magnitude) component of luminosity evolution to be invoked to explain the rapid evolution of the quasar luminosity function (e.g. Boyle et al. 2000).  Our results are much more consistent with a picture in which the increased availability of fuel (or, equivalently, increased frequency of mergers) at $z \\simeq 2$ results in a substantial increase in the number density of active black holes, along with a moderate increase in the fueling efficiency of a typical observed quasar. An increase of a factor of 2-3 in typical fueling efficiency between $z \\simeq 0.2$ and $z \\simeq 2$ is certainly permitted by the results of detailed studies of black-hole masses in nearby RQQs (Dunlop et al. 2001; McLure et al. 2001), which suggest that on average RQQs at $z \\simeq 0.2$ are radiating at only $\\simeq 20$\\% of their Eddington limit. It is thus not unreasonable to find that, in an era of greater fuel availability, our sample of $z \\simeq 2$ RQQs, matched in nuclear optical output to our low-redshift sample, could be largely produced by black-holes (and hence hosted by spheroids) with a characteristic mass a few times smaller than in the comparison $z \\simeq 0.2$ sample. Importantly, however, an observed trend towards moderately higher fueling efficiences with increasing $z$ does not require one to invoke some redshift-dependent change in quasar black-hole accretion mechanism.  On the contrary, the observed modest mass reduction is in fact predicted by new models which attempt to explain observed quasar evolution as a consequence of varying quasar birthrate, combined with declining light-curves for individual quasars (Miller, Percival \\& Lambert, in preparation). In these models, the increased birth-rate at $z \\simeq 2$ leads to a statistical bias such that quasars of a given absolute magnitude are more likely to be observed earlier in (i.e. closer to the peak of) their declining light-curves, and therefore closer to maximum luminosity than at low redshift, when quasar birth is relatively rare.  In answer to the second question, one possible explanation is the effect of the joint selection criteria of high radio luminosity and moderate optical luminosity used in the definition of our RLQ sub-samples. There is now growing evidence that luminous radio sources are only produced by a very massive subset of the black-hole population which powers quasars in general. In particular, the results of Dunlop et al. (2001) and McLure \\& Dunlop (2001) indicate that the RLQs in our $z \\simeq 0.2$ sample are powered by black-holes which are typically 3 times more massive than the power sources of RQQs, and are confined to the mass range $M > 10^9 {\\rm M_{\\odot}}$. If this remains true at high redshift, and if the black-hole spheroid relationship remains basically unchanged, then radio-based selection will effectively guarantee host galaxies of similar mass at all redshifts. If, as we have done in this study, one also insists that the optical luminosities of the RLQs under study are comparable at $z \\simeq 2$ and $z \\simeq 0.2$, then it would not be unexpected that such subsamples of objects would show no evidence for the redshift-dependent change in fueling efficiency found for the RQQs. This does not of course mean that RLQs would not, on average, be more efficiently fueled at high redshift, but rather that the selection criteria used here would mitigate against us observing such an effect in our sample."
    },
    "0010/astro-ph0010557_arXiv.txt": {
        "abstract": "Despite the suspected binarity for a significant fraction of Galactic lenses, the current photometric surveys detected binary microlensing events only for a small fraction of the total events.  The detection efficiency is especially low for non-caustic crossing events, which comprise majority of the binary lensing events, due to the absence of distinctive features in their light curves combined with small deviations from the standard light curve of a single point-mass event.  In addition, even they are detected, it will be difficult to determine the solution of the binary lens parameters due to the severe degeneracy problem.  In this paper, we investigate the properties of binary lensing events expected when they are astrometrically observed by using high precision interferometers.  For this, we construct vector field maps of excess centroid shifts, which represent the deviations of the binary lensing centroid shifts from those of a single lensing events as a function of source position.  From the analysis of the maps, we find that the excess centroid shifts are substantial in a considerably large area around caustics.  In addition, they have characteristic sizes and directions depending strongly on the source positions with respect to the caustics and the resulting trajectories of the light centroid (astrometric trajectories) have distinctive features, which can be distinguished from the deviations caused by other reasons.  We classify the types of the deviations and investigate where they occur.  Due to the strong dependency of the centroid shifts on the lens system geometry combined with the distinctive features in the observed astrometric trajectories, astrometric binary lensing observations will provide an important tool that can probe the properties of Galactic binary lens population. ",
        "introduction": "Searches for Galactic dark matter by monitoring light variations of stars located in the Magellanic Clouds and Galactic bulge caused by gravitational microlensing have been or are being carried out by several groups (MACHO: Alcock et al.\\ 2000; EROS: Afonso et al.\\ 1999; OGLE: Udalski, Kubiak, \\& Szyma\\'nski 1997).  The searches have been very successful and the total number of candidate events now reaches up to 1,000 (P.\\ Popowski 2000, private communication).  Among these events, a notable fraction of events are believed to be caused by binary lenses  (Udalski et al.\\ 1994; Bennett et al.\\ 1995; Alard, Mao, \\& Guibert 1995; Rhie \\& Bennett 1996; Pratt et al.\\ 1996; Albrow et al.\\ 1996; Dominik \\& Hirshfeld 1996; Alcock et al.\\ 1999a, 1999b).  Detecting binary lensing events is important because one can obtain valuable information about Galactic binary lens population such as the distributions of binary mass ratios and separations.   Due to the difference in the lens system geometry, the light curve of a binary lensing event deviates from the single-peak symmetric one of an event caused by a single point-mass lens.  The most important difference in the geometry of a binary lens system from that of a single lens system is the formation of caustics.  When a source star crosses the caustics, an extra pair of images appear (or disappear) and its flux is greatly amplified, resulting characteristic sharp spikes in the light curve. If the source does not cross the caustics, on the other hand, the light curve is less perturbed and in most cases has no distinctive features such as the spikes of the caustic crossing event.  Di Stefano \\& Perna (1997) pointed out that the detection efficiency of non-caustic crossing binary lensing events is significantly lower than that of caustic crossing events due to the absence of distinctive features in their light curves combined with small deviations.  This was well demonstrated by the recent MACHO result where non-caustic crossing events comprise only $\\sim 25\\%$ of the total detected binary lensing events despite their predicted higher frequency of occurrence (Alcock et al.\\ 1999b).  Another problem in investigating binaries from microlensing is that even if an event is detected, it is difficult to determine the solution of the binary lens parameters, especially for non-caustic crossing events, due to the severe degeneracy problem (Mao \\& Di Stefano 1995).  The level of degeneracy is so high for most gently perturbed light curves of non-caustic crossing events that many of these events allow fits to models of the binary companion ranging from equal-mass binary to planets (Di Stefano \\& Perna 1997).  Although the level of degeneracy for caustic crossing events is relatively low due to the characteristic spike feature in their light curves, they cannot be totally free from the degeneracy problem (Dominik 1999; Albrow et al.\\ 1999). Therefore, for the acquisition of accurate information about Galactic binaries from microlensing, it will be essential to devise a new method that can effectively detect binary lensing including non-caustic crossing events and can resolve their degeneracy problem.   In addition to photometry, microlensing can also be detected and observed astrometrically.  When a source star is lensed by a single point-mass lens, its image is split into two.  The typical separation between the two images for a Galactic event caused by a stellar mass lens is on the order of a milli-arcsec.  This separation is too small for direct observation of the individual images even with the {\\it Hubble Space Telescope}.  However, if microlensing is observed by using several planned high-precision interferometers from space-based platform, e.g.\\ the {\\it Space Interferometry Mission} (SIM, http://sim.jpl.nasa.gov), and ground-based large telescopes, e.g.\\ the Keck and the Very Large Telescope, one can measure the astrometric displacements in the source star image light centroid, $\\bdelta\\btheta_{\\rm c}$.  Once the centroid shifts are measured and their trajectory is constructed, one can determine the lens proper motion, from which one can better constrain the lens mass compared to the photometrically obtained Einstein ring radius crossing time $t_{\\rm E}$.  Astrometric microlensing observation was first proposed nearly simultaneously by Miyamoto \\& Yoshii (1995), H\\o\\hskip-1pt g, Novikov, \\& Polnarev (1995), and Walker (1995) and further analysis considering actual performances of specific instrument has been done by Boden, Shao, \\& Van Buren (1998).  A modified idea of monitoring a nearby high proper motion star working as a lens instead of monitoring a source star for the accurate determination of the lens mass was proposed by Miralda-Escud\\'e (1996) and Paczy\\'nski (1998). For a single lensing event, the light centroid traces out an elliptical trajectory during the event (Walker 1995; Jeong, Han, \\& Park 1999).  If an event is caused by a binary, on the other hand, the resulting trajectory of the light centroid (astrometric trajectory) deviates from the elliptical one of the single lensing event (Chang \\& Han 1999), which is analogous to the deviations in the photometric light curve.   In this paper, we investigate the properties of binary lensing events when they are astrometrically observed by using high precision interferometers. For this, we construct vector field maps of excess centroid shifts, $\\Delta\\bdelta\\btheta_{c}(\\xi,\\eta)$, which represent the deviations of the binary lensing centroid shifts from those of a single lensing event as a function of source position $\\bzeta=(\\xi,\\eta)$.  From the analysis of the maps, we find that the excess centroid shifts are substantial in a considerably large area around caustics.  In addition, they have characteristic sizes and directions depending strongly on the source positions with respect to the caustics and the resulting astrometric trajectories have distinctive features which can be distinguished from the deviations caused by other reasons.  We classify the types of the deviations and investigate where they occur. ",
        "conclusions": "The findings about the astrometric properties of binary lensing events from the analysis of the vector field maps of excess centroid shifts for various types of binary lens systems are summarized as follows. \\begin{enumerate} \\item The lens binarity causes large deviations in the astrometric trajectory from the elliptical one of a single lensing event.  The excess centroid shifts are substantial even in the regions considerably away from the caustics, implying that the distortions will be significant even for a significant fraction of non-caustic crossing binary lensing events.  \\item Depending on the locations with respect to lens caustics, the excess centroid shifts have very characteristic sizes and directions.  As a source passes through these regions, the resulting astrometric trajectories have distinctive features, which are categorized by concave, convex, and twisting distortions.  \\item The binary-lens induced deviations in the astrometric trajectory can be distinguished from those caused by other effects, such as the parallax, blending, and finite source effects. \\end{enumerate} Therefore, astrometric microlensing observations will be an efficient method to detect events caused by binary lenses and to constrain their parameters with accuracy.    {\\bf Acknowledgements}: This work was supported by grant KRF-99-041-D00442 of the Korea Research Foundation.  \\bigskip \\clearpage"
    },
    "0010/astro-ph0010411_arXiv.txt": {
        "abstract": "We have measured the Li abundance of 18 stars with $-2~^<_\\sim$~[Fe/H]~$^<_\\sim~-1$ and 6000~K~$^<_\\sim~T_{\\rm eff}~^<_\\sim$~6400~K, a parameter range that was poorly represented in previous studies.  We examine the Galactic chemical evolution (GCE) of this element, combining these data with previous samples of turnoff stars over the full range of halo metallicities.  We find that $A$(Li) increases from a level of $~\\sim 2.10$ at [Fe/H]~=~$-3.5$, to $\\sim 2.40$ at [Fe/H]~=~$-1.0$, where $A$(Li)~= log$_{10}$($n$(Li)/$n$(H))~+~12.00.  We compare the observations with several GCE calculations, including existing one-zone models, and a new model developed in the framework of inhomogeneous evolution of the Galactic halo.  We show that Li evolved at a constant rate relative to iron throughout the halo and old-disk epochs, but that during the formation of young-disk stars, the production of Li relative to iron increased significantly.  These observations can be understood in the context of models in which post-primordial Li evolution during the halo and old-disk epochs is dominated by Galactic cosmic ray fusion and spallation reactions, with some contribution from the $\\nu$-process in supernovae.  The onset of more efficient Li production (relative to iron) in the young disk coincides with the appearance of Li from novae and AGB stars.  The major challenge facing the models is to reconcile the mild evolution of Li during the halo and old-disk phases with the more efficient production (relative to iron) at [Fe/H]~$>~-0.5$. We speculate that cool-bottom processing (production) of Li in low-mass stars may provide an important late-appearing source of Li, without attendant Fe production, that might explain the Li production in the young disk. ",
        "introduction": "%  Lithium plays several valuable roles as a diagnostic of stellar and Galactic evolution.  As the only metal synthesized in significant quantities in the big bang, $^7$Li provides a rare constraint on the baryon density of the universe (e.g., Ryan et al. 2000).  As an element destroyed in stars where the temperature exceeds $2.5\\times 10^6$~K, its survival at the stellar surface indicates the degree of exchange of material between the surface and interior via convection, diffusion and other processes.  Thirdly, as a product of spallation and fusion reactions and of stellar sources, it provides a measure of the chemical evolution of the Galaxy.  In practice, the three roles of Li cannot be treated in isolation.  The primordial (big bang) abundance cannot be determined without knowing the sources and sinks of the element, and the degree of mixing for stars of different metallicity cannot be determined from the observations unless the contribution of Galactic production is known.  Normally, one attempts to reduce the complexity of the problem by isolating one or two parts.  For example, the near-constancy of the $^7$Li abundance in warm halo stars over a range of effective temperature and metallicity led Spite \\& Spite (1982) to conclude that destruction of Li in those stars and its production in the course of Galactic chemical evolution were negligible. As a result, they argued that it was proper to consider the observed Li abundance as the primordial one, hardly altered.  The view that Li in these objects was unaltered was supported empirically by the small spread in Li abundances and by classical stellar evolution models (e.g., Yale `standard' models [Deliyannis, Demarque, \\& Kawaler 1990]) which showed negligible levels of pre-main sequence and main sequence destruction even over the long ($\\simeq$13 Gyr) lives of the objects.  (This contrasted with the considerable destruction seen in some young open clusters [e.g., Hobbs \\& Pilachowski 1988], but is understood as the depth of the surface convection zone being less in stars of lower metallicity, thus not reaching the depths required for burning at $T \\ge 2.5\\times 10^6$~K).  Challenges to the Li survival hypothesis have come from both theoretical and observational sources.  More complex (and hopefully more realistic, but also more uncertain) stellar evolution models, involving rotationally-induced mixing, were found to be able to deplete significant fractions of the Li in these objects.  Early models suggested as much as 90\\% could be lost (Pinsonneault, Deliyannis, \\& Demarque 1992), though later work suggested perhaps half might be destroyed (Pinsonneault et al. 1999).  (The downward revision of the figure was driven partly by the models and partly by observational data.) Coupled to this theoretical work were claims of significant scatter in the observed abundances, inconsistent with a single primordial value (Deliyannis, Pinsonneault, \\& Duncan 1993; Thorburn 1994). The most recent observations show, however, that the intrinsic scatter in a sample of 22 halo field turnoff stars is $\\sigma_{\\rm int}~<~0.02$~dex, and does not support the proposition of more than 0.1 dex $^7$Li depletion by the rotational mixing mechanism (Ryan, Norris, \\& Beers 1999). Although this field star sample places very tight limits on the intrinsic spread in $^7$Li, there is evidence of at least some star-to-star differences in the halo.  Boesgaard et al. (1998) find a spread in $^7$Li abundance amongst subgiants in M92 (see Figure 1), Deliyannis (1999, priv.comm.) finds $^7$Li differences between the extremely metal-poor field stars G64-12 and G64-37, stars which otherwise appear very similar to one another.  Furthermore, there exist a small number of very Li-deficient stars which are otherwise indistinguishable from halo Li ``preservers'' (Hobbs, Welty, \\& Thorburn 1991; Thorburn 1992; Spite et al. 1993; Norris et al. 1997; Ryan, Norris, \\& Beers 1998).  However, the different (dense) stellar environment of the M92 stars and small volume of the Galaxy it samples, and the rarity of the other cases, leads us to view the small observed spread of Li abundance in the recent halo-star sample ($\\sigma_{\\rm int}~<~0.02$) as representative of the majority of the halo.  In addition to discussions of the intrinsically thin Spite halo Li plateau, claims have been made of the existence of dependencies of observed Li abundance upon $T_{\\rm eff}$ and [Fe/H].  Thorburn (1994), Norris, Ryan, \\& Stringfellow (1994), and Ryan et al.  (1996a) all found the halo observations to require significantly non-zero coefficients to fits of the form $A{\\rm (Li)}~=~A_0~+~A_1{{T_{\\rm eff}}\\over{100~{\\rm K}}}~+~A_2{\\rm [Fe/H]}$, where $A$(Li)~= log$_{10}$($n$(Li)/$n$(H))~+~12.00.  Typical estimated values of the coefficients were $A_1$~=~0.03 and $A_2$~=~0.14.  The coefficient on $T_{\\rm eff}$, $A_1$, may depend crucially on the adopted temperature scale.  The optical photometric scales used in the cited studies were challenged by Bonifacio \\& Molaro (1997) who used temperatures derived from application of the infrared flux method (IRFM), whereupon they concluded that both $A_1$ and $A_2$ were consistent with zero, which is to say that the Spite plateau is flat.  The IRFM scale has often been proposed as less likely to be affected by metallicity-dependent systematic errors (Saxner \\& Hammarb\\\"ack 1985; Magain 1987), but the uncertainties in the $T_{\\rm eff}$ of any individual star are still considerable, with $\\sigma_{T_{\\rm eff}}~\\simeq~100$~K (Alonso, Arribas, \\& Mart\\'inez--Roger 1996; Bonifacio \\& Molaro 1997).  The other photometric scales can at least lead to small internal errors, $\\sigma_{T{\\rm eff}}~\\simeq~30-40$~K (Ryan et al. 1999), and a mean error of 55~K in the present sample, but possibly with less reliable external systematics.  The trade-off is that the IRFM may deliver better systematics but at the expense of introducing greater internal scatter. Hopefully, improvements in the systematics of the optical temperature scales and in the internal errors of the IRFM scale will be achieved, and we will be able to clarify the size of the $A_1$ term in the near future.  Bonifacio \\& Molaro (1997) also found the $A_2$ term (metallicity coefficient) to be consistent with zero, but the work by Ryan et al. (1999), which achieved errors as small as $A$(Li)~=~0.03 dex for most stars, again found a significant value: $A_2~=~0.118~\\pm~0.023$.  Ryan et al. traced the main difference between these two results to some substantial differences in the stellar metallicities adopted (from the literature) by the two studies.  After comparing with a third [Fe/H] estimate (the homogeneously applied estimator of metallicity obtained by Beers et al. 1999), they argued that the [Fe/H] values adopted by Bonifacio \\& Molaro had sufficiently large errors to obscure the $A_2$ analysis.  Ryan et al. further argued that measurements of $^6$Li in HD~84937 and BD~+26$^\\circ$2578 (Smith, Lambert, \\& Nissen 1993, 1998; Cayrel et al. 1999; Hobbs \\& Thorburn 1994, 1997), and now also in HD~140283 (Deliyannis \\& Ryan 2000), likewise evidenced the contribution of Galactic chemical evolution (GCE) to Li production even in stars with [Fe/H]~$\\sim~-2.5$, since $^6$Li is thought to be exclusively post-primordial.  The contributions to the total Li inferred from $A_2$ and from $^6$Li were found to be compatible.  From application of the inhomogeneous GCE model of Suzuki, Yoshii, \\& Kajino (1999), Suzuki, Yoshii, \\& Beers (2000a) argue that a non-zero slope in the relationship between $A$(Li) and [Fe/H] (of the same order of magnitude as that observed) must arise in the early Galaxy from Li production associated with the spallation reactions that give rise to Be and B.  In their earlier study, Ryan et al. (1996a) noted that an observational bias existed in the available Li data for halo stars.  In the quest for the primordial lithium abundance, observers had studied progressively more metal-poor stars, but had examined rather fewer at [Fe/H]~$\\sim~-1.5$. Moreover, those that were examined at higher [Fe/H] were invariably cooler than the more metal-poor ones, potentially complicating the analysis of the coefficients $A_0$, $A_1$, and $A_2$ by the inadvertent introduction of collinearity in the predictor variables.  To address both of these difficulties, we set out to measure a sample of hotter (6000~K~$^<_\\sim~T_{\\rm eff}~^<_\\sim$~6400~K), more metal-rich ($-2~^<_\\sim$~[Fe/H]~$^<_\\sim~-1$) stars.  The sample selection, observations, and abundance analysis are discussed in the following sections.  We then combine the new data on 14 stars, which correct the previous paucity of warmer, higher metallicity halo stars with existing observations, and examine GCE of Li.  In addition to these 14 stars we report on four stars with $A$(Li)~$<~1.7$, two of which are newly discovered extremely Li-deficient halo stars.  These exceptional stars are discussed in detail in a separate paper (Ryan et al. 2001). ",
        "conclusions": "%  We have measured the Li abundances of 18 stars with $-2~<$~[Fe/H]~$<~-1$ and 6000~K~$<~T_{\\rm eff}~<$~6400~K, populating a previously poorly-sampled region of the $T_{\\rm eff}$, [Fe/H] plane. Of these, four proved to be highly Li-deficient, with $A$(Li)~$<~1.7$ (and are discussed in the companion paper). The remaining 14 were found to conform to the same trends of $A$(Li) with $T_{\\rm eff}$ and [Fe/H] identified earlier by Ryan et al. (1996a), removing any doubts that the results of that study were affected by the selection biases inherited in the observational samples.  It is unclear whether the $T_{\\rm eff}$ trend we have identified is intrinsic or due to the photometric effective temperature scales used, but this uncertainty was largely circumvented by restricting our attention to stars in a narrow $T_{\\rm eff}$ range by excluding those with $T_{\\rm eff}~<~6000$~K. This sub-sample of turnoff halo stars, supplemented with objects from previous studies occupying the same temperature range, revealed a {\\it significant increase} of $A$(Li) over the metallicity range of the halo, from $A$(Li)~=~2.10 at [Fe/H]~=~$-3.5$ to $A$(Li)~=~2.40 at [Fe/H]~=~$-1.0$.  We examined various GCE models in an attempt to understand the halo and disk phases of Li production, and showed, with a simple linear-evolution model, that the net Li production rate relative to iron must increase substantially during young-disk evolution.  A very satisfactory match to the halo and old-disk data was found in the three-component (primordial, GCR, and $\\nu$-process) model of Fields \\& Olive (1999a,b; Ryan et al. 2000). The additional sources of stellar nucleosynthesis in the young disk ([Fe/H]~$>~-0.5$)  are well represented by the models of Romano et al. (1999), whose main contributors are novae, and to a lesser extent, AGB stars. A hybrid of the two models provided the best current match to the halo and disk data together. In addition, a new model of halo GCE in an inhomogeneous framework, extending the work of Suzuki et al. (1999), was presented which was equally capable of modelling the halo data.  While none of these models presents a {\\it perfect} fit to {\\it all} epochs of Galactic evolution, the match between the models and data is sufficiently good to believe that the models are viable, albeit not uniquely so.  The primary remaining challenge is to reproduce the efficient production of Li during late stages of disk evolution. A simple fiducial model demonstrates that Li production relative to iron production increases significantly at [Fe/H]~$>~-0.5$. The disk model of Romano et al. currently comes closest to predicting this, but we also speculate that the recently identified cool-bottom processing (production) of Li in low-mass red giants (Sackmann \\& Boothroyd 1999) may provide a late-appearing source of Li without attendant Fe production."
    },
    "0010/astro-ph0010319_arXiv.txt": {
        "abstract": "Clusters of galaxies can be used for very different kinds of cosmological tests. I review a few of the methods: determination of cluster masses and dark matter content, metal enrichment and its connection to the origin of the intra-cluster gas, and the dynamical state of clusters. The progress in the different fields expected from observations with the new X-ray satellites XMM and Chandra is pointed out. ",
        "introduction": "In recent years clusters of galaxies have been shown to be excellent diagnostic tools for cosmological research. They provide a variety of different ways to determine cosmological parameters like the mean density $\\Omega$ of the universe, the baryon density $\\Omega_{baryon}$, the cosmological constant $\\Lambda$  or the Hubble constant $H_0$. Galaxy clusters are special for several reasons. (1) They are the largest bound structures in the universe. With sizes of a few Mpc and masses around $10^{15} \\msol$ they constitute already a considerable fraction of the universe. (2) They are closed systems. As no matter can leave the cluster potential well the mix of baryons and non-baryons should be representative for the universe as a whole. (3) The crossing time -- the time it takes a galaxy to move from one end of the cluster to the other -- is of the order of a Hubble time. Therefore the information of the formation process is not completely wiped out, but there are still traces left, which provide insights into the early universe. (4) Clusters are observable out to large redshifts. Comparison of properties of distant and nearby clusters therefore yields information about evolutionary effects, which are predicted to be different in different cosmological scenarios. (5) They can be used as tracers for the mass distribution in the universe, which is another quantity to constrain cosmological models. ",
        "conclusions": "Clusters of galaxies can be used for very different types of cosmological tests. In particular observations with the new X-ray satellites launched recently, XMM and Chandra, will provide a large step forward in several respects. The unprecedentedly high spatial resolution of Chandra (1\") provides detailed X-ray images to investigate cluster morphology and derive very accurate profiles. The high sensitivity of XMM makes it the ideal instrument to find and study distant galaxy clusters in order to investigate evolutionary effects. Improved spectral resolution allows for much better temperature and metallicity measurement. And -- very importantly -- XMM and Chandra are capable of performing spatially resolved spectroscopy with high accuracy, which is necessary for temperature maps, mass determination and metallicity distributions. The combination of these data with observations in other wavelengths will open new horizons in cosmology."
    },
    "0010/astro-ph0010633_arXiv.txt": {
        "abstract": "\\begin{center}{\\large\\bf Abstract}\\end{center} During the last years it has become clear that global $O(N)$ defects and $U(1)$ cosmic strings do not lead to the pronounced first acoustic peak in the  power spectrum of anisotropies of the  cosmic microwave background which has recently been observed  to high accuracy. Inflationary models cannot easily accommodate the low second peak indicated by the data. Here we construct causal scaling seed models which reproduce the first and second peak. Future, more precise CMB anisotropy and polarization experiments will however be able to distinguish them from the ordinary adiabatic models. ",
        "introduction": "Inflation and topological defects are two classes of models to explain the origin of large scale structure in the universe. In inflationary models, for fixed cosmological parameters the fluctuation spectrum is determined by the initial conditions. In models with topological defects or other types of seeds, fluctuations in the cosmic plasma and in the geometry are continuously induced by the gravitational coupling to the seed energy momentum tensor.  Cosmic microwave background~(CMB) anisotropies provide a excellent link between theoretical predictions and observational data. They allow us to distinguish between inflationary perturbations and models with defects by purely linear analysis.  On large angular scales, both classes of models predict an approximately scale-invariant Harrison-Zel'dovich spectrum \\cite{h,z}. For inflationary models this can be seen analytically. Scale-invariance for defects was discovered numerically \\cite{SPT,DZ,Shellard}; simple analytical arguments are given  {\\em e.g.} in \\cite{Roma}.  On smaller angular scales ($10' \\lsim \\theta \\lsim 2^{\\circ}$), the predictions of inflation and global $O(N)$ defects are different. While inflationary models predict a series of 'acoustic peaks', global  $O(N)$ defects show a low amplitude broad 'hump'~\\cite{DGS,Seljak,DKM}. For local $U(1)$ cosmic strings, the result is not so clear. Depending on the detailed modeling of local cosmic strings, the resulting acoustic peaks are quite different. The peak can be entirely absent~\\cite{Shellard} or present and even quite substantial, but at an angular harmonic $\\ell \\sim 400$ -- $500$~~\\cite{Albrecht,Joao,Vachaspati}.  Recent experiments~\\cite{toco,B97,B98,MAX} have measured  CMB anisotropies which are fully compatible with a flat adiabatic inflationary model on the scale of the first peak and incompatible with the above mentioned defect models. However, the second peak is too low for values of the baryon density that are within the constraints inferred from standard nucleosynthesis. Combining the recent BOOMERanG and MAXIMA-I data with informations from the distribution of galaxies, a value of $\\Omega_bh^2=0.032\\pm0.004$ was found in \\cite{jaffe} (see also \\cite{pierpol,tegman,tegman2,lange}), which is incompatible at nearly $3 \\sigma$ with the value $\\Omega_bh^2=0.0189 \\pm 0.0019$ \\cite{burles} inferred mainly from measurements of primordial deuterium from Ly-alpha absorption systems in the continuum emission of $3$ high redshift quasars.  Even if it is fair to say that the possibility of systematic errors in all these data sets needs further investigation, several, mainly phenomenological, mechanisms have been put forward to solve the problem of the low second peak. The simplest is clearly to modify standard nucleosynthesis so that a higher value of the baryon density parameter, which leads to a suppression of even peaks becomes acceptable~\\cite{Ale,ori,kapli,Lesg,hansen,dibari}. Another suggestion is to modify one of the 'pillars' of the inflationary model, the nearly scale-invariant primordial spectrum of fluctuations by adding features on it \\cite{kinney,louise,subir}. Also models with $\\Omega_bh^2=0.019$ and with a 'red' tilted spectral index $n \\sim 0.9$, even if not preferred with respect to the $\\Omega_bh^2=0.03$ and $n=1$ models, give a reasonable $\\chi^2$-fit to present data (see, e.g. \\cite{kana,kmr,covi,moro}). Furthermore, in Ref.~\\cite{ave,bat} it has been  found that a time-varying fine-structure constant can increase the compatibility between CMB and BBN data. Finally, a combination of inflation with topological defects which can contribute to the Sachs-Wolfe plateau and to the first peak but not to the second or third peak, has also been proposed~\\cite{Cont,Mairi} as a possible resolution to the problem of the low secondary peaks.   Here we want to investigate whether generic defect models, the so-called 'causal scaling seeds'  models, can reproduce the new data. A couple of years ago, Neil~Turok constructed a model with scaling causal seeds which perfectly reproduced the CMB anisotropy spectrum of inflationary models~\\cite{Turokmodel}. Other synthesized causal seed models with various heights of the acoustic peaks are discussed in~\\cite{MR,DKLS}. Spergel \\& Zaldarriaga argued that causal seeds can nevertheless be distinguished from inflationary models by the induced polarization~\\cite{SZ}. Our investigations confirm and extend this result. But here we shall not only play with some parameters describing the model, but we also vary cosmological parameters, especially the total curvature which basically determines the angular diameter distance and thereby the angular scale onto which the peaks in the power spectrum are projected. ",
        "conclusions": "In this paper we have shown that causal scaling seed models for structure formation can reproduce the recent CMB anisotropy data~\\cite{B98,MAX}. A first very simple closed model (family I) can be brought in reasonable agreement with all but the last two Maxima-1 points for a baryon density which is not compatible with the nucleosynthesis constraint. It is interesting to note that present data already slightly  disfavors closed isocurvature models since they have a peak which is narrower than what is preferred by the data. A somewhat more refined model  (family II) is, for a suitable choice of parameters, in excellent agreement with all data points in a flat, $\\La$-dominated universe. The cosmological parameters of our best fit model~I are $\\Om_m=\\Om_{cdm}+\\Om_b=0.35,~\\Omega_\\La=0.85, h=0.65,~h^2\\Om_b=0.019$ and those of model~II are $\\Om_m=\\Om_{cdm}+\\Om_b=0.4,~\\Omega_\\La=0.6, h=0.65,~h^2\\Om_b=0.019$. This model is preferred by the data with respect to the 'concordance model' with $\\Om_m=0.3,~\\Omega_\\La=0.7,~h=0.65,~h^2\\Om_b=0.019$ and inflationary initial conditions. These models can, however, be clearly distinguished from inflationary models by future experiments either measuring the secondary peaks or the polarization spectrum. The first one, a closed model,  has smaller inter-peak distances than flat inflationary models  (see fig.~\\ref{ct}), a definitive lower amplitude of temperature fluctuations for $\\ell \\ge 650$ and a {\\it greater} r.m.s. amplitude of polarization for $50 \\le \\ell \\le 550$; in the second one the first peak $\\ell \\sim 150$ in the polarization spectrum  is not present (see fig.~\\ref{cp}), which is a consequence of causality, and the polarization amplitude is generally lower in the band $0 \\le \\ell \\le 800$.  To achieve this agreement we have suppressed vector and tensor perturbations and have assumed perfectly coherent fluctuations. We believe that it is quite improbable that topological defects from a GUT phase transition have such a behavior. Nevertheless, there might be some other scale-invariant causal physical mechanism (e.g. some spherically symmetric 'neutrino explosions', see Ref.~\\cite{Turokmodel}) leading to seeds of this or similar type. Clearly, we only have a satisfactory model of structure formation if also the physical origin of the 'seeds' is clarified. However, the point of this work was not to find ``the correct model of large scale structure formation'' but mainly to investigate, in a phenomenological but at the same time physically motivated way, to what extent the present values of the cosmological parameters derived from accurate CMB data analysis can still be plagued by the assumption of the underlying theoretical model. We have seen {\\em e.g.} that flatness, $\\Om_m+\\Om_\\La=1$ is not mainly supported by the position of the first peak but by its {\\em width}. Clearly, once secondary peaks are unambiguously detected, the inter peak distance will represent another direct measure of the total curvature.  This investigation is rather important especially if some of the parameters obtained assuming the standard inflationary model are in significant disagreement with complementary, more direct observations, as the high $\\Omega_bh^2$ value seems to suggest. While present CMB data can be regarded as a triumph for a scenario based on primordial adiabatic fluctuations, we have presented here phenomenological models, based on isocurvature fluctuations, that also give a good fit to the CMB data. Fortunately, the concrete models proposed here have peculiar characteristics that future CMB experiments will be able to detect\\footnote{A systematic study of the precision with which cosmological parameters can be determined by the Planck satellite if the model space is enlarged to allow for arbitrary isocurvature models {\\em without seeds}, is presented in Ref.~\\cite{Bucher}}. \\vspace{0.2cm}\\\\ {\\bf Acknowledgments}\\hspace{0.5cm} We thank Louise Griffith, Mairi Sakellariadou, Jean-Philippe Uzan and Julien Devriendt for stimulating discussions.\\\\ This work is supported by the Swiss National Science Foundation."
    },
    "0010/astro-ph0010405_arXiv.txt": {
        "abstract": "A dense-enough gas-accumulation evolves, over a few~Myr of intensifying star formation, to an embedded cluster.  If it contains a sufficient amount of mass, O~stars form and explosively expel the remaining gas, whereas poorer clusters reduce their embryonic gas content more gradually. The sudden expulsion of gas unbinds most of a rich cluster, but a significant fraction of it can condense by two-body interactions to become an open cluster despite a star-formation efficiency as low as 30~per cent.  Poorer clusters survive their gradual mass loss more easily, but have short, relaxation-limited life-times. Pleiades-like clusters may thus form as nuclei of expanding OB associations, by filling their tidal radii and having large (1--2~pc) core-radii. A 'main-sequence' of clusters is thus established. Ultimately, a cluster dies an explosive death through the ever shortening relaxation time, and leaves a remnant that consists of about 4--10 stars arranged in a highly hierarchical and thus long-lived system. Dynamical mass segregation in very young clusters is extremely rapid, and heats a cluster substantially, which is partially off-set by the cooling from the disruption of primordial binaries. ",
        "introduction": "\\noindent Star clusters fascinate because some are beautifully evident to the unaided eye, and because their birth, life and death remain mysterious. Wonderful examples 'magically' lying together in one quadrant on the summer sky (southern hemisphere) are M42 (the Orion Nebula Cluster, ONC), and the Pleiades and Hyades Clusters. Their ages are $\\tau_{\\rm cl}\\approx1$~Myr, 100~Myr and 600~Myr, respectively, and latest research indicates they may form approximately one evolutionary sequence. Such clusters may be the origin of a significant proportion of Galactic-field (GF) stars, which is one reason why we want to understand their behaviour.  In this contribution (KII), the aim is to convey some theoretical aspects concerning the birth, evolution and death of open clusters, but I stress that much remains to be worked out in this exciting field. Complementary texts are Kroupa (2000a, KI; 2000b, KIII).  A few quantities useful for roughly assessing the global state of a cluster are the half-mass diameter crossing-time [Myr],  \\hfill $t_{\\rm cr} = 4.2\\,\\left(M_{\\rm st}/\\left(\\epsilon\\,100\\,M_\\odot\\right) \\right)^{-1/2} \\left(R_{0.5}/1\\,{\\rm pc}\\right)^{3/2}$, \\hfill (1)  \\noindent the median relaxation-time [Myr] for a purely stellar system (Binney \\& Tremaine 1987, BT)  \\hfill$t_{\\rm rel} = \\left(21/ \\ln(0.4\\,N)\\right) \\left(M_{\\rm st} / 100\\,M_\\odot\\right)^{1/2} \\left(\\overline{m}/1\\,M_\\odot \\right)^{-1} \\left(R_{0.5} / 1 \\, {\\rm pc}\\right)^{3/2}$, \\hfill (2)  \\noindent and the three-dimensional velocity dispersion [pc/Myr] in the cluster,  \\hfill $\\sigma = 0.47\\,\\left(M_{\\rm st}/\\left(\\epsilon\\,100\\,M_\\odot\\right)\\right)^{1/2} \\left(R_{0.5}/1\\,{\\rm pc} \\right)^{-1/2}$. \\hfill (3)  \\noindent Here, $R_{0.5}$ is the half-mass radius, $M_{\\rm st}/\\epsilon = M_{\\rm g}+M_{\\rm st}$, $M_{\\rm g}$ being the gas mass and $\\epsilon$ the star-formation (sf) efficiency, and $M_{\\rm st}=N\\,\\overline{m}$ is the mass of the cluster containing $N$ stars that have an average mass $\\overline{m} = 0.36 \\, M_\\odot$, assuming the universal (GF) IMF (Kroupa 2000c) with stellar masses in the range $0.01-50\\,M_\\odot$, $G=4.49\\times10^{-3}\\,{\\rm pc}^3/(M_\\odot\\,{\\rm Myr}^2)$, and 1~pc/Myr~$\\approx$~1~km/s.  For one massive star with $m=10\\,(20)\\,M_\\odot$, the GF IMF implies $N_{10 \\left(20\\right)}=371\\,(1155)$ stars with $m\\in\\left[0.01,10\\right]\\, \\left(\\left[0.01,20\\right]\\right) \\,M_\\odot$. These numbers are useful for determining if an embedded cluster is likely to contain stars massive enough to remove the natal gas within a time $\\tau_{\\rm g}<t_{\\rm cr}$, i.e. 'explosively'. In what follows, {\\it poor} clusters contain $N\\simless10^3$ stars, whereas {\\it rich} clusters contain $N\\simgreat10^3$~stars.  Typical numbers for a few well-known clusters are $\\tau_{\\rm cl}\\simless1$~Myr, $N = 5000-10000, R_{0.5}\\approx0.45$~pc (ONC), $\\tau_{\\rm cl}\\simless1$~Myr, $N\\approx200, R_{0.5}\\approx0.5$~pc ($\\rho$~Oph, still embedded in gas), $\\tau_{\\rm cl}\\approx100$~Myr, $N\\approx3000, R_{0.5}\\approx3$~pc (Pleiades), and $\\tau_{\\rm cl}\\approx600$~Myr, $N\\approx2000, R_{0.5}\\approx4$~pc (Hyades). ",
        "conclusions": ""
    },
    "0010/astro-ph0010296_arXiv.txt": {
        "abstract": "We test the effects of magnetic fields on the mass determination of galaxy clusters with the X-ray method. As this method takes into account only thermal pressure, additional non-thermal pressure support, e.g. caused by a magnetic field, can result in an underestimation of the mass. SPH models of galaxy clusters including magnetic fields of various strengths in two different cosmologies are used for the test. We compare the true mass of the model clusters with the mass determined from simulated X-ray images of the same models. The effect of the magnetic field on the mass in relaxed clusters is found to be negligible compared to other uncertainties, even at small radii, where the strongest effects are expected. However in one model, in which the cluster is in the process of merging, we find an effect of up to a factor of 2  caused by the magnetic field. In relaxed clusters the mass estimate, averaged over many models, is at most 15\\% smaller in models with relatively high magnetic field of $2\\mu$G compared to models without a magnetic field with a small dependence on the cosmology chosen. Therefore we conclude that the presence of a magnetic field cannot be responsible for the discrepancies of a factor of 2-3 found between X-ray and lensing mass in the central regions of relaxed clusters. ",
        "introduction": "Clusters of galaxies -- the most massive bound objects in the universe -- are excellent diagnostic tools for cosmological parameters. In particular, measuring the masses of clusters provides a way to determine the amount and the distribution of dark matter and thus $\\Omega_\\mathrm{m}$. The comparison with the baryonic component yields an estimate for $\\Omega_\\mathrm{baryon}$, which is another important cosmological parameter, as current measurements are in contradiction with primordial nucleosynthesis for a $\\Omega=1$ universe (White et al. 1993).   Cluster masses can be measured in various ways, via the gravitational lensing effect, via the assumption of virial equilibrium and with X-ray observations using the assumption of hydrostatic equilibrium. In some clusters these different methods yield masses which can differ up to a factor of three in the central region of the cluster with the lensing mass being typically larger, e.g. A2218 (Kneib et al. 1995; Squires et al. 1996), MS0440+0204 (Gioia et al. 1998),  RXJ1347-1145 (Fischer \\& Tyson, 1997, Schindler et al. 1997; Sahu et al. 1998), Cl0939+4713 (Seitz et al. 1996; Schindler et al. 1998). Various reasons for these discrepancies have been suggested.  The presence of substructure and cluster mergers are obvious ways to explain the differences, but tests with numerical simulations showed that the assumptions of hydrostatic equilibrium and spherical symmetry are well justified in relaxed clusters of galaxies (Evrard et al. 1996; Schindler 1996) and cannot account for the large discrepancies in relaxed clusters. Also the effects of small-scale density variations were found to be much smaller than the observed discrepancies (Mathiesen et al. 1999). As a further possibility for the discrepancies, in particular in the cluster centres,  Loeb \\& Mao (1994) suggested that the intra-cluster gas might not only be supported by thermal pressure, but also by considerable magnetic pressure. As magnetic pressure is not taken into account in the X-ray mass determination method it potentially leads to an underestimation of the mass.   Whereas Loeb \\& Mao (1994) suggested that strong magnetic fields can be hidden to Faraday rotation measurements if the magnetic fields are tangled, Dolag et al. (1999) found in simulations that even clusters with an overall small magnetic field can be penetrated partially by regions of high magnetic fields. These simulations are based on a consistent model for the cluster magnetic field and therefore lead to realistic magnetic field configurations within the assumed cosmological scenarios. Here the magnetic fields strengths and structure is given by the formation of the cluster in a cosmological environment. Although on average the magnetic pressure in there models is much smaller than the thermal pressure ($\\approx 5\\%$ (Dolag et al. 2000b)), the models show some scatter around this mean value. There are domains of high magnetic fields approaching or sometimes even exceeding equipartition with the thermal energy. The simulations predict the number, locations and sizes of these domains depending on the formation history and dynamical state of the cluster, which provides the possibility to test the importance of these domains. So far the influence of these domains on the mass determination is not clear.  In this work we use the models by Dolag et al. (1999) to test the influence of such magnetic fields on the mass determination with the X-ray method. We compare the true cluster mass, which is known exactly for the model clusters, with the mass derived from a simulated X-ray observation of the same model cluster.  After the description of the method in Sect.~2 we present the results in Sect.~3 and the conclusions in Sect.~4. ",
        "conclusions": "A magnetic field can affect the mass determination in a cluster of galaxies in principle in two ways, by the isotropic and the anisotropic component of the magnetic pressure. The isotropic component provides an additional non-thermal pressure support, hence tends to lower the temperature in the cores of clusters. Depending on the gas mass fraction, this effect can amount to $\\approx 5\\%$ of the central temperatures for EdS cosmologies (Dolag et al. 2000). Consequently, the mass is underestimated by the X-ray method because this additional pressure is not taken into account. The anisotropic component of the magnetic pressure slightly changes the density and temperature distribution in the inner region of the cluster by redirecting gas flows. In this way the gas is prevented from reaching hydrostatic equilibrium. This may also lead to a possible misinterpretation of the X-ray emission when determining the mass.  To test these effects, we performed cosmological MHD simulations of galaxy clusters in different cosmologies, to examine the effect of magnetic fields in galaxy clusters on the mass reconstruction via the X-ray method. The range of initial magnetic field strengths was chosen so as to reproduce the observed Faraday-rotation measurements at the final redshift of $z=0$.  We find one extreme merger model, in which the sum of all effects caused by the magnetic field can reduce the reconstructed mass by a factor of 2 in the central region. This model cluster is in the process of merging and has therefore a particularly high magnetic field. As the mass determination for merging clusters canbe unreliable, because they are far from hydrostatic equilibrium, this adds up to already large uncertainties, so that we can just emphasise again, that X-ray mass estimates of merging clusters must be regarded with caution.  \\begin{figure} \\centerline{\\epsfxsize=0.99\\hsize\\epsffile{plot3.ps}} \\caption[]{Ratio of true and X-ray mass for one of the EdS models, in which the subclusters are in the process of merging. For each model, the three projection directions are shown in the same line style. The ordering from over- to underestimating the mass according to the magnetic field strengths (different line styles) is a result of the combination of all effects due to the magnetic field mentioned in the text. For this extreme merger the magnetic effects are considerably larger than the usual scatter of mass profiles from different projection directions. } \\label{fig:compbig9} \\end{figure}  In relaxed clusters magnetic fields  do not generally lead to any under- or overestimate of masses and they cannot be the dominant reason for the mass discrepancy between X-ray mass and lensing mass found in the central region of some clusters. The maximum difference of averaged mass profiles between models with high magnetic field and no magnetic field is 15\\%, while observational discrepancies between different methods of a factor of 2-3 are found. It is more likely that projection effects play an important role. As the gravitational lensing effect is sensitive to all the mass along the line-of-sight while the X-ray mass determines only the mass in the potential well, other mass concentrations along the line-of-sight can lead to a higher lensing mass compared to the X-ray mass (as probably seen in Cl0500-24 (Schindler \\& Wambsganss 1997)). Another effect can be that cooling flows lead to wrong determinations of temperatures when the temperature profile of the cluster is not available (Allen 1998). Also the $\\beta$ model does not always provide a good fit to the profile at all radii (Bartelmann \\& Steinmetz 1996). All these effects will be analysed in detail in a forthcoming paper."
    },
    "0010/astro-ph0010069_arXiv.txt": {
        "abstract": "We have studied whether the jet in a collapse-driven supernova can be a key process for the rapid-process (r-process) nucleosynthesis. We have examined the features of a steady, subsonic, and rigidly rotating jet in which the centrifugal force is balanced by the magnetic force. As for the models in which the magnetic field is weak and angular velocity is small, we found that the r-process does not occur because the final temperature is kept to be too high and the dynamical timescale becomes too long when the neutrino luminosities are set to be high. Even if the luminosities of the neutrinos are set to be low, which results in the low final temperature, we found that the models do not give a required condition to produce the r-process matter. Furthermore, the amount of the mass outflow seems to be too little to explain the solar-system abundance ratio in such low-luminosity models. As for the models in which the magnetic field is strong and angular velocity is large, we found that the entropy per baryon becomes too small and the dynamical timescale becomes too long. This tendency is, of course, a bad one for the production of the r-process nuclei. As a conclusion, we have to say that it is difficult to cause a successful r-process nucleosynthesis in the jet models in this study. ",
        "introduction": "\\label{intro} \\indent   It is one of the most important astrophysical problems that the sites where the rapid-process (r-process) nucleosyntesis occurs are not still known exactly. There are, at least, three reasons that make the study on r-process nucleosynthesis important. One of them is a very pure scientific interest. The mass numbers of the products of r-process nucleosynthesis are very high ($A$ = 80--250), which means that the most massive nuclei in the universe are synthesized through the r-process. You can guess easily that the situation in which the r-process nuclei are synthesized is a very peculiar one in the universe. We want to know where, when, and how the r-process nuclei are formed. Second reason is that some r-process nuclei can be used as chronometers. For example, the half-lives of $\\rm ^{232}Th$ and $\\rm ^{238}U$ are 1.405$\\times 10^{10}$ yr and 4.468$\\times 10^{9}$ yr, respectively. So if we can predict the mass-spectrum of the products of r-process nucleosynthesis precisely, we can estimate the ages of metal-poor objects which contain the r-process nuclei by observing its abundance ratio. Third reason is that some r-process nuclei can be used as tools of the study on the chemical evolution in our Galaxy (e.g., \\cite{ishimaru99}), which has a potential to reveal the history of the evolution of our Galaxy itself. Due to the reasons mentioned above, the study on the r-process nucleosynthesis is very important.   The conditions in which the r-process nucleosynthesis occurs successfully are (e.g., \\cite{hoffman97}): (i) neutron-rich ($n_n \\ge \\rm 10^{20} \\; cm^{-3}$), (ii) high entropy per baryon, (iii) small dynamical timescale, and (iv) small $Y_e$. This is because r-process nuclei are synthesized throuth the non-equilibrium process of the rapid neutron capture on the seed nuclei (iron-group elements). In other words, a highly dense, explosive, and neutron--rich site with high entropy will be a candidate for the location where the r-process nucleosynthesis occurs.   The most probable candidates for the sites are collapse-driven supernovae (e.g., \\cite{woosley94}) and/or neutron star mergers (e.g., \\cite{freiburghaus99}). This is because these candidates are thought to have a potential to meet the requirements mentioned above. However, we think that the collapse-driven supernovae are thought to be more probable sites than the neutron star mergers, because metal poor stars already contain the r-process nuclei (e.g., \\cite{freiburghaus99}). In fact, Mcwilliam et al. (1995) reported that the abundance of Eu can be estimated in 11 stars out of 33 metal-poor stars. These observations prove that r-process nuclei are produced from the early stage of the star formation in our Galaxy. Taking the event rate of collapse-driven supernovae ($10^{-2} \\; \\rm yr^{-1} \\; Gal^{-1}$; \\cite{vandenbergh91}) and neutron star merger ($10^{-5} \\; \\rm yr^{-1} \\; Gal^{-1}$; \\cite{vanden96}; \\cite{bethe98}) into consideration, collapse-driven supernovae are favored since they can supply the r-process nuclei from the early stage of the star formation in our Galaxy. Also, Cowan et al (1999) reported that the abundance ratio of r-process nuclei in metal poor stars are very similar to that in the solar system. This proves that r-process nuclei are synthesized through the similar conditions. This will be translated that, at least, most of the r-process nuclei are from one candidate. So we assume in this paper that most of the r-process nuclei are synthesized in the collapse-driven supernovae.   There are many excellent and precise analytic and/or numerical calculations on the r-process nucleosynthesis in the collapse-driven supernovae. However, it would be able to be said that there is no report that the r-process nuclei can be reproduced completely. For example, Takahashi et al. (1994) performed numerical simulations assuming Newtonian gravity and reported that entropy per baryon in the hot bubble is about 5 times smaller than the required value. Qian and Woosley (1996; hereafter QW96) also reported analytic treatments of the neutrino-driven winds from the surface of the proto-neutron star. At the same time, their analytical treatments are tested and confirmed by numerical methods. However, the derived entropy by their wind solutions are shown to fall short, by a factor of 2--3, of the value required to produce a strong r-process (\\cite{hoffman97}). Otsuki et al. (2000) have surveyed the effects of general relativity parametrically. They reported that r-process can occur in the strong neutrino-driven winds ($L_{\\nu} \\sim 10^{52}$ erg $\\rm s^{-1}$) as long as a massive ($\\sim$2.0 $M_{\\odot}$) and compact ($\\sim$ 10 km) proto-neutron star is formed. It is very interesting because such a solution can not be found in the frame work of Newtonian gravity (\\cite{qian96}). Such a solution is confirmed by the excellent numerical calculations (\\cite{sumiyoshi99}). However, the equation of state (EOS) of the nuclear matter has to be too soft to achieve such conditions. In fact, the r-process nuclei can not be produced in the numerical simulations with an normal EOS (\\cite{sumiyoshi99}).   There is only one report that r-process nucleosynthesis occurred successfully. That is the work done by Woosley et al (1994; here after WWMHM94). In their numerical simulation, the entropy per baryon becomes higher and higher as the time goes on. Finally, at very late phase of neutrino-driven wind ($\\sim 10$ s after the core-collapse), successful r-process occurs. However, there are some questions on their results. One is that the reason why the entropy per baryon becomes high at the late phase is unclear. In fact, when we adopt the analytic formulation of QW96, it is concluded that such a high entropy should not be obtained. It is true that general relativistic effects are included in WWMHM94, but such a high entropy could not be obtained in the work done by Otsuki et al (2000). So the discrepancy between WWMHM94 and QW96 can not be explained by the general relativistic effects alone. Also, WWMHM94 has a problem that the nuclei whose mass numbers are $\\sim$ 90 are much produced in the early stage of the neutrino-driven winds. So if we try to reproduce the mass-spectrum of the solar system abundances, we have to abandon the matter which is made at the early stage of the neutrino-driven winds. Even worse, the successful mass-spectrum at the late phase of neutrino-driven winds is destroyed when the reactions of neutral-current neutrino spallations of nucleons from $\\rm ^{4}He$ are taken into consideration (\\cite{meyer95}). In their paper, it is reported that (30--50)$\\%$ increase in the entropy is needed in order to restore the $A$ = 195 peak, which is extremely large modifications to the model. So, although the work done by WWMHM94 is the very remarkable and interesting one, it would not be concluded that the problem of r-process nucleosynthesis has been solved completely.   Due to the reason mentioned above, it will be natural to think that there may be an (some) effect(s) that will help the r-process nucleosynthesis. There are some papers that mention the effects of jet (\\cite{symbalisty84}; \\cite{shimizu94}; \\cite{nagataki00}) that is generated around the polar region of the highly rotating (the period $T\\sim$1ms) and/or magnetized ($\\sim 10^{15}$ G) proto-neutron star (\\cite{leblanc70}; \\cite{symbalisty84}; \\cite{shimizu94}; \\cite{yamada94}; \\cite{fryer99}). Since the physical conditions in the jet are different from those in a spherical explosion, the products of the nucleosynthesis are expected to be different. In fact, it is reported that the products of the explosive nucleosynthesis in a jet-like explosion in Si-- and O-- rich layer are much different from those in a spherical explosion (\\cite{nagataki97}; \\cite{nagataki00}). So it is natural to think that the products of explosive nucleosynthesis in the Fe core (that is, in the hot bubble) will be also changed due to the effects of the jet. This is the reason why we should examine whether the effects of the jet can help the synthesis of r-process nuclei or not.   Since there is no numerical simulation on the r-process nucleosynthesis in the jet during the neutrino-driven wind phase, our final goal is to perform such realistic numerical simulations. However, such a numerical simulation will be a heavy task. We also think that the results of such numerical simulations will not be understood completely without analytical studies. So, before performing such numerical simulations, we examine the physical conditions of the jet using a simple model. In this paper, we study a steady, subsonic, and rigidly rotating flow of the jet that is driven by neutrinos. We study the effects of the jet on the physical conditions such as the entropy per baryon and the dynamical timescale. Finally, we discuss whether the effects of the jet can help the synthesis of r-process nuclei.   In section~\\ref{model}, we explain the formulation for the jet in the hot bubble. In section~\\ref{result}, we show the results. Summary and discussions are presented in section~\\ref{summary}. ",
        "conclusions": "\\label{summary} \\indent   We have studied whether the jet in a collapse-driven supernova can be a key process for the r-process nucleosynthesis. We have studied the effects of a jet using a simple model, because the results of a realistic numerical simulation concerning with such a theme will not be understood completely without analytical study. Although our final goal is to perform such realistic numerical simulations, this work is a necessary process to understand the effects of a jet on the r-process nucleosynthesis.   We have studied two cases, that is, the cases in which the magnetic fields are weak/strong. In both cases, we assumed that the flow is steady and subsonic.   As for the weak field solutions, we have concluded that high-luminosity models can not work because $T_b$ becomes high and dynamical timescale becomes too long. Also, low-luminosity models do not present a required condition to produce the r-process nuclei in the framework of Newtonian gravity. Although a successful solution may be obtained when we include the effects of jet and general relativity at the same time, the total mass of the outflow seems to be too little to explain the solar-system abundance ratio in such low-luminosity models. On the other hand, we found that the entropy per baryon is small and the dynamical timescale is long in the strong field solution. We can tell that this tendency is a bad one for the production of r-process nuclei. The truth is that the solutions studied in this paper are inapplicable to the case of strong magnetic fields. However, similar tendency may be also found in the exact solutions. If so, we will be able to conclude that strong field solutions are unsuitable for r-process nucleosynthesis. As a conclusion, we have to say that it seems to be difficult to cause a successful r-process nucleosynthesis in the jet models in this study.   We have to emphasize that there are some assumptions in this study. So, we can not say that we have proved that a successful r-process nucleosynthesis does not occur in a neutrino-driven jet in a collapse-driven supernova. For example, we assumed that the flow is subsonic and there is no critical point, which is the common assumption in the previous studies on the r-process nucleosynthesis (\\cite{qian96}; \\cite{otsuki00}). However, we think that we do not need to restrict the solutions in such a way, that is, there may be a transition point at which Eqs.~(\\ref{eqn5}),~(\\ref{eqn9}),~(\\ref{eqn11}), and~(\\ref{eqn12}) break down. In most cases, the transition point will be a shock front. It means that the flow will gain entropy at the transition point, which will be a good sense to produce the r-process nuclei. The problem whether the flow contains transition points or not depends sensitively on the initial velocity on the surface of the proto-neutron star. So our final goal is to determine physically the velocity at the surface of the proto-neutron star. It means that the $\\dot{M}$ should be not given as an input parameter. It should be an output parameter. We have to investigate the mechanism for determining the outflow velocity at the surface of the proto-neutron star for further discussions. We also assumed that the flow is steady. As seen in subsection~\\ref{strong}, the density gradient becomes nearly zero and acceleration due to the strong magnetic fields does not occur in the framework of the steady flow. So, the situation may be different and acceleration occurs in the case of the unsteady flow. We should investigate the features of the unsteady jet flow. It will be investigated by numerical tests assuming a simple environment. We will perform such numerical tests in the near future. At the same time, we will try to perform realistic numerical simulations for the jet induced explosion in a collapse-driven supernova and seek a final answer whether the jet in a collapse-driven supernova is a key process for the r-process nucleosynthesis or not."
    },
    "0010/hep-ph0010240_arXiv.txt": {
        "abstract": "We discuss the flavor conversion of supernova neutrinos in the three-flavor mixing scheme of neutrinos. We point out that by neutrino observation from supernova one can discriminate the inverted hierarchy of neutrino masses from the normal one if $s_{13}^2 \\gsim \\text{a few} \\times 10^{-4}$, irrespective of which oscillation solution to the solar neutrino problem is realized in nature. We perform an analysis of data of SN1987A and obtain a strong indication that the inverted mass hierarchy is disfavored unless $s_{13}^2 \\lsim \\text{a few} \\times 10^{-4}$. ",
        "introduction": " ",
        "conclusions": ""
    },
    "0010/astro-ph0010225_arXiv.txt": {
        "abstract": "During supernova explosions, strange stars with almost bare quark surfaces may be formed. Under certain conditions, these stars could be rapidly spun down by the torque exerted by the fossil disks formed from the fall-back materials. They may also receive large kicks and hence, have large proper motion velocities. When these strange stars pass through the spherical ``Oort'' comet cloud formed during the pre-supernova era, they will capture some small-scale comet clouds and collide with some comet-like objects occasionally. These impacts can account for the repeating bursts as observed from the soft gamma repeaters (SGRs). According to this picture, it is expected that SGR 1900+14 will become active again during 2004-2005. ",
        "introduction": "Soft Gamma-ray Repeaters (hereafter SGRs) and Anomalous X-ray Pulsars (hereafter AXPs) are two groups of enigmatic sources. They share the following properties: 1. They all have long rotation periods (clustered within 5-12 seconds) and large spin-down rates (see, e.g. Mereghetti \\& Stella 1995\\markcite{ms95}; Kouveliotou et al. 1998, 1999\\markcite{k98}\\markcite{k99}); 2. Most of them are associated with supernova remnants, indicating that they are young objects (for reviews, see Hurley 1999\\markcite{h99}; Mereghetti 1999\\markcite{m99}); 3. No optical, infrared or radio counterparts have been identified (e.g. Eikenberry \\& Dror 2000\\markcite{ed00}; Lorimer \\& Xilouris 2000\\markcite{lx00}); 4. They all have soft persistent pulsed X-ray emission with luminosities of $L_{\\rm x} \\sim 10^{35}-10^{36}~{\\rm ergs~s^{-1}}$, well in excess of the spin down energy of these sources (e.g. Thompson 2000\\markcite{t00} for a review). The main difference between both types of the objects is that SGRs show occasional soft gamma-ray bursts while AXPs do not. It is also found that SGRs usually have larger proper motion velocities than AXPs according to their relative positions with respect to the cores of their supernova remnants (Hurley 1999). The main characteristics of the SGR bursts include: 1. Most of the bursts have super-Eddington luminosities with $L_{\\rm b}\\sim 10^{38}-10^{42}~{\\rm erg~s^{-1}}$; 2. The fluence distribution of the bursts is a power-law, and there is no correlation between the burst intensity and the time intervals between the bursts (G\\\"og\\\"us et al. 1999; 2000\\markcite{g99}\\markcite{g00}); 3. Two giant flares have been detected from SGR 0526-66 (the March 5, 1979 event) and SGR 1900+14 (the August 27, 1998 event), which share some common properties (see Thompson 2000 for a review); 4. Most bursts have soft spectra with characteristic energy around 20-30 keV.  The popular model for SGRs and AXPs is the magnetar model, which can account for almost all the phenomena listed above (Duncan \\& Thompson 1992\\markcite{dt92}; Thompson \\& Duncan 1995, 1996\\markcite{td95}\\markcite{td96}; Thompson 2000). However, the differences between SGRs and AXPs are not straightforwardly interpreted since these objects are not intrinsically different objects within the magnetar picture. It also remains unclear how some other issues, e.g., the non-systematic discrepancy between the characteristic ages derived assuming dipolar spindown and the ages of the associated supernova remnants, no clear positive dependence between $L_{\\rm x}$ and the polar surface field strength $B_{\\rm p}$, etc., can be properly addressed. On the other hand, a fossil-disk-accretion model for AXPs recently emerges from the independent studies by Chatterjee et al. (Chatterjee, Hernquist \\& Narayan 2000\\markcite{chn00}; Chatterjee \\& Hernquist 2000\\markcite{ch00}) and Alpar (1999, 2000)\\markcite{a99}\\markcite{a00}. The neutron stars in such a scenario have normal magnetic fields as the Crab pulsar. The model can interpret the AXP phenomenology well, but the bursts from the SGRs are difficult to interpret. On the observational ground, Marsden et al. (2000\\markcite{mlrh00}) observed that the SGRs and the AXPs are located in a much denser environment than the normal pulsars. They hence argue that the peculiar behaviors of the SGRs and AXPs may be due to their ``nurture'' from the environment rather than due to their special ``nature'' (i.e. magnetars) as compared with the normal pulsars. However, no plausible idea was proposed to connect the ``nurture'' to the phenomenology of these sources, especially the bursting behavior of the SGRs.  In this Letter, we attempt to propose a model to understand the bursting behavior of the SGRs without introducing the magnetar idea. We propose that the central objects of the SGRs are ``bare'' strange stars with normal magnetic fields ($10^{12}- 10^{13}$ G). We assume that these strange stars are born directly from supernova explosions from some massive progenitors, and they have experienced a spindown history as that having been proposed for the AXPs within the fossil disk model (Chatterjee et al. 2000; Alpar 2000). According to this model, some fallback materials from the supernova ejecta will form a fossil disk around the strange star. The SGRs/AXPs are just such strange/neutron stars that have experienced the ``propeller'' phase ($r_{\\rm c} \\ll r_{\\rm m} < r_{\\rm l}$), and are now in the ``tracking'' phase ($r_{\\rm c} \\lesssim r_{\\rm m} < r_{\\rm l}$) when infall of the materials onto the surface is possible and the star is X-ray bright. Here $r_{\\rm l}$, $r_{\\rm m}$ and $r_{\\rm c}$ are the light cylinder, the magnetospheric radius, and the corotating radius, respectively. In our picture, AXPs may be still neutron stars. We will attribute the SGR bursts to their occasionally collisions with some comet-like objects in the dense environment of the SGRs. We will show how various SGR properties as reviewed above could be accounted for within this picture. Our model differs from some other strange star SGR models (e.g. Alcock et al. 1986b; Cheng \\& Dai 1998; Dar \\& de Rujula 2000). ",
        "conclusions": "In this Letter, we propose that the peculiar behaviors of the SGRs are due to both their Nature (bare strange stars) and their Nurture (the Oort Cloud in the dense environment). Instead of invoking the magnetar hypothesis, we adopt the strange star hypothesis to interpret some interesting features of the SGRs. It is worth pointing out that the periodic activity does not depend on the nature of the central star. Although some authors argue that the bursting phenomenology (e.g. super-Eddington luminosity) can be also interpreted by colliding comets with a neutron star (e.g. Katz 1996), we think that a bare strange star is a cleaner interpretation due to the reasons discussed above. An important criterion to differentiate our model from the magnetar model is the activity period. If SGR 1900+14 will be turned on again in 2004, the magnetar model is then not favored, since it may be hard to find a mechanism to trigger the magnetic field decay instabilities periodically. If it turns out that some problems (e.g. super-Eddington luminosity, baryonic contamination, and large proper motion velocity) are not solvable within the neutron star impacting model, the bursts from SGR 1900+14 in time then present a support to the strange star hypothesis and will bring profound implications for fundamental physics.  According to this picture, there might be some other bare strange stars which may also have super-Eddington bursts when they collide with comet-like objects. However, they must have passed through the Oort Cloud and/or in a much less dense environment, so that the chance to detect repeating bursts is rare. Single bursting events are possible and they may account for a small portion of the short, soft bursts in BASTE data. The association of SGR 1806-20 with a radio plerion may not be compatible with the present picture, but recent results indicate that the non-thermal radio core of the supernova remnant G10.0-0.3 may be associated with another luminous blue variable rather than with the SGR (Hurley 1999)."
    },
    "0010/astro-ph0010013_arXiv.txt": {
        "abstract": "We have mapped the thermal emission line of SiO ($v$ = 0; $J$ = 2--1) associated with the quadrupolar molecular outflow driven by the very cold far-infrared source IRAS 16293$-$2422. The SiO emission is significantly enhanced in the northeastern red lobe and at the position $\\sim$50$''$ east of the IRAS source. Strong SiO emission observed at $\\sim$50$''$ east of the IRAS source presents evidence for a dynamical interaction between a part of the eastern blue lobe and the dense ambient gas condensation, however, such an interaction is unlikely to be responsible for producing the quadrupolar morphology. The SiO emission in the northeastern red lobe shows the spatial and velocity structure similar to those of the CO outflow, suggesting that the SiO emission comes from the molecular outflow in the northeastern red lobe itself. The observed velocity structure is reproduced by a simple spatio-kinematic model of bow shock with a shock velocity of 19--24 km s$^{-1}$ inclined by 30--45$^{\\circ}$ from the plane of the sky. This implies that the northeastern red lobe is independent of the eastern blue lobe and that the quadrupolar structure is due to two separate bipolar outflows.  The SiO emission observed in the western red lobe has a broad pedestal shape with low intensity. Unlike the SiO emission in the northeastern red lobe, the spatial extent of the SiO emission in the western red lobe is restricted to its central region. The spatial and velocity structures and the line profiles suggest that three different types of the SiO emission are observed in this outflow; the SiO emission arises from the interface between the outflowing gas and the dense ambient gas clump, the SiO emission coming from the outflow lobe itself, and the broad SiO emission with low intensity observed at the central region of the outflow lobe. ",
        "introduction": "Thermal emission from the SiO molecule is observed in shocked regions associated with molecular outflows from young stellar objects, in which the SiO abundance is enhanced by several orders of magnitude (e.g. \\markcite{Bac91,Mik92,Mar92}Bachiller, Mart\\'{\\i}n-Pintado, \\& Fuente 1991; Mikami et al. 1992; Mart\\'{\\i}n-Pintado, Bachiller, \\& Fuente 1992). Such an abundance enhancement of SiO is interpreted as the release of Si or SiO from dust grains by means of strong shocks (e.g. \\markcite{Mik92,Mar92}Mikami et al. 1992; Mart\\'{\\i}n-Pintado et al. 1992). Since the SiO emission has been seldom detected in quiescent dark clouds because of its very low abundance (of the order of 10$^{-12}$; \\markcite{Ziu89,Mar92}Ziurys, Friberg, \\& Irvine 1989; Mart\\'{\\i}n-Pintado et al. 1992), the SiO emission is generally believed to be a good tracer of shocked molecular gas.  Strong SiO emission observed in low-mass star forming regions is mainly associated with outflows driven by very cold far-infrared sources referred to as the ^^ ^^ Class 0$\"$ sources (e.g., \\markcite{Bac92, Bac96}Bachiller \\& G\\'{o}mez-Gonz\\'{a}lez 1992; Bachiller 1996). On the basis of high-spatial resolution mappings of the SiO lines, it has been revealed that both spatial and velocity structures of the shocked gas are closely related to the history and an ejection mechanism of the outflows. The representative cases are the outflows driven by the Class 0 sources in the L1157 and L1448 dark clouds; the SiO maps present a picture that these outflows are driven by  collimated and episodic jets ejected from the central star (\\markcite{Zha95, Gue98, Dut97}Zhang et al. 1995; Gueth, Guilloteau, \\& Bachiller 1998; Dutrey, Guilloteau, \\& Bachiller 1997). Therefore, mapping of the SiO emission is expected to provide an opportunity to study the origin of the molecular outflows with complex morphology.  In this paper, we present the map of the SiO ($v$ = 0;$J$ = 2--1) emission in the quadrupolar outflow associated with IRAS 16293$-$2422 (hereafter referred to as I16293). I16293 is known to be a proto-binary system with a projected separation of 840 AU (\\markcite{Mun92,Wal93}Mundy et al. 1992; Walker, Carlstrom, \\& Bieging 1993) embedded in the nearby (160 pc) $\\rho$ Ophiuchi molecular cloud. This outflow consists of two pairs of lobes; the brighter one extending along the east-west direction and the other one along the northeast-southwest direction (\\markcite{Wal99, Miz90}Walker et al. 1988; Mizuno et al. 1990). To explain the conspicuous quadrupolar morphology of this outflow system, \\markcite{Wal93}Walker et al. (1993) proposed that the two pairs of lobes correspond to the two independent bipolar outflows driven by two independent sources. On the other hand, \\markcite{Miz90}Mizuno et al. (1990) showed  that the outflow is dynamically interacting with the dense ambient gas clump, and suggested that a single outflow lobe could be split into two lobes by the interaction. To examine which idea is plausible, and to probe the effect of the complex outflow on its surroundings, we investigate  the spatial and kinematic structure of the shocked gas traced by the SiO emission. ",
        "conclusions": "\\subsection{Dynamical Interaction between the Outflow and the Dense Gas Clump}  As mentioned in the previous section, the SiO emission in the E1 clump is likely to arise from the region where the outflow blown out from the protostar impinge on the dense ambient gas clump 16293E. Since the line profile observed in the E1 clump that shows a strong cut-off at the ambient velocity and a gradual redshifted wing (Figure 3b), it is suggested that the SiO emission in the E1 clump arises from the quiescent material accelerated by the shock. The dynamical interaction of the outflow with 16293E was also pointed out by \\markcite{Miz90}Mizuno et al. (1990). They suggested that the blueshifted gas blown eastwards from its driving source is interacting with 16293E, and that such an interaction splits a single blueshifted lobe into the NER lobe and the eastern blue (EB) lobe, resulting in the quadrupolar morphology. However, the observed SiO emission is not blueshifted but redshifted; this clearly contradicts the above scenario.  The redshifted SiO emission in the E1 clump suggests a possibility that the redshifted gas in the NER lobe is blown out to the eastward from its driving source and changes its direction to the northeast by means of a dynamical interaction with 16293E. In this case, the velocity field of the dense gas in 16293E is expected to be disturbed by the momentum provided by the outflow. On the basis of the CO observations, \\markcite{Miz90}Mizuno et al. (1990) estimated the momentum of the outflowing gas in the NER lobe to be 0.5 $M_{\\odot}$ km s$^{-1}$. Since the mass of 16293E is estimated to be $\\sim$1.6 $M_{\\odot}$ from the H$^{13}$CO$^+$ data by assuming the local thermodynamical equilibrium (LTE) condition with the excitation temperature $T_{\\rm ex}$ = 15 K and optically-thin H$^{13}$CO$^+$ emission, the gas of 16293E is expected to show the velocity shift of $\\sim$0.3 km s$^{-1}$ toward the red from the rest of the cloud. However, there is no evidence of such a velocity shift in the H$^{13}$CO$^+$ data. Therefore, we consider that the change of outflow direction by means of a dynamical interaction is unlikely. In addition, the velocity field observed in the E2 clump is explained by the bow shock impinging on the ambient medium (see {\\S}4.2), suggesting that the outflowing gas in the NER lobe is blown out to the northeast direction from its origin and does not change its direction on its way to the E2 position.  Then, which outflow component is interacting with 16293E and is responsible for the strong SiO emission at the E1 position? A recent interferometric CO ($J$ = 1--0) map has shown that the east-west pair of lobes, which corresponds to the EB and WR lobe pair, has an axis close to the plane of the sky and consists of fan-shaped lobes, each of which contains both blueshifted and redshifted components (Hirano et al. in preparation). This implies that the EB lobe contains a considerable amount of redshifted gas in the far side of the lobe. Therefore, we consider that the redshifted gas in the EB lobe is interacting with 16293E and is producing the redshifted SiO emission observed at E1.   \\subsection{Simple Kinematic Model of the NER Lobe}  As shown in Figure 1, the spatial distribution of the SiO clump E2 resembles that of the NER lobe in the single-dish CO maps. In particular, the map of the high-velocity SiO emission is quite similar to that of the high-velocity CO $J$ = 2--1 emission presented by \\markcite{Cec98}Ceccarelli et al. (1998). Channel maps presented in Figure 2 reveal that the high-velocity SiO emission in the E2 clump increases its velocity downstream. Such a velocity increase downstream is also observed in the CO ($J$ = 2--1) emission in the NER lobe (\\markcite{Wal88}Walker et al. 1988). These results indicate that the high-velocity SiO emission in the E2 clump has both spatially and kinematically similar characteristics to the CO outflow in the NER lobe, suggesting that this SiO emission comes from the NER lobe itself. Unlike the CO emission, the SiO emission in the E2 clump has another velocity component whose velocity is close to the cloud systemic velocity. The presence of two velocity components is explained if the low- and the high-velocity SiO emission components come from the near side and the far side of the NER lobe, respectively. Such two velocity components corresponding to the near sides and the far sides of the lobes were also seen in the position-velocity diagrams of the CO emission in the Mon R2 outflow (\\markcite{Mey91}Meyers-Rice \\& Lada 1991). In the case of the NER lobe in the I16293 outflow, the low-velocity component has not been identified in the CO maps because it may be screened by overwhelming emission from the quiescent ambient gas.  In Figure 5, we show the position-velocity diagram of the SiO emission along the major axis of the NER lobe (P.A. = 60$^{\\circ}$). The SiO emission in the NER lobe increases its velocity dispersion as an increasing distance from I16293, and the largest velocity dispersion is observed toward slightly downstream of the E2 peak. As pointed out by \\markcite{Ben96}Bence, Richer, \\& Padman (1996), such a velocity feature is explained if the SiO emission arises from the bow shock generated by the high-velocity jet impinging onto the ambient medium. Recently, several authors have proposed the models that the molecular outflows are created through the propagation of the large bow shocks (e.g. \\markcite{Rag93,Mas93}Raga \\& Cabrit 1993; Masson \\& Chernin 1993). In the following, we use a simple kinematical model of the bow shock proposed by \\markcite{Har87}Hartigan, Raymond, \\& Hartmann (1987) and compare the velocity features derived from the model with those observed.  The geometry of the bow shock is schematically depicted in Figure 6. We assume that the bow shock has axial symmetry around the jet axis $z$ with a shape described by a function $z(r) = Ar^2 + Br^4$, where $r$ is the distance from the jet axis, and $A$ and $B$ are numerical constants. In the frame of the exciting source, shock is propagating at a velocity of $V_{\\rm s}$ into the ambient pre-shock medium moving at $V_{\\rm amb}$. In the frame of the bow shock, material impacts the shock surface with velocity of $V_{\\rm s} - V_{\\rm amb}$. Since the velocity component normal to the surface of the bow shock $v_{\\rm n} = (V_{\\rm s} - V_{\\rm amb}) {\\rm cos}{\\xi}$ diminishes because of quick energy release via radiation, only the velocity component parallel to the surface $v_{\\rm p} = (V_{\\rm s} - V_{\\rm amb}) {\\rm sin}{\\xi}$ is conserved. Therefore, the post shock velocity is $v_2 = (V_{\\rm s} - V_{\\rm amb}) {\\rm sin} {\\xi}$, and its angle is ${\\theta} = ({\\pi}/2) - {\\xi}$. The radial velocity $V_{\\rm r}$ in the frame of the exciting source is given by \\begin{equation} V_{\\rm r} = (V_{\\rm s} - V_{\\rm amb}) {\\rm sin}{\\xi} {\\rm sin} ({\\xi} {\\pm} {\\phi}) - V_s {\\rm cos} {\\phi}, \\end{equation} where $\\phi$ is the angle between the line of sight and the jet axis, and the + and $-$ signs refer to the top and bottom halves of the curve in Figure 6, respectively. As demonstrated by \\markcite{Har87}Hartigan et al. (1987), the total velocity dispersion $\\Delta V$ of the bow-shock feature and the median velocity $V_{\\rm median}$ are given by $\\Delta V = V_{\\rm s} - V_{\\rm amb}$ and $V_{\\rm median} = - [(V_{\\rm s} + V_{\\rm amb})/2] \\times {\\rm cos}{\\phi}$, respectively. In the case of the NER lobe, the total velocity dispersion is $\\Delta V {\\sim}$16 km s$^{-1}$, and $V_{\\rm median}$ measured in the frame of the exciting source ($V_{\\rm LSR} \\approx$4 km s$^{-1}$) is $\\approx$8 km s$^{-1}$. To estimate the inclination angle $\\phi$, we compare the radial velocity distribution calculated from equation (1) with the observed position-velocity diagram. If we assume that the bow shock has a shape described as $z(r) = 0.42 r^2 + 0.136 r^4$, which is designated as modified Raga shape in \\markcite{Har87}Hartigan et al. (1987), the observed feature is reproduced with $\\phi{\\sim}135^{\\circ}$ (i.e. the angle between the outflow axis and the plane of the sky {\\it i} is $\\sim$45$^{\\circ}$), $V_{\\rm s}$ = 19.3 km s$^{-1}$, and $V_{\\rm amb}$ = 3.3 km s$^{-1}$ (solid curve in Figure 5). In the case of the bow shock with a parabolic shape, $z = r^2$, we obtained $\\phi {\\sim} 120^{\\circ}$ ({\\it i} $\\sim$30$^{\\circ}$), $V_{\\rm s}$ = 24.0 km s$^{-1}$, and $V_{\\rm amb}$ = 8.0 km s$^{-1}$ (dash-dotted curve in Figure 5). Although the parameters depend on the assumed shape of the bow-shock surface, the observed velocity features can be explained if the axis of the flow is inclined by 30--45$^{\\circ}$ from the plane of the sky and the shock velocity is $\\sim$19--24 km s$^{-1}$. The dynamical time scale of the E2 shock is estimated to be (5--7) $\\times$ 10$^3$ yr. It should be noted that the presence of the low-velocity component which is close to the cloud systemic velocity is a natural consequence in the context of the bow-shock outflow models.  As shown in Figure 5, the velocity pattern observed in the NER lobe is reproduced by the simple bow-shock model, as in the case of the SiO emission observed toward the red lobe of the bipolar outflow in the L1448 dark cloud (\\markcite{Dut97}Dutrey et al. 1997). This implies that the NER lobe is one of the bipolar outflow lobes and is not a splinter of the EB lobe.  The bow-shock model cannot explain the high-velocity emission of $V_{\\rm LSR} {\\sim}$18 km s$^{-1}$ observed farther downstream of the E2 peak (R.A. offset $>$ 120$''$). Since the E2 peak is not located toward the end but toward the middle of the NER lobe, the bow shock at the E2 peak is unlikely to be a terminal shock. Therefore, the NER lobe could have two-shock structure created by episodic ejection events as in the case of the blueshifted lobe of the L1157 outflow (e.g. \\markcite{Zha95,Gue98}Zhang et al. 1995; Gueth et al. 1998). To assess this possibility, higher-resolution observations with interferometer are required.  \\subsection{Three Types of SiO Emission Observed in the I16293 Outflow}  As discussed in the previous two subsections, the properties of the E1 and E2 clumps are different from each other. The SiO emission in the E1 clump arises from the interface between the outflow and the dense ambient gas clump, and that in the E2 clump is likely to come from the outflow lobe itself.  The properties of the SiO emission observed at W2 appear to be similar to those of the emission from E1, because the position of W2 corresponds to the periphery of the WR lobe at which the outflowing gas faces the ambient medium. The line profile of the SiO emission observed at W2 implies that most of the shocked gas has the velocity close to the ambient velocity.  On the other hand, the SiO line profile observed at W1, where the high-velocity CO emission is strongest in the WR lobe, shows a broad pedestal shape with its width measured at 1$\\sigma$ level is 12 km s$^{-1}$. Unlike the NER lobe, the CO emission in the WR lobe shows no systematic displacement of emission peaks with increasing velocity, and has a peak at the position close to that of W1 at all velocities (\\markcite{Wal88}Walker et al. 1988). The broad width of the SiO emission is likely to reflect the large velocity dispersion of the outflowing gas along the line of sight toward the W1 position. Unlike the E2 clump, the spatial extent of the SiO emitting region in the W1 is restricted to the small area of $\\sim$20$''$ in size, which is much smaller than the size of the WR lobe observed in the CO. Such a broad SiO emission with low intensity is also observed in the central region of the outflows in S140 and IC1396N (Hirano et al. in preparation).  To summarize, in the I16293 outflow, the SiO emission lines having three different properties are observed. The first one is the enhanced SiO emission at the interface between the outflowing gas and the dense ambient gas clumps (E1 and W2), which shows the line profile peaks at the velocity close to the cloud systemic velocity. The second one is the SiO emission showing similar spatial extent and kinematic structure as the CO outflow, and considered to arise from the outflow lobe itself (E2). The third one is the broad SiO emission with low intensity observed toward the central region of the outflow lobe (W1).  \\subsection{SiO Abundances}  To estimate the SiO abundance in the shocked gas, we assumed optically thin emission, and calculated the column densities of the SiO molecules by assuming the LTE condition. The permanent dipole moment of the SiO molecule is employed to be 3.1 Debye. In the following, we assumed the excitation temperature $T_{\\rm ex}$ to be 15 K, which is almost the same as the rotational temperature estimated from the NH$_3$ (1, 1) and (2, 2) transitions (\\markcite{Miz90}Mizuno et al. 1990). It is possible that the excitation temperature of the SiO in the shocked gas is higher than that assumed here, because recent observations of high-excitation ammonia have revealed that the shocked gas in several outflows is heated to more than 100 K (e.g. \\markcite{Bac93,Ume99}Bachiller, Mart\\'{\\i}n-Pintado, \\& Fuente 1993; Umemoto et al. 1999). If we adopt 100 K as an excitation temperature, the column density of the SiO becomes a factor of three higher than that calculated for $T_{\\rm ex}$ = 15 K.  By using the total integrated intensities (from $V_{\\rm LSR}$ = 2 to 20 km s$^{-1}$), the SiO column densities, $N$(SiO), at the E1 and E2 positions are derived to be 4.3$\\times$10$^{13}$ cm$^{-2}$ and 8.4$\\times$10$^{13}$ cm$^{-2}$, respectively, whereas it is estimated to be $\\simeq  2.0\\times 10 ^{13}$ cm$^{-2}$ at both the W1 and W2 positions. At the IRAS position, we obtain  $N$(SiO) to be $1.0\\times 10 ^{13}$ cm$^{-2}$. When we adopt an H$_2$ column density averaged over the core of 0.2 pc in radius, 3.6$\\times$10$^{22}$ cm$^{-2}$, derived from the C$^{18}$O observations of \\markcite{Miz90}Mizuno et al. (1990), the SiO abundance $X$(SiO) is estimated  to be 1.2$\\times$10$^{-9}$ toward the E1 peak, 5.2$\\times$10$^{-10}$ toward W1 and W2, and 2.8$\\times$10$^{-10}$ at the IRAS position. It should be noted that the abundance values estimated here are the beam-averaged values. If the SiO emission observed at the IRAS position comes from the small area of $<$3$''$ size, the $X$(SiO) becomes a factor of $\\sim$50 higher than that estimated here, which is consistent with the SiO abundance value derived by \\markcite{Cec00}Ceccarelli et al. (2000) from their LVG analysis.  To estimate the SiO abundance in the outflowing gas in the NER lobe, we compare the  column density of the high-velocity SiO (from 8 to 20 km s$^{-1}$ ) with that of H$_2$ derived from the CO data observed by \\markcite{Miz90}Mizuno et al. (1990). At the E2 peak, the $N$(SiO) and $N$(H$_2$) of the high-velocity gas are estimated to be 7.3$\\times$10$^{13}$ cm$^{-2}$ and 8.8$\\times$10$^{20}$ cm$^{-2}$, respectively. We thus obtain $X$(SiO) in the high-velocity gas to be 8.3$\\times$10$^{-8}$ at the E2 peak. This indicates that the SiO abundance in the NER lobe is comparable to that observed in the blueshifted lobe of the L1157 outflow (\\markcite{Mik92}Mikami et al. 1992) and is enhanced by a factor of $\\sim$10$^{4}$ as compared to the upper limits in the dark clouds (\\markcite{Zir89}Ziurys et al. 1989). If we assume that all the Si atoms (Si/H$_2$ $\\sim$7$\\times$10$^{-5}$) is depleted on grains in the unperturbed component, at least $\\sim$0.1 \\% of Si in the interacting region should be released in the gas phase in order to explain the SiO abundance in the NER lobe."
    },
    "0010/astro-ph0010539_arXiv.txt": {
        "abstract": "This article summarizes the structural parameters of the Galactic bulge and disk, and discusses the interpretation of the bulge microlensing observations and the determination of the Milky Way's luminous mass from the terminal velocity curve and the Oort limit. The bulge is a rotating bar with corotation radius around 4 kpc.  The NIR disk has a short scale-length.  The measured surface density of the local disk is in good agreement with the prediction of a maximum NIR disk model for the Milky Way.  The preliminary new value for the microlensing optical depth of clump giant sources is within $1.7\\sigma$ of the prediction by the maximum NIR disk model, while the optical depth for all sources is still significantly higher. These results imply that cold dark matter cannot dominate inside the solar radius. ",
        "introduction": "\\label{secintro}  The Milky Way is a normal galaxy seen under a magnifying glass: the ages, metallicities, and kinematics of its stars can be observed in much greater detail than in any other galaxy. Combined with a model for the large-scale structure and potential, dynamical models for specific stellar populations can be constructed which link our local neighbourhood with distant parts of the Galaxy. Thus we can study in a unique way the imprints of the evolution and formation processes that shaped the Milky Way, and presumably other galaxies as well.  As is well-known, our viewpoint from within the Galactic disk also has some drawbacks, in that many of the Milky Way's large scale properties have been difficult to obtain. Only in the last decade has it become possible to map the old stellar population in the Galactic bulge and inner disk, using infrared observations to penetrate the dust layers in the disk. Models are needed to interpret these data and link them to gas kinematical and other observations. Two important results are that the inner Galaxy contains a rotating bar/bulge, and that the scale-length of the disk is significantly shorter than previously assumed.  Microlensing surveys have now found hundreds of events towards the Galactic bulge. From such data and from local stellar-kinematic measurements, direct constraints on the mass of the Milky Way's bulge and disk can be obtained, which are difficult to come by in external galaxies, and which can be used to test cosmological models of dark matter halo and galaxy formation. ",
        "conclusions": ""
    },
    "0010/astro-ph0010363_arXiv.txt": {
        "abstract": "We have made high-resolution, high-sensitivity dynamic spectra of a sample of strong pulsars at 430~MHz with the Arecibo radiotelescope. For 4 pulsars we find faint but sharply delineated features in the secondary spectra.  These are examples of the previously observed ``crisscross'' or ``multiple drift slope'' phenomenon presumed to be due to multiple imaging of the pulsar by the interstellar medium.  The unprecedented resolution and dynamic range of our observations allow a deeper level of analysis. Distances to the dominant scattering screen along the line of sight are determined and are shown to agree well with those inferred from other scintillation phenomena. Multiple imaging of the pulsar by the ISM is required. A compact central image surrounded by a faint scattering halo, roughly circularly symmetric, is consistent with the data. Scattering from filaments may also be consistent. The angular extent of the scattering material parallel to the direction of the pulsar velocity is $\\sim 5$~mas, corresponding to a linear extent of $\\sim 2$~AU. Further observations of these features should allow better discrimination between models and an identification of the scattering structures. ",
        "introduction": "Scattering by density inhomogeneities in the ionized interstellar medium causes multiple rays to interfere, producing frequency structure in the spectra of spatially coherent sources such as pulsars \\cite{sch68,ric90}. The relatively high transverse velocities of most pulsars cause the spectra to change significantly on timescales of minutes to hours. When the spectrum is monitored over time, the resulting two-dimensional array or ``dynamic spectrum''  is often dominated by a random pattern of scintillation maxima with characteristic bandwidth and timescale \\cite{cwb85}. On occasion, however, more organized features are present. The best studied of these are periodic fringe patterns in the dynamic spectra due to the interference between two or more discrete bundles of rays or, equivalently, rays arriving from multiple images on the sky \\cite[ hereafter RLG]{cw86,wc87,gbr99,rlg97}.  A less well-studied pattern in pulsar dynamic spectra consists of a crisscross of intensity maxima that are variably spaced perpendicular to the ridge lines of the maxima. Cordes and Wolszczan (1986) drew attention to these ``multiple drift rates'' in dynamic spectra, building on earlier work (Hewish 1980; Roberts \\& Ables 1982; Hewish, Wolszczan, \\& Graham 1985; Cordes, Weisberg, \\& Boriakoff 1985). \\nocite{hew80} \\nocite{ra82} \\nocite{hwg85} \\nocite{cwb85} Numerous examples of the phenomenon are evident in the dynamic spectra presented by Gupta, Rickett, \\& Lyne (1994) in their multi-year study of 8 pulsars, \\nocite{grl94} but the autocorrelation analysis that they employed is not well suited to its full elucidation. The remarkable episode of ``fringing'' in the dynamic spectra of PSR~B0834+06 that RLG explore is also accompanied at all epochs by this phenomenon.  The central result of our study is to show that this crisscross pattern has a much clearer representation in transform space, where it is seen to be a high-Q phenomenon with a characteristic parabolic curvature. We discuss the constraints that the presence of this pattern places on the image geometry and speculate on the physical structures that might give rise to this scattering geometry. ",
        "conclusions": ""
    },
    "0010/astro-ph0010649_arXiv.txt": {
        "abstract": "{ We report on the {\\it XMM-Newton} observations of the BL Lac object MS~0737.9+7441 during the performance verification phase. A simple power--law fit provides an adequate description of the integrated spectrum in the 0.2--10 keV energy band. The photon index  is slightly steeper in the EPIC pn data with $\\Gamma = (2.38 \\pm 0.01)$ compared to the EPIC MOS data ($\\Gamma = (2.28 \\pm 0.01)$). The difference is most probably due to the present uncertainties in the calibration of the EPIC MOS and EPIC pn data sets. We report evidence for intrinsic absorption in the distant BL Lac above the Galactic column ($N_{\\rm H, Gal} = 3.2 \\cdot 10^{20} \\rm{cm^{-2}}$) which is $N_{\\rm H, fit}^{\\rm z=0.315} \\rm = (2.70 \\pm 0.20) \\times 10^{20}\\ cm^{-2}$ in the EPIC pn data and $N_{\\rm H, fit}^{\\rm z=0.315} \\rm = (3.25 \\pm 0.25) \\times 10^{20}\\ cm^{-2}$ in the EPIC MOS data assuming neutral gas and solar abundances. The flux  variations are found to be of the order of 10 \\%. No significant spectral variability is detected. ",
        "introduction": "The BL Lac object MS 0737.9+7441 was discovered in the {\\it Einstein} Observatory Extended Medium-Sensitivity Survey (EMSS; Gioia et al. 1990, Stocke et al. 1991) with a flux of $\\rm f_X = (9.8 \\,\\pm 0.6) \\cdot 10^{-12} \\ erg \\ cm^{-2} \\ s^{-1}$ in the energy range between 0.3 and 3.5 keV. Its redshift is z = 0.315 (Morris et al. 1991). In the ROSAT All-Sky-Survey observations the source had a count rate of 0.49 counts s$^{-1}$ during an exposure of 456 seconds. Perlman et al. (1996) obtained a best fit power--law photon index of $\\rm \\Gamma = 1.91$ from the observed hardness ratios assuming Galactic absorption $N_{\\rm H, Gal} = 3.2 \\cdot 10^{20} \\rm{cm^{-2}}$. Lamer et al. (1996) examined pointed ROSAT-PSPC data of MS 0737.9+7441 with a count rate of 0.51 $\\pm$ 0.01 counts per second for an exposure of 8782 seconds. Using a power--law model the best fit parameters are  $N_{\\rm H, fit} = (4.16 \\pm 0.48) \\cdot 10^{20} \\rm{cm^{-2}}$ and $\\Gamma = 2.39 \\pm 0.11$. In a {\\it BeppoSAX} observation (Wolter et al. 1998) MS 0737.9+7441 was detected in the LECS instrument with 37.1 $\\pm$ 7.8 net counts during an exposure of 3075 seconds, the MECS detector net  counts were 735.9 $\\pm$ 30.6 in 23279 seconds. By assuming a simple power law the best fit parameters are $\\Gamma = 2.53^{+0.28}_{-0.23}$ and $N_{\\rm H, fit} = (25.8^{+49.3}_{-21.6}) \\cdot 10^{20} \\rm{cm^{-2}}$. A broken power law resulted  in photon indices of $\\Gamma{_1} = 1.17$ (which they classify as uncertain) and $\\Gamma{_2} = 2.43^{+0.18}_{-0.16}$. The  break energy is 1.05 (1.27--1.61) keV.  In the following we report on the {\\it XMM-Newton} observation of MS~0737.9+7441 obtained during the performance verification program. Two exposures of approximately 20 ksec and 60 ksec were performed on the source between April 12, 2000 and April 13, 2000, either side of the orbital apogee gap. Around the middle of the second PN exposure the camera suffered a short telemetry break. The PN camera was in full frame mode throughout the observation. Both MOS cameras were in large window ($300 \\times 300$ pixels) mode. All cameras employed their respective thin--1 filters.  The Hubble parameter was chosen to be $H_0 = \\rm 70\\ km\\ s^{-1}\\ Mpc^{-1}$ and a cosmological deceleration parameter of $q_0 = \\rm \\frac{1}{2}$ have been adopted throughout the paper. ",
        "conclusions": "XMM-Newton observations reveal that a featureless simple power law model with absorption by neutral hydrogen provides an adequate fit to the data. The photon indices as measured with the EPIC pn and EPIC MOS are slightly different with 2.38 $\\pm$ 0.01 and 2.28 $\\pm$ 0.01, respectively, most probably due to present calibration uncertainties for the EPIC detectors. The spectral residua between 0.5 and 0.7 keV might be attributed to uncertainties in the presently available EPIC pn response matrix not allowing us to make definitive statements on the possible presence of soft X-ray emission lines, e.g. emission due to O~VIII (653 eV) and Fe~XXVII (726 eV) would fall in this energy range. We confirm the presence of intrinsic absorption in the source, which is about $N_{\\rm H, fit} = 3.0 \\cdot 10^{20}\\ \\rm cm^{-2}$ in the source frame (($2.70 \\pm 0.20) \\cdot 10^{20}\\ \\rm cm^{-2}$ for EPIC pn  and ($3.25 \\pm 0.25) \\cdot 10^{20}\\ \\rm cm^{-2}$ for EPIC MOS). The soft X-ray absorption detected in MS 0737.9+7441 appears to be fairly similar to other studies. Beckmann and Wolter (2000) found a mean value for the intrinsic absorption at soft X-ray energies of about $N_{\\rm H, fit} = \\rm 1.0 \\cdot 10^{20}\\ \\rm cm^{2}$ with the largest value of $1.0 \\cdot 10^{21}\\ \\rm cm^{-2}$. The intrinsic absorption found in the host galaxy of the BL Lac MS 0737.9+7441 is also similar to the soft X-ray absorption of radio-quiet active galaxies (cf. Table 1 of Boller, Brandt \\& Fink 1996 and Table 2 of Walter \\& Fink 1993). The flux variations of about 10 \\% are relatively small for a BL Lac object (compare Giommi et al. 1990). No significant spectral variations are detected during the observations."
    },
    "0010/astro-ph0010155_arXiv.txt": {
        "abstract": "We report on the results of  spectral analysis of the dipping source XB~1254--690 observed by the BeppoSAX satellite. We find that the X-ray dips are not present during the BeppoSAX observation, in line with recent RXTE results. The broad band (0.1--100 keV) energy spectrum is well fitted by a three-component model consisting of a multicolor disk blackbody with an inner disk temperature of $\\sim 0.85$ keV, a comptonized spectrum with an electron temperature of $\\sim 2$ keV, and a  bremsstrahlung at a temperature of $\\sim 20$ keV. Adopting a distance of 10 kpc and taking into account a spectral hardening factor of $\\sim 1.7$ (because of electron scattering which modifies the blackbody spectrum emitted by the disk) we estimated that the inner disk radius is $R_{\\rm in} \\sqrt{\\cos i} \\sim 11$ km, where $i$ is the inclination angle of the system with respect to the line of sight. The comptonized component could originate in a spherical corona or boundary layer, surrounding the neutron star, with  optical depth $\\sim 19$. The bremsstrahlung emission, contributing  $\\sim 4\\%$ of the total luminosity, probably  originates in an extended accretion disk corona with radius $\\sim 10^{10}$ cm. In this scenario we calculated that the optical depth of this region is $\\sim 0.71$ and its mean electron density is $N_e \\sim 2.7 \\times 10^{14}$ cm$^{-3}$. This last component might also be present in other low mass X-ray binaries, but, because of its low intrinsic luminosity, it is not easily observable. We also find an absorption edge at $\\sim 1.27$ keV with an optical depth of $\\sim 0.15$. Its energy could correspond to the L-edge of Fe XVII, or K-edge of Ne X or neutral Mg. ",
        "introduction": "About 10 Low Mass X-ray Binaries (LMXB) are known to show periodic dips in their X-ray light curves and most of them also show X-ray burst activity. The dip intensities, lengths and shapes  change from source to source, and, for the same source, from cycle to cycle. Most of the  characteristics of the dips are consistent with a model in which the central X-ray source is temporarily obscured by material in the outer edge of the accretion disc. This  thicker region could be formed by the impact of the gas stream from the Roche-lobe filling companion star on the outer rim of the accretion disc.  Among the dipping sources, XB~1254--690, that also shows type-I X-ray bursts, is peculiar because it showed a cessation of its regular dipping activity (Smale \\& Wachter 1999).  This source was observed for the first time by HEAO 1 A2 instrument (Mason et al. 1980) when an optical burst of $\\sim 20$ s duration occurred. The optical counterpart to XB~1254--690, GR Mus, was identified by Griffiths et al. (1978) to be a faint blue object (V=19.1), implying a ratio of X-ray to optical luminosity $L_x/L_{\\rm opt} \\sim 200$, similar to those of other compact X-ray binaries with Population II companions (Lewin \\& Joss, 1983).  EXOSAT observations showed the presence of type-I X-ray bursts and led to the discovery of recurrent X-ray dips with a period of $3.88 \\pm 0.15$ hr (Courvoisier, Parmar \\& Peacock 1984; Courvoisier et al. 1986).  The  duration of the dips was $\\sim 0.8$ hrs, representing $\\sim 20\\%$ of the binary cycle, with a reduction of the X-ray flux between 20 and 90\\%.  During the dips, X--ray variabilities are present on time scales 1--300 s. This complex time structure indicates that the obscuring region is clumpy and of variable size. A modulation with the same period of the X-ray dips was also observed in optical light curves, with the optical minimum occurring $\\sim 0.2$ cycles after the X-ray dip (Motch {\\it et al.} 1987). The presence of dips and the lack of eclipse of the X-ray source give a constraint on the inclination angle of the source that is between 65$^{\\circ}$ and 73$^{\\circ}$ (Courvoisier et al. 1986, Motch {\\it et al.} 1987). X-ray bursts and X-ray dips were also present in a Ginga observation performed on August 1990 (Uno et al. 1997). On the other hand the dips were not present in a RXTE observation in 1997 between April 28 and May 1 (Smale \\& Wachter, 1999). The cessation of the dipping activity indicates that the angular size of the disk edge (at the impact point with the accretion stream)  decreased from $17^\\circ-25^\\circ$ to less than $10^\\circ$ (Smale \\& Wachter, 1999). There is no evidence of a simultaneous variation of the accretion rate or the location of the outer rim of the accretion disk (Smale \\& Wachter, 1999).  The energy spectrum of XB~1254--690 in the 1--10 keV energy range, observed by {\\it EXOSAT}, was fitted by thermal bremsstrahlung with $kT = 5.5$ keV, and the corresponding flux was $\\sim 6 \\times 10^{-10}$ ergs cm$^{-2}$ s$^{-1}$ (Courvoisier et al. 1986), although other models could not be excluded. The spectrum obtained by {\\it Ginga} in the 2--35 keV range was fitted by a blackbody plus a power law with exponential cutoff, or by a multicolor disk blackbody plus a comptonized blackbody, where the electron temperature of the hot plasma is fixed at 100 keV and the optical depth is $\\sim 0.2$ (Uno et al. 1997).  Neither model gave acceptable residuals. The fits were partially improved including extra components like a Gaussian line centered at 6.3--6.7 keV.  We report here the results of a broad band (0.1--100 keV) spectral analysis performed on data from the BeppoSAX satellite.  The broad energy band of BeppoSAX narrow field instruments allows us to well constrain the spectrum of this source, and to detect the presence of a high energy component. ",
        "conclusions": "We analyzed data from a BeppoSAX observation of XB~1254--690 performed in 1998 December  22 and 23. Neither X--ray bursts nor X--ray dips were  present during this observation. The energy spectrum is described by a multicolor disk blackbody, with an inner temperature of $\\sim 0.85$ keV and an inner disk radius of $R_{\\rm eff} \\sqrt{\\cos i}\\sim 11$  km, a comptonized spectrum, and bremsstrahlung emission. The comptonized spectrum is produced in a relatively cool ($\\sim 2$ keV) region of moderate optical depth ($\\tau \\sim 19$ for a spherical geometry), probably surrounding the neutron star. This region could be identified with an inner optically thick corona or with the boundary layer between the accretion disk and the neutron star.  The presence of reflection signatures is not detected with high statistical significance probably because of the high inclination of the system. The bremsstrahlung component, probably produced in an extended, optically thin ADC, has a temperature of $\\sim 20$ keV. The corresponding emission region has a mean electron density $N_{\\rm e} \\sim 2.7 \\times 10^{14}$ cm$^{-3}$, and an optical depth $\\tau \\sim 0.71$. We observe an absorption edge at $\\sim 1.27$ keV, but the nature of this edge is not clear, given that we do not observe other edges in the spectrum. We do not observe an iron emission line at 6.4--6.7 keV, with an upper limit on the equivalent width of 18 eV for the iron line at 6.4 keV and 15 eV for the iron line at 6.7 keV."
    },
    "0010/astro-ph0010424_arXiv.txt": {
        "abstract": "The remnant of SN~1006 has an X-ray spectrum dominated by nonthermal emission, and pre-ASCA observations were well described by a synchrotron calculation with electron energies limited by escape. We describe the results of a much more stringent test: fitting spatially integrated \\asca \\gis (0.6 -- 8 keV) and \\rxte~\\pca~(3 -- 10 keV) data with a combination of the synchrotron model \\sresc newly ported to XSPEC and a new thermal shock model \\vshock.  The new model can describe the continuum emission above 2 keV well, in both spatial distribution and spectrum.  We find that the emission is dominantly nonthermal, with a small but noticeable thermal component: Mg and Si are clearly visible in the integrated spectrum.  The synchrotron component rolls off smoothly from the extrapolated radio spectrum, with a characteristic rolloff frequency of $3.1 \\times 10^{17}$ Hz, at which the spectrum has dropped about a factor of 6 below a powerlaw extrapolation from the radio spectrum.  Comparison of TeV observations with new TeV model images and spectra based on the X-ray model fits gives a mean post-shock magnetic field strength of about 9 $\\mu G$, implying (for a compression ratio of 4) an upstream magnetic field of 3 $\\mu G$, and fixing the current energy content in relativistic electrons at about $7 \\times 10^{48}$ erg, resulting in a current electron-acceleration efficiency of about 5\\%. This total energy is about 100 times the energy in the magnetic field. The X-ray fit also implies that electrons escape ahead of the shock above an energy of about 30 TeV.  This escape could result from an absence of scattering magnetohydrodynamic waves above a wavelength of about $10^{17}$ cm. Our results indicate that joint thermal and nonthermal fitting, using sophisticated models, will be required for analysis of most supernova-remnant X-ray data in the future. \\bigskip ",
        "introduction": "\\label{sec:introduction}     Since supernova shocks are one of the few mechanisms known to be capable of providing adequate energy to supply the pool of Galactic cosmic rays, supernova remnants (SNRs) have long been suspected as the primary site of Galactic cosmic ray acceleration, at least up to the slight steepening in the cosmic-ray spectrum at a few $10^{15}$ eV, known as the ``knee.'' Direct evidence for energetic particles comes from radio observations of synchrotron emission from 1--10 GeV electrons. However, in the cosmic rays observed at Earth, at a few GeV electrons are about 50 times less numerous than cosmic-ray ions, whose spectrum is an unbroken power law from $10^9$ eV to 10$^{15}$ eV. These considerations raise two questions: 1) Do SNRs accelerate ions? 2) Are they capable of accelerating {\\it any} particles to energies of 10$^{15}$ eV?  This paper addresses the second question by demonstrating the presence of electrons of energies of hundreds of TeV in the remnant of SN 1006 AD.  High-energy electrons can produce X-rays via nonthermal bremsstrahlung and synchrotron radiation, and gamma-rays from inverse-Compton upscattering of any photons present.  Relativistic protons can produce $\\gamma$-rays from the decay of $\\pi^0$ particles produced in inelastic collisions with background gas.  These processes have recently been studied in detail by \\citet{Sturner97}, \\citet*{Gaisser98} and \\citet{Baring99}.  The analysis of X-ray and gamma-ray observations of shell supernova remnants may give direct evidence bearing on both questions above.  In this paper we will interpret observed X-rays and TeV gamma-rays as primarily synchrotron radiation and inverse-Compton upscattered cosmic microwave background radiation, respectively, and will obtain maximum electron energies and electron shock acceleration efficiencies -- crucial information for understanding shock acceleration in general, and the origin of Galactic cosmic rays.  The earliest evidence of nonthermal X-ray emission in a shell supernova remnant came from the featureless spectrum of SN~1006 AD (G327.6--1.4) \\citep{Becker80}, explained as the loss-steepened extrapolation of the radio synchrotron spectrum by \\citet{Reynolds81}. However, early data were poor and the models were simplistic.  Thermal models by \\citet*{Hamilton86} seemed able to produce a featureless X-ray spectrum, given a high degree of elemental stratification in ejecta. However, observations by \\asca~ \\citep{Koyama95} showed unmistakable evidence for nonthermal emission in the rims, along with thermal, line-dominated emission in the interior.  \\citet{Reynolds96} demonstrated that detailed, self-consistent synchrotron models could be constructed which adequately described the pre-\\asca~ integrated spectrum of SN~1006 as the diminishing extension of the radio spectrum, reproducing the slope reported by \\citet{Koyama95} in the rim.  While \\citet{Laming98} proposed a modified thermal bremsstrahlung model based on the model of \\citet{Hamilton86}, even the author concluded that it is not an appropriate fit for SN~1006.   Since then, several more shell SNRs have been shown to have nonthermal, non-plerionic emission.  In addition to  SN~1006 \\citep{Koyama95}, G347.3-0.5 (RJ J1713.7-3946) \\citep{Koyama97, Muraishi00,Slane99} has nearly featureless X-ray spectra clearly dominated by nonthermal emission.  Some remnants, while not dominated by nonthermal emission in the \\asca~band, show other evidence pointing to its presence. \\osse~ observations of Cassiopeia A at 400-1250 keV \\citep{The96} are well described by a broken power law, steepening to higher energies \\citep{Allen97a}. RCW 86 \\citep*{Vink97} shows anomalously low abundances when fit with thermal models.  When the fit includes a synchrotron model the abundances fall within expected ranges \\citep{Borkowski99}.  An important consequence of synchrotron X-ray emission was pointed out by \\citet{Pohl96} and \\citet{MastdeJ96}.  Electrons capable of producing keV synchrotron photons in magnetic fields of a few microgauss will also produce high-energy photons by inverse-Compton scattering any photon fields.  In particular, inverse-Compton upscattering of cosmic-microwave-background photons can produce TeV gamma rays, and \\citet{Pohl96} explicitly predicted a range for the expected TeV flux of SN~1006 (subsequently found by the CANGAROO air \\v Cerenkov telescope). The predicted flux depends only on the electron distribution, so it can be used, in conjunction with observed synchrotron fluxes, to deduce a mean magnetic-field strength in a remnant. Subsequent extensive calculations by \\citet{Sturner97}, \\citet{Gaisser98}, and \\citet{Baring99} included this effect.  However, nonthermal X-ray emission from shell supernova remnants could be due to bremsstrahlung from nonrelativistic electrons with energies of a few tens of keV \\citep{Asvarov90} rather than synchrotron radiation from electrons with energies of hundreds of TeV.  Careful models (e.g., \\citealp{Baring99} as well as \\citealp{Reynolds96}) show that in neither case are straight power-laws to be expected. Synchrotron spectra should all be steepening (convex up) while bremsstrahlung spectra, resulting from the lowest energy cosmic ray electrons departing from the Maxwellian tail, ought to be flattening (concave up). \\citet{Allen97a} showed that the \\asca~spectrum from SN~1006 steepens to higher energies, and the discovery of the predicted TeV gamma rays by the \\cangaroo\\ collaboration \\citep{Tanimori98} provided strong evidence that the nonthermal X-ray emission in SN~1006 is in fact synchrotron radiation.  However, nonthermal bremsstrahlung is expected to be important for some remnants, and as more observations become available above 10 keV, it will be necessary to discriminate between synchrotron and nonthermal bremsstrahlung.  (While most analyses, e.g., \\citet{MastdeJ96}, conclude that the TeV emission from SN~1006 is inverse-Compton, \\citet{Aharonian99} and \\citet*{Berezhko99} assert that the $\\pi^0$-decay mechanism may be partly responsible.)  Despite the strong evidence for curvature, a straight power law is still commonly regarded as a defining characteristic of X-ray synchrotron radiation in shell supernova remnants, as it is in Crab-like supernova remnants and active galactic nuclei.  While power laws will certainly be a better description of synchrotron emission produced by the particle spectra expected in SNR shock acceleration than quasi-thermal exponential continua, all SNR X-ray synchrotron continua are expected to be substantially curved, even over a relatively narrow energy range. However, in almost all published studies, nonthermal SNR X-ray emission has been described by a power law. This purely phenomenological treatment, while characterizing the emission, fails to explain its origin. Power laws cannot continue forever rising as the frequency drops, a fact camouflaged by the rapidly dropping response of X-ray instruments at low energies. Broken power laws can avoid this ``infrared catastrophe,'' but they still duck the question of the origin of the emission and its relation to the lower-energy synchrotron spectrum. Since no Galactic supernova remnant has X-ray emission bright enough to lie on the unbroken extrapolation of the radio spectrum \\citep{Reynolds99}, extrapolating from radio measurements will grossly over-predict the X-ray flux, as demonstrated in Figure \\ref{fluxes}. The spectrum must drop significantly below the observed power law in the radio to be consistent with the X-ray flux measurements. A good physical model must both fit this turnover and explain the mechanism causing the spectral curvature.   Some spectral modeling of SNR synchrotron X-rays has been done using other descriptions.  While a power law represents a spectrum with no curvature, the maximum curvature likely to result from shock acceleration comes from an electron spectrum with an exponential cutoff.  In the approximation in which each electron radiates all its energy at the peak frequency of the single-particle synchrotron emissivity (the ``$\\delta$-function approximation''), this implies a photon spectrum $S(E) \\propto \\exp((E/E_m)^{-1/2})$. \\citet{Aharonian99} modeled synchrotron emission from SN~1006 using this expression, with a constant maximum energy $E_m$.  As was demonstrated in \\citet{Reynolds98} the $\\delta$-function approximation is not accurate well above the turnover energy. In addition, the electron distribution will vary throughout the remnant.  A convolution of the electron synchrotron emissivity with the calculated electron distribution at each location in the remnant is required. The model described in this paper includes the full electron emissivity.    It is important to understand and accurately characterize the synchrotron emission in SNRs. If the emission mechanism in those SNRs with clear nonthermal continua is synchrotron radiation, then many SNRs could have varying amounts of synchrotron X-ray emission, contributing part of the continuum emission alongside the thermal lines and continuum. As demonstrated by RCW 86 \\citep{Borkowski00a}, such an unmodeled component stymies thermal fits by preventing accurate measurements of shock temperatures and elemental abundances. Understanding either the thermal or nonthermal X-ray emission from SNRs requires appropriate modeling of both components. This paper applies a model that assumes Sedov dynamics and a power-law distribution of electrons up to a maximum energy \\emax, which can vary with time and with location in the remnant.  In general, the maximum energy attained by electrons in the shock acceleration process could be limited by one of several factors --- electrons above this energy could escape from the remnant, the remnant could be so young it has not had time to accelerate particles beyond some \\emax, or \\emax~could be the energy at which radiative losses balance further acceleration.  Normally, a remnant's spectrum would first be limited by age, and then by radiative losses, unless escape prevented either of the other limits from being reached.  In this paper we analyze new \\rxte \\ observations, along with archival \\asca \\ observations, in Section \\ref{sec:observations}. We discuss the escape model, {\\it sresc}, in Section \\ref{sec:model}, along with our new gamma-ray modeling, and compare \\sresc~to thermal models and a thermal+power law model in Section \\ref{sec:results}. In Section \\ref{sec:discuss}, we discuss the elemental abundances we infer from the joint thermal-nonthermal model, and compare them to predictions by \\citet*{Nomoto84} and \\citet{Iwamoto99} and discuss the implications of our results for the efficiency of electron acceleration.  We summarize our conclusions in Section \\ref{sec:conclusions}. ",
        "conclusions": "\\label{sec:conclusions}  \\begin{enumerate} \\item The escape-limited synchrotron model, {\\it sresc}, provides a good fit to integrated spectral observations of SN~1006 by \\rxte~ and \\asca~ -- the highest quality data to date. In addition, the spatial prediction closely matches the X-ray image in the energy range where the model applies.  \\item The \\sresc model provides a significant improvement over the power-law models:  \\begin{enumerate}  \\item \\sresc provides a more accurate description of the emission based on physical principles. \\item \\sresc allows a clearer separation of thermal and nonthermal emission, constrained by radio observations \\end{enumerate}  \\item We believe an adequate description of the synchrotron emission leads to more accurate temperature and abundance measurements in the remaining thermal model, though better thermal inferences require both spatially resolved spectra (and models) and codes appropriate for ejecta dominated by heavy elements. Evidence for enhanced abundances suggests that at least part of the thermal X-ray emission comes from ejecta.  \\item The TeV gamma-ray observations allow us to determine the energy in relativistic electrons to be about $7 \\times 10^{48}$ erg; the current shock efficiency at accelerating electrons is about 5\\%, and the energy in relativistic particles is much greater than that in magnetic field.  \\item We estimate that the ambient magnetic field is about 3 $\\mu$Gauss, and that the MHD wave spectrum near SN 1006 drops substantially in amplitude above a wavelength of about $10^{17}$ cm.   \\end{enumerate}  We have demonstrated that {\\it sresc}, in conjunction with inhomogeneous thermal models, can describe the full remnant emission.  These models (\\sresc and \\srcut in more general situations, as discussed in Section \\ref{sec:model}) represent a significant improvement over power-law models in describing the physics of synchrotron X-ray emission. We can now proceed to spatially-resolved spectral modeling of SN~1006. The true test of the escape limited synchrotron model will be its application to the spatially resolved data sets in our forthcoming paper. We will develop specialized versions of the \\sresc model to describe spatially distinct areas of SN~1006."
    },
    "0010/astro-ph0010612_arXiv.txt": {
        "abstract": "This paper presents the \\xmm\\ first-light observations of the Hickson-16 compact group of galaxies. Groups are possibly the oldest large-scale structures in the Universe, pre-dating clusters of galaxies, and are highly evolved. This group of small galaxies, at a redshift of 0.0132 (or 80 Mpc) is exceptional in the having the highest concentration of starburst or AGN activity in the nearby Universe. So it is a veritable laboratory for the study of the relationship between galaxy interactions and nuclear activity. Previous optical emission line studies indicated a strong ionising continuum in the galaxies, but its origin, whether from starbursts, or AGN, was unclear. Combined imaging and spectroscopy with the EPIC X-ray CCDs unequivocally reveals a heavily obscured AGN and a separately identified thermal (starburst) plasma, in NGC 835, NGC 833, \\& NGC 839. NGC 838 shows only starburst thermal emission. Starbursts and AGN can evidently coexist in members of this highly evolved system of merged and merging galaxies, implying a high probability for the formation of AGN as well as starbursts in post-merger galaxies.   \\begin{keywords} galaxies: active -- galaxies: starburst -- X-rays: galaxies \\end{keywords} ",
        "introduction": "The Hickson-16 galaxy group (or HCG-16) comprises seven galaxies with a mean recession velocity (Ribeiro \\et \\cite{ribeiro}) of 3959$\\pm$66km~s$^{-1}$ and a velocity dispersion of 86$\\pm$55~km~s$^{-1}$, centered on the position $\\alpha=02\\degg 09\\arcm 33\\arcs$, $\\delta=-10\\degg 09\\arcm 46.7\\arcs$ (J2000). The four central members of the group, originally identified by Hickson (Hickson \\cite{hickson}), all fall within the 30\\arcm\\ field of view of EPIC, they all show evidence for mergers and a strong ionising continuum; they have well resolved optical nuclei (Mendes de Oliveira \\et 1998). However the nature of the ionising continuum is unclear. Optical emission line diagnostics (de Carvalho \\& Coziol \\cite{decarv},  Veilleux \\& Osterbrock \\cite{veil}) suggest that in NGC 838 and NGC 835 there is a starburst nucleus, whilst in NGC 835, NGC833 and NGC 839 there is a low ionisation narrow emission line region (or LINER 2 nucleus). This could arise either from an AGN or from starburst activity: [OIII]/H$\\beta$ values for the nuclei are all $<2.5$ (de Carvalho \\& Coziol \\cite{decarv}). The EPIC instrument is able to observe directly the ionising continuum, and distinguish clearly between optically thin thermal emission from a starburst, and non-thermal hard X-ray emission from an AGN. These observations can therefore be used to determine the nature of the ionising continuum, helping to clarify the relationship between mergers, the triggering of starbursts, and the creation and fueling of black holes.  \\begin{figure*} \\resizebox{\\hsize}{!}{\\rotatebox{-90}{\\includegraphics{HCG16_om.ps}}} \\caption{The OM colour image of HCG 16 derived from 1000 second exposures in V and UV bands. A bright foreground star has been removed from the lower right of the image. Blue here represents the UV band.} \\end{figure*} ",
        "conclusions": "Direct X-ray spectroscopy is the best way to identify hidden AGN in galaxies, and here the EPIC cameras on XMM-Newton have produced clear evidence for active, massive black holes in three out of four galaxies in HCG-16. The presence of a similar active nucleus in NGC 838 is unlikely, unless it is very heavily absorbed. The nature of the ionising continuum in the four galaxies has been elucidated: there is thermal emission from starburst activity in three, (possibly four) of the galaxies, and in three of them there is a coexisting active black hole. While LINER-1 galaxies with broad H$\\beta$ lines do harbour black holes (Terashima \\et \\cite{tera-one}, \\cite{tera-two}) this is the first {\\it direct} evidence that black holes power LINER2 galaxies. These AGN are at the low end of the luminosity scale, consistent with their small size (Magorrian \\et \\cite{mag}).  In NGC 833, the accreting black hole X-ray luminosity arguably exceeds the FIR luminosity. This is very unusual, even compared with much more luminous AGN, and may indicate the stripping of dust and gas by past interactions with other galaxies (Mendes de Oliveira \\et \\cite{mendes}). The EPIC X-ray study of this nearby and evolved system of small galaxies indicates a high fraction of active black holes coexisting with starbursts. This is consistent with optical studies of compact groups (e.g. Coziol \\et \\cite{coziol}), where a large fraction of galaxies with nuclear activity is found. The observations may also provide a link between normal galaxies, where black holes may be inactive, and Seyfert galaxies and quasars, where black holes dominate, and where galaxy mergers may be implicated in the onset of black hole activity (Boyce \\et \\cite{boyce}, Bahcall \\et \\cite{bahcall}). Further observations with XMM-Newton have the potential to determine more precisely the fraction of nearby galaxies that harbour active, low-luminosity black holes."
    },
    "0010/astro-ph0010338_arXiv.txt": {
        "abstract": "The {\\it Chandra X-ray Observatory} recently discovered an X-ray point source near the center of Cassiopeia A, the youngest known Galactic supernova remnant. We have conducted a sensitive search for radio pulsations from this source with the Very Large Array, taking advantage of the high angular resolution of the array to resolve out the emission from the remnant itself. No convincing signatures of a dispersed, periodic source or of isolated dispersed pulses were found, whether for an isolated or a binary source. We derive upper limits of 30 and 1.3 mJy  at 327 and 1435 MHz for its phase-averaged pulsed flux density. The corresponding luminosity limits are lower than those for any known radio pulsar with age less than $10^{4}$ years. Our search sensitivities to single pulses were 25 and 1.0 Jy at 327 and 1435 MHz. For comparison, the Crab pulsar emits roughly 80 pulses per minute with flux densities greater than 100 Jy at 327 MHz and 8 pulses per minute with flux densities greater than 50 Jy at 1435 MHz. These limits suggest that Cas~A belongs to the growing population of young neutron stars that are radio quiet ",
        "introduction": "Cassiopeia~A, with an age of only 320 years, is the youngest known Galactic supernova remnant. The supernova that gave rise to this remnant was probably discovered in 1680, when Flamsteed (1725) observed a 6th magnitude star within $13'$ of Cas~A's location (\\cite{ash80}). Because its abundances of heavy elements are consistent with those expected from the explosion of a massive star (\\cite{hughes00}), Cas~A is thought to be the remnant of a Type II supernova. Given the stellar initial mass function, a large fraction of Type II supernovae are expected to leave behind neutron stars. While many searches at various wavelengths for the compact remnant in Cas~A have been undertaken, only upper limits had been established until the recent {\\it Chandra} discovery of an X-ray point source (hereafter XPS) near the center of the remnant (\\cite{tan99}). This XPS was later confirmed in archival {\\it ROSAT} (\\cite{asc99}) and {\\it Einstein} (\\cite{pav99}) images. The XPS is within $5''$ of the expansion center of the remnant (\\cite{van83}). For a remnant distance of $3.4^{+0.3}_{-0.1}$ kpc, calculated through radial velocity, proper motion, and age data (\\cite{reed95}), this offset corresponds to a maximum transverse velocity of roughly 200 km s$^{-1}$, a typical velocity for young pulsars (\\cite{cor98}).  Pavlov et al. (2000) find that the X-ray spectrum of the XPS is fit well by a power law with index $2.6 - 4.1$, with luminosity $(2-60) \\times 10^{34}$ ergs s$^{-1}$. Similarly, Chakrabarty et al. (2001) estimate a power law index of $2.8 - 3.6$ and luminosity of $(2 - 16) \\times 10^{34}$ ergs s$^{-1}$. These indices are much steeper than the X-ray spectra of Active Galactic Nuclei (AGN). Furthermore, given the measured surface density of AGN, the probability of finding an AGN that close to the center of the SNR is negligible. For these reasons, Pavlov et al. (2000) conclude that the XPS is almost certainly associated with the Cas~A remnant. The fitted power law indices for the XPS are also much steeper than those measured for the 6 young X-ray pulsars having ages less than $10^{4}$ years, whose spectral indices range from $1.1-1.7$ (\\cite{deepto01}). In addition, the estimated XPS luminosities are only marginally consistent with measured young pulsar X-ray luminosities, which range from $3\\times 10^{35}$ to $4\\times 10^{36}$ ergs s$^{-1}$. It is also significant that, unlike 5 of the 6 young pulsars with ages less than $10^{4}$ years, Cas~A shows no evidence for a synchrotron nebula surrounding the point source.  While there is much evidence for and against the neutron star nature of the XPS, the definitive proof would be the detection of pulsations. No X-ray pulsations have been found in data taken with the {\\it Chandra} High Resolution Camera. However, a time-tagging problem caused the timing accuracy to be degraded from 16 $\\mu$s to $\\approx$ 4 ms, allowing searches for only broad pulses with periods greater than 20 ms (\\cite{deepto01}). The detection of radio pulsations, however, would be just as conclusive. For this reason, we have conducted a search for radio pulsations from the XPS using the Very Large Array (VLA) at frequencies of 327 and 1435 MHz. There have been several previous searches for a radio pulsar in Cas~A (\\cite{davies70,seir80,woan93,lorimer98}). Our VLA search is many times more sensitive, as having an accurate position for the XPS allows us to resolve out the strong emission from the remnant, which with 327- and 1435-MHz flux densities of 6400 and 2000 Jy is one of the strongest radio sources known.  \\bigskip ",
        "conclusions": "We have calculated upper limits of 30 and 1.3 mJy at 327 and 1435 MHz for the phase-averaged pulsed flux density from the XPS in Cas~A. We did not detect any single, dispersed pulses above our sensitivity of 25 and 1.0 Jy (assuming a pulse width of 1~ms) at 327 and 1435 MHz. We now compare our upper limit to those from previous Cas~A searches. Davies \\& Large (1970) carried out an unsuccessful 408-MHz search for single pulses from Cas~A. Seiradakis \\& Graham (1980) quoted a 1420-MHz upper limit of 1.6 mJy for pulsed emission from Cas~A. However, their limit does not account for the additional system temperature contributed by the nebula, significant for their beamwidth of $9'$. More recently, Woan \\& Duffett-Smith (1993) published an upper limit of 80 mJy at 408 MHz and Lorimer et al. (1998) published an upper limit of 46 mJy at 606 MHz. For a typical pulsar spectrum, our 1435-MHz upper limit is an order of magnitude better than these previously published limits, due to the known position of the XPS and the power of the VLA to resolve out the excess emission from the nebula.  Our upper limit of 1.3 mJy at 1435 MHz translates to a luminosity of 15 mJy kpc$^{2}$ for a neutron star at a distance of 3.4 kpc. This is lower than the mean 1400-MHz luminosity of 30 mJy kpc$^{2}$  estimated for young pulsars (\\cite{frail93}) and is lower than the luminosity of any known radio pulsar with characteristic age less than 10$^{4}$ years (see Figures 1 and 2 of Brazier \\& Johnston 1999). This suggests that the XPS in Cas~A is either not a pulsar, is a radio-quiet pulsar, or is a radio pulsar beamed away from us. While evidence suggests that pulsar beams are wider at short periods (see e.g.~Biggs (1990)) and that they are therefore likely to intersect the Earth's line of sight, individual pulsars show mixed properties. For instance, the Crab pulsar shows narrow radio components that may be part of a large beam, but the profile of another young pulsar in the Large Magellanic Cloud, PSR B0540-69, is very broad (\\cite{man93}). Furthermore, it is difficult to disentangle true beam width from viewing angle for these pulsars. It is certainly possible that, like the Geminga pulsar, the XPS shows no radio pulses but is an X-ray, and perhaps gamma-ray, pulsar. Although there is no EGRET point source near the position of the XPS, gamma-ray searches with future, more sensitive instruments, could be successful. Further measurements of the spectrum and time variability of the XPS will also help determine which of the above scenarios is the case.  This search adds to the growing body of evidence that there are manifestations of young neutron stars other than the standard Crab-like radio pulsar. Although it is widely accepted that radio pulsars are born in supernovae, fewer than 10\\% of searches targeted at supernova remnants have found young (age $<$ 25 kyr) pulsars (\\cite{kaspi96,gorham96,lorimer98}). Some of this may be due to scattering at low frequencies in the Galactic plane and the decreased sensitivity due to the bright remnants, but it also hints at the unrecognized diversity of the neutron star population. While some neutron stars are detected as radio pulsars, others may be born as anomalous X-ray pulsars, soft gamma-ray repeaters, quiescent neutron stars, or other classes of objects not yet discovered (\\cite{kaspi2000,gott00}). The pathways to and percentages of these different classes of objects are not yet clear. However, future {\\it Chandra} detections of compact X-ray sources, and large-scale radio pulsar surveys, like the Parkes Multibeam Survey (\\cite{lyne2000}), will help to answer these questions."
    },
    "0010/astro-ph0010562_arXiv.txt": {
        "abstract": "We present progress results of a new survey for bright QSOs ($V<14.5$, $R<15.4$, $B_J<15.2$) covering the whole sky at high galactic latitudes, $|b|>30$. The surface density of QSOs brighter than $B_J=14.8$ turns out to be $2.9 \\pm 0.8 \\cdot 10^{-3} deg^{-2}$. The optical Luminosity Function at $0.04 < z \\le 0.3$ shows significant departures from the standard pure luminosity evolution, providing new insights in the modelling of the QSO phenomenon. ",
        "introduction": "Why a survey for local QSOs?\\\\ Because in the epoch of 2dF and SDSS, of the thousands of QSOs at $z\\sim 2$, it is important to have precise information about the spatial density and the clustering of nearby AGN to provide zero-point and leverage for the study of the QSO evolution. As in other fields, it is paradoxical that we know much better the properties of the high-z QSO population than the local one. The technical problem is however not trivial since, as the PG survey (Schmidt \\& Green 1983) made clear, it is necessary to cover the whole sky to obtain sufficient statistics where the present data are unsatisfactory: at $z<0.3$ and magnitudes around $13-15$. Such a task has become relatively simpler in recent times, thanks to the availability of large high-quality databases. ",
        "conclusions": ""
    },
    "0010/astro-ph0010081_arXiv.txt": {
        "abstract": "We investigate stability of Toomre-Hayashi model for a self-gravitating, rotating gas disk. The rotation velocity, $ v _\\varphi $, and sound speed, $ c _{\\rm s} $, are spatially constant in the model. We show that the model is unstable against an axisymmetric perturbation irrespectively of the values of $ v _\\varphi $ and $ c _{\\rm s} $. When the ratio, $ v _\\varphi / c _{\\rm s} $, is large, the model disk is geometrically thin and unstable against an axisymmetric perturbation having a short radial wavelength, i.e., unstable against fragmentation. When $ v _\\varphi / c _{\\rm s} $ is smaller than 1.20, it is unstable against total collapse. Toomre-Hayashi model of $ 1.7 \\, \\la \\, v _\\varphi / c _{\\rm s} \\, \\la \\, 3 $ was thought to be stable against axisymmetric perturbations in earlier studies in which only radial motion was taken into account.  Thus, the instability of Toomre-Hayashi model having a medium $ v _\\varphi / c _{\\rm s}  $ is due to not change in the surface density but that in the disk height. We also find that the singular isothermal perturbation is stable against non-spherical peturbations. ",
        "introduction": "\\citet{toomre82} and \\citet*{hayashi82} found  an analytic model for a self-gravitating gas cloud. The model describes a disk in which the self-gravity balances with the centrifugal force and gas pressure. The rotation velocity, $ v _\\varphi $, and the sound speed, $ c _{\\rm s} $, are assumed to be constant. The density distribution of the model is expressed as \\begin{equation} \\rho \\, (r, \\, \\theta) \\; = \\; \\frac{\\gamma ^2 c _{\\rm s} ^2} {2 \\pi G r ^2} \\, \\left\\lbrack \\frac{\\cosh \\zeta}{\\cosh (\\gamma \\zeta)} \\right\\rbrack ^2 \\; , \\label{th-model} \\end{equation} where \\begin{equation} \\zeta \\; = \\; \\ln \\, \\left( \\frac{1 \\, + \\, \\cos \\, \\theta}{\\sin \\, \\theta} \\right) \\; , \\end{equation} and \\begin{equation} \\gamma \\; = \\; 1 \\, + \\, \\frac{v _\\varphi ^2}{2 c _{\\rm s} ^2} \\; , \\end{equation} in the spherical polar coordinates, ($ r $, $ \\theta $, $ \\varphi $).  This simple model is useful for studying dynamics of self-gravitating gas disk.  Earlier researchers \\citep*[see, e.g.,][]{kalnajs71, lemos91, syer96, evans98a, evans98b, goodman99, shu00} studied further simplified model, i.e., the razor-thin power law disk model, which is equivalent to the vertically averaged Toomre-Hayashi model. Besides of its simplicity, it is a good approximation to a centrifugally supported disk formed by collapse of a rotating gas cloud \\citep*[see, e.g.,][] {matsumoto97, saigo98}. The centrifugally supported disk will evolve into single or binary stars associated with gas disks. The evolution may be driven by instabilities of the disk. Since the Toomre-Hayashi model is an exact equilibrium solution, it is ideal for linear stability analysis; we need not worry about inaccuracy of the unperturbed state.  In this paper we concentrate on the simplest mode, the axisymmetric one.  We will show that the Toomre-Hayashi model is unstable for any $ \\gamma $, i.e., any $ v _\\varphi / c _{\\rm s} $.  This result is different from prediction by \\citet{hayashi82} and  earlier analyses by \\citet{lemos91} and \\citet{shu00}.  While two-dimensional perturbation is taken into account in our analysis, only radial perturbation was considered in earlier analysis.  We believe that our analysis will provide insights for stability of self-gravitating disk.  The rest of our paper has the following structure.  In \\S 2 we derive energy principle for axisymmetric perturbation around the Toomre-Hayashi disk.  In \\S 3 we obtain the marginally stable mode for the Toomre-Hayashi disk. In \\S 4 we show that it is unstable for any $ \\gamma $ and discuss implications of our stability analysis. Main conclusions are summarized in \\S 5. ",
        "conclusions": "We discussed the stability of the Toomre-Hayashi model against axisymmetric perturbations in this paper.  Our approach was conservative since we focused on the marginally stable mode.  We did not obtain an unstable mode.  Nevertheless we had some new findings.  This is because we used least approximations and took account of the vertical structure exactly.  Although our analysis is restricted in the axisymmetric mode, the result will be helpful for more general analysis on the stability of self-gravitating disk.  For later use we summarize the main results in the following.  \\begin{enumerate} \\item The Toomre-Hayashi model is unstable against axisymmetric perturbation irrespectively of its model parameter, $ \\gamma $. \\item The wavelength of the unstable perturbation is proportional to the radius.  The ratio of the wavelength to the radius is larger for smaller $ \\gamma $. \\item Change in the vertical height can induce instability even when the surface density does not change. \\item It is prospected that an unstable perturbation may grow in proportion of time in the power of time. \\item The Toomre-Hayashi model of $ \\gamma \\, = \\, 0 $, i.e., the singular isothermal sphere is proven to be stable against a non-spherical perturbation. \\end{enumerate}   The authors thank Tetsuro Konishi for telling us the asymptotic solution shown in \\S 3. This research is financially supported in part by the Grant-in-Aid for Scientific Research on Priority Areas and that for Encouragement of Young Scientists of the Ministry of Education, Science, Sports and Culture of Japan (Nos. 10147105, 11134209, 1202103, and 12740123)."
    },
    "0010/astro-ph0010048_arXiv.txt": {
        "abstract": "The ``drifting'' subpulses exhibited by some radio pulsars have fascinated both observers and theorists for 30 years, and have been widely regarded as one of the most critical and potentially insightful aspects of their emission.  Moreover, Ruderman \\& Sutherland (1975), in their classic model, suggested that such regular modulation was produced by a system of subbeams, rotating around the magnetic axis under the action of $\\bf{E}$$\\times$$\\bf{B}$ drift.  Such ``drift'' sequences have thus been thoroughly studied in a number of pulsars, but it has proven difficult to verify the rotating-subbeam hypothesis, and thus to establish an illuminating connection between the phenomenon and the actual physics of the emission.  Here, we report on detailed studies of pulsar B0943+10, whose nearly coherent sequences of ``drifting'' subpulses have permitted us to identify their origin as a system of subbeams that appear to circulate around the star's magnetic axis.  We introduce several new techniques of analysis, and we find that both the primary and secondary features in the star's fluctuation spectra are aliases of their actual values.  We have also developed a method of tracing the underlying pattern responsible for the observed sequences, using a ``cartographic'' transform and its inverse, permitting us to study the characteristics of the polar-cap emission ``map'' and to confirm that such a ``map'' in turn represents the observed sequence.  We apply these techniques to the study of three different Arecibo observations: a 1992 430-MHz sequence which includes a transition from the star's highly organized ``B'' profile mode to its disorganized ``Q'' mode; a 1972 430-MHz ``B''-mode sequence; and a 1990 111-MHz ``B''-mode sequence.  The ``B''-mode sequences are consistent in revealing that the emission pattern consists of 20 subbeams, which rotate around the magnetic axis in about 37 periods or 41 seconds.  Even in the ``Q'' mode sequence, we find evidence of a compatible circulation time.   The similarity of the subbeam patterns at different radio frequencies strongly suggests that the radiation is produced within a set of columns, which extend from close to the stellar surface up though the emission region and reflect some manner of a ``seeding''phenomenon at their base.  The subbeam emission is then tied neither to the stellar surface nor to the field.  While the origin of the ``memory'' responsible for the stability of the pattern over several circulation times is unknown, the hollow-conal form of the average pattern is almost certainly the origin of the conal beam forms observed in most pulsars. ",
        "introduction": " ",
        "conclusions": ""
    },
    "0010/astro-ph0010518_arXiv.txt": {
        "abstract": "The Southern Galactic Plane Survey (SGPS) is an \\HI\\ and 1.4-GHz continuum survey of the 4th quadrant of the Galaxy at arcmin resolution.  We present here results on linearly polarized continuum emission from an initial 28-square-degree Test Region for the SGPS, consisting of 190 mosaiced pointings of the Australia Telescope Compact Array, and covering the range $325\\fdg5 < l < 332\\fdg5$, $-0\\fdg5 < b < +3\\fdg5$. Complicated extended structure is seen in linear polarization throughout the Test Region, almost all of which has no correlation with total intensity. We interpret the brightest regions of polarized emission as representing intrinsic structure in extended polarization, most likely originating in the Crux spiral arm at a distance of 3.5~kpc; fainter polarized structure is imposed by Faraday rotation in foreground material.  Two large areas in the field are devoid of polarization. We argue that these voids are produced by foreground \\HII\\ regions in which the magnetic field is disordered on scales of $\\sim0.1-0.2$~pc.  We also identify a depolarized halo around the \\HII\\ region RCW~94, which we suggest results from the interaction of the \\HII\\ region with a surrounding molecular cloud. ",
        "introduction": "\\label{sec_intro}  Soon after linear polarization was first detected from celestial radio sources, it became apparent that the entire Galactic Plane was a significant source of polarized emission (\\cite{sw61}). There appear to be at least two components to this emission (e.g.\\ \\cite{jfr87}; \\cite{dhjs97}): polarization from discrete supernova remnants (SNRs), and a diffuse polarized background produced by the interaction of the relativistic component of the interstellar medium (ISM) with the Galactic magnetic field.  In the presence of magnetic fields and free electrons, electromagnetic radiation propagates in two circularly-polarized orthogonal modes of different velocities. Linearly polarized radiation passing through a magneto-ionized medium will thus emerge with its position angle rotated, an effect known as {\\em Faraday rotation} (\\cite{far44}; \\cite{cp61}). If a source emits linear polarization with an intrinsic position angle $\\Theta_0$, then the measured position angle $\\Theta$ at an observing wavelength $\\lambda$ is given by: \\begin{equation} \\Theta = \\Theta_0 + {\\rm RM}~\\lambda^2. \\label{eqn_rm} \\end{equation} In this expression, the {\\em rotation measure}\\ (RM), in units of rad~m$^{-2}$, is defined by : \\begin{equation} {\\rm RM}~ = K \\int B \\cos \\theta~n_e dl, \\end{equation} where $K = 0.81$~rad~m$^{-2}$~pc$^{-1}$~cm$^3$~$\\mu$G$^{-1}$, $B$, $\\theta$ and $n_e$ are the magnetic field strength, inclination of the magnetic field to the line of sight and electron density respectively, and the integral is along the line of sight from the observer to the source.  Faraday rotation can be a powerful probe of Galactic sources, and has been used to determine the magneto-ionic properties of both individual objects (e.g.\\ \\cite{hc80}; \\cite{mldg84}) and of the Galaxy as a whole (e.g.\\ \\cite{sk80}; \\cite{hq94}). Extragalactic sources (\\cite{ccsk92}), pulsars (\\cite{hmq99}) and  extended polarized emission from the entire Galaxy (\\cite{spo84}) have all been used as background sources against which Faraday rotation can be studied. However, with a single-dish telescope all such studies have their limitations --- studies of extragalactic sources suffer from confusion in the Galactic Plane, pulsars are of too low spatial density to study specific regions of the sky in detail, and the spatial resolution with which extended emission can be imaged is generally poor.  In more recent years, instruments and techniques have been developed which allow interferometers to map large regions of the sky in polarization. Consequent polarimetric surveys with both the Westerbork and DRAO Synthesis Telescopes do not suffer from any of the problems described above for single-dish observations, and can thus study polarization and Faraday rotation at high spatial resolution. These studies have revealed several unusual features with no counterparts either in total intensity or at any other wavelength (\\cite{wdj+93};  \\cite{gldt98}; \\cite{gld+99}; \\cite{hkd00}).  These results demonstrate that there are components of the ISM which can only be studied through polarimetry.  The above-mentioned surveys have been carried out at high Galactic latitudes and/or in the 2nd Galactic quadrant, and thus provide important information on the magneto-ionic properties of local gas and of the outer Galaxy.  However, it is also of considerable interest to study the polarimetric properties of the inner Galaxy, in which there are undoubtedly regions of stronger magnetic fields, higher densities and increased levels of turbulence. The complexity of polarization in this part of the sky has been demonstrated by the spectacular single-dish polarization surveys of Duncan \\etal (1997a\\nocite{dhjs97}, 1999\\nocite{drrf99}), but a full understanding of the emission in these regions requires the higher angular resolution which only an interferometer can provide.  The Australia Telescope Compact Array (ATCA) is ideally suited to such a study. Its antipodean location allows it to observe almost the entire 4th Galactic quadrant, the high polarimetric purity of its receivers ensures that polarized signals are uncontaminated by leakage from total intensity, its mosaicing capability allows it to easily map large parts of the sky, its multi-channel continuum mode allows accurate studies of Faraday rotation, and its compact antenna configurations give it good sensitivity to extended structure. Motivated by these capabilities, we are in the process of mapping the entire Galactic Plane south of declination $\\sim-35^\\circ$ in 1.4-GHz polarization; observations are being made simultaneously in the \\HI\\ line. Together with single-dish \\HI\\ data being taken with the 64-m Parkes radiotelescope (\\cite{mdg+00a}), all these data form the {\\em Southern Galactic Plane Survey}\\ (SGPS; \\cite{dmg+99}; \\cite{mdg+00c})\\footnote{see also the SGPS WWW page, at {\\tt http://www.astro.umn.edu/\\~{ }naomi/sgps.html .}}, the ATCA component of which will cover the Galactic coordinates $253^\\circ < l < 358^\\circ$, $-1^\\circ < b < 1^\\circ$. The polarimetric component of this data-set will have a spatial resolution of $\\sim$1~arcmin, and a theoretical sensitivity of $\\sim$0.25~mJy (where 1~jansky~$= 10^{-26}$~W~m$^{-2}$~Hz$^{-1}$).  Before embarking on the full SGPS, we carried out ATCA observations on a small Test Region, covering the range $325\\fdg5 < l < 332\\fdg5$, $-0\\fdg5 < b < +3\\fdg5$. A preliminary image of this region, at a resolution of $\\sim$5~arcmin, was presented by Dickey (1997\\nocite{dic97}). We present here our entire data-set on polarization from the SGPS Test Region, at the final survey resolution of 1~arcmin. In Section~\\ref{sec_obs} we describe the aspects of SGPS observations and data-reduction relevant to the polarimetric component of the survey, while in Section~\\ref{sec_results} we present the distribution of polarized emission and of rotation measure which we have determined from these data. In Section~\\ref{sec_discuss}, we interpret these results in terms of intrinsic polarization and Faraday rotation in the Galactic Plane. The present paper focuses on extended regions of linear polarization; in a separate paper by Dickey \\etal (in preparation), we show how the RMs measured towards point sources in the SGPS Test Region can constrain models of the Galactic magnetic field. ",
        "conclusions": "The ATCA's sensitivity, spatial resolution and spectral flexibility have allowed us to study linear polarization and Faraday rotation in the SGPS Test Region in unprecedented detail. Even though this field covers less than 7\\% of the full survey, a variety of structures are apparent from this preliminary study.  We identify the following main features seen in polarization:  \\begin{itemize}  \\item Diffuse linear polarization seen throughout the Test Region, with structure on scales of degrees, and completely uncorrelated with emission seen in total intensity. The brightest polarized features most likely represent intrinsic structure in the source of emission, but fainter components are best explained as being imposed by Faraday rotation of a uniformly polarized background by foreground material. The rotation measures determined towards this emission generally lie between --100 and 0 rad~m$^{-2}$, which indicates that this emission originates at a distance of $\\sim$3.5~kpc, most likely in the Crux spiral arm of the Galaxy.  \\item Two large regions devoid of polarization, each several degrees in extent, and with no counterparts in total intensity. We have considered one of these voids in detail, and have shown that the depolarization observed through it can be explained by turbulence within it on scales $\\sim0.2$~pc.  We find that the properties of the void are consistent with it being a nearby ($\\sim250$~pc) \\HII\\ region of density $n_e \\sim 20$~cm$^{-3}$ and magnetic field $B \\sim 5$~$\\mu$G, possibly associated with the O star HD 144695.  \\item A halo of depolarization surrounding the \\HII\\ region RCW~94. We interpret this as delineating a region of lower electron density surrounding the \\HII\\ region. We propose that this region corresponds to a halo of ionized material resulting from the interaction of RCW~94 with a surrounding molecular cloud.  \\end{itemize}  The results presented here demonstrate the value of interferometric polarimetry as a probe of the ISM.  The full SGPS will allow us to accumulate many more examples of the types of features we have described here, with which we will be able to better determine their properties and distributions.  Furthermore, we expect that the large-scale properties of polarized emission will be a function of position within the Galactic Plane (e.g.\\ Duncan \\etal 1997a, 1999\\nocite{dhjs97,drrf99}), reflecting differing orientations of the spiral arms with respect to the line of sight. Overall, we anticipate that the SGPS will provide the opportunity for a comprehensive study of the magneto-ionic medium of the inner Galaxy, on scales ranging from sub-parsec turbulence up to the global structure of the spiral arms and disc."
    },
    "0010/astro-ph0010204_arXiv.txt": {
        "abstract": "We present the galaxy-galaxy angular correlations as a function of photometric redshift in a deep-wide galaxy survey centered on the Hubble Deep Field South. Images were obtained with the Big Throughput Camera on the Blanco 4m telescope at CTIO, of 1/2 square degree in broad-band $uBVRI$, reaching $\\sim 24$~mag. Approximately 40,000 galaxies are detected in the survey.  We determine photometric redshifts using galaxy template fitting to the photometry.  Monte Carlo simulations show that redshifts from these data should be reliable out to $z\\sim 1$, where the 4000~\\AA~break shifts into the $I$~band. The inferred redshift distribution, $n(z)$, shows good agreement with the distribution of galaxies measured in the HDF North and the Canada-France Redshift Survey.  After assigning galaxies to redshift bins with width $\\Delta z = 0.33$, we determine the two point angular correlation function in each bin.  We find that the amplitude of the correlation, $A_w$, drops across the three bins to redshift $z\\sim 1$. Simple $\\epsilon$~models of clustering evolution fit this result, with the best agreement for $\\epsilon=0$.  Hierarchical cold-dark-matter models best fit in a low density, $\\Lambda$-dominated universe. ",
        "introduction": "The Hubble Deep Field (Williams et al.\\ 1996; HDFN) presented the deepest view of the universe in what is now the best studied 5 square arcminutes of the sky. With the addition of the Space Telescope Imaging Spectrograph (Woodgate et al.\\ 1998; STIS) and the Near Infrared Camera and MultiObject Spectrometer (Thompson et al. 1998; NICMOS), the opportunity was presented to expand on the HDF program with an ambitious observation of the field around the z=2.25 QSO J2233-6033 (Boyle 1997). This Hubble Deep Field South (Williams et al.\\ 2000; HDFS) provided a spectroscopic investigation of the matter along the line of sight to the QSO (Ferguson et al.\\ 2000), the deepest image ever taken at optical wavelengths (Gardner et al.\\ 2000), and deep optical and IR images of parallel fields (Casertano et al.\\ 2000; Fruchter et al.\\ 2000).  In 1997 September, we obtained broad-band $uBVRI$ imaging of half of a square degree around the HDFS. We have cataloged over 40,000 galaxies\\footnote[5]{The data are publically available on the web at http://hires.gsfc.nasa.gov/$\\sim$research/hdfs-btc/.}. The first results from the project (Palunas et al.\\ 2000a, hereafter Paper I) were a study of QSOs in this field. Here we present a measurement of the evolution of large-scale structure in the universe, using the angular correlations of galaxies as a function of photometric redshifts determined from our multi-color data. We will also publish the catalog in Palunas et al.\\ (2000b; hereafter Paper III).  In the standard hierarchical model of galaxy formation and evolution, galaxy clustering traces overdense regions in the distribution of dark matter.  High-redshift galaxies occur at the most extreme peaks in the density field and are thus biased tracers of mass---i.e., luminous matter is more strongly clustered than dark matter (Kaiser 1984). Subsequently, the clustering of the dark matter increases more rapidly than that of galaxies, so that the two distributions approach one another and are more similar today than in the past.  In other words, the galaxy clustering bias is less.  The way in which the bias decreases is a function of the cosmological parameters $\\Omega, \\Lambda,$~and $P(k)$~(e.g.  Kauffman et al. 1999).  Therefore, by measuring the evolution of the galaxy-galaxy correlation function, we can constrain models of structure formation.  Clustering has often been estimated from catalogues of two-dimensional sky coordinates of galaxies in order to infer the three-dimensional, physical correlation function, $\\xi(r)$, using Limber's equation (e.g., Peebles 1980).  An important recent study of this type is the APM Galaxy Survey (Maddox et al. 1996).  On the other hand, spectroscopic surveys have allowed direct measurement of $\\xi(r)$~for local galaxies over a wide field (e.g. Lin et al.\\ 1996). Studies at high redshift have had to rely on pencil-beam redshift surveys (e.g. Giavalisco et al.\\ 1998) or wide-field photometric surveys in a single filter (Postman et al.\\ 1998).  Photometric redshifts now make it possible to combine the best of both techniques. While not accurate enough to determine $\\xi(r)$, they are sufficient to assign the galaxies to redshift bins and study the evolution of \\acf~with redshift directly, rather than inferring the evolution from varying the magnitude limit (Connolly, Szalay, \\& Brunner 1998; Arnouts et al. 1999; Brunner, Szalay, \\& Connolly 2000).  We have used this method to trace the evolution of clustering to z=1.  In section 2, we summarize the observations and data reduction.  In section 3 we discuss the techniques used in the photometric redshift analysis.  In section 4 we present the properties of the estimated redshifts. In section 5 we discuss the correlation of the cataloged galaxies as a function of redshift.  We present our conclusions in section 6. Unless otherwise noted, a cosmology of $H_0=65$ km s$^{-1}$~Mpc$^{-1}$, $\\Omega_M=0.3$, $\\Omega_{\\Lambda}=0.7$~is assumed. ",
        "conclusions": "\\label{sec:  conclusions}  Our data are consistent with a popular scenario for the evolution of clustering with redshift.  As we noted above, more data will help distinguish between these cases.  Kauffmann et al.\\ (1999) discuss the predictions of hierarchical CDM models on the observed clustering of galaxies.  They explain that in a low density, $\\Lambda$-dominated universe ($\\Lambda$CDM), there is a ``dip'' (their phrasing) in the amplitude of galaxy clustering around $z\\sim 1$.  This dip occurs between the era of strong dark matter clustering traced by unbiased luminous galaxies (low redshift) and the era in which the observable galaxies are highly biased tracers of the densest (earliest clustered) regions.  In a high density universe, the dip is absent, since structures form later and galaxies are more highly biased at higher redshifts; in that case, the clustering strength is constant for $z<1$~and then rises sharply with redshift.  CDM models also predict that the dip is weaker for redder, more luminous galaxies.  Arnouts et al.\\ (1999) demonstrate similar features in a wider array of CDM models, but with the same distinction between high density and low density universes.  In our data, we find evidence for a decline out to $z\\sim 1$, though the lack of higher redshift points prevents us from testing with our data alone whether this is a dip or an indication of a continuing decline. However, a reliable data point at $z\\sim 3$~from the Lyman break galaxies (Giavalisco et al.\\ 1998) suggests that a dip must occur somewhere in the intervening redshift range.  The strength of this effect is magnified by the fact that at higher redshifts we appear to be seeing blue galaxies, for which the dip is expected to be stronger.  Our data are consistent with $\\Lambda$CDM predictions for a low density universe.  Although the strength of the evolution is debated, this scenario has been suggested by several studies (e.g. Postman et al.\\ 1998, Le Fevre et al.\\ 1996).  Using the photometric redshift technique, Arnouts et al.\\ (1999) find this kind of clustering evolution in the HDF (but note that Connolly et al.\\  1998 do not).  These results should be taken as a first look at the power of using the photometric redshift technique to measure clustering evolution on a sample of tens of thousands of galaxies on large scales.  With the new instrumentation now becoming available on medium and large telescopes, this experiment will be repeated with higher accuracy out to larger redshifts.  In particular, substantial improvements will occur when it is practical to obtain deep near-IR data over similar area."
    },
    "0010/astro-ph0010174_arXiv.txt": {
        "abstract": "We have obtained 1.0---2.5$\\mu$m spectra at R$\\sim$600 of 14 disk dwarfs with spectral types M6 to L7. For four of the dwarfs we have also obtained infrared spectra at R$\\sim$3000 in several narrow intervals. In addition, we present new L$^{\\prime}$ photometry for four of the dwarfs in the sample, which allows improved determinations of their bolometric luminosities.  While obtaining the photometry we resolved the L-dwarf Denis-P J 0205-1159 into an identical pair of objects separated by 0\\farcs 35. The spectra, together with the published energy distribution for one other L5 dwarf, are compared to synthetic spectra generated by upgraded model atmospheres.  Good matches are found for 2200$\\ge$$T_{\\rm eff}$~K$\\ge$1900 (spectral types around M9 to L3), but discrepancies exist at $T_{\\rm eff}\\geq$2300~K (M8) and for $T_{\\rm eff}\\leq$1800~K (L4--L7).  At the higher temperatures the mismatches are due to incompleteness in the water vapor opacity linelist.  At the lower temperatures the disagreement is probably due to our treatment of dust: we assume a photospheric distribution in equilibrium with the gas phase and neglect any diffusion mechanisms. We derive effective temperatures for the sample from the comparison with synthetic spectra and also by comparing our observed total intrinsic luminosities to structural model calculations (which are mostly independent of the atmosphere but are dependent on the unknown masses and ages of the targets).  The two derivations agree to about 200~K except for the faintest object in the sample where the discrepancy is larger.  Agreement with other temperature determinations is also $\\pm\\sim$200~K, except for the L7 dwarf. ",
        "introduction": "The last few years have seen rapid advances in both the observational data on low--mass stars and brown dwarfs and in the theoretical models of these objects. During this period we have been obtaining infrared spectra of a sample of halo and disk dwarfs approaching and below the stellar/sub--stellar boundary.  This paper extends our spectroscopic study of M dwarfs presented in \\cite{l99} (hereafter L00) to lower effective temperatures ($T_{\\rm eff} <$ 2500~K). The data are compared to synthetic spectra generated from an upgraded version of Allard \\& Hauschildt's NextGen model atmospheres (\\cite{all98b}).  In \\S 2 we present the target sample and in \\S 3 we describe the instruments used to obtain the spectroscopic data. In \\S 4 we present imaging results:  new L$^{\\prime}$ photometry, confirmation of binarity of Denis-P J 0205-1159, and color--color diagrams. In \\S 5 we describe the spectra and calculate integrated fluxes and bolometric magnitudes. \\S 6 describes the models and the comparison process and the results of the comparison of the data to the synthetic spectra are presented in \\S 7.  In \\S 8 we present derivations of effective temperature based on structural models. Our conclusions are given in \\S 9. ",
        "conclusions": "Our derived temperatures agree  with other derivations of $T_{\\rm eff}$ for early--L type dwarfs within the likely errors. \\cite{pav00} use both the NextGenDusty  and \\cite{tsu00} atmospheric structures with their own spectral synthesis code to fit red spectra; they have to include an unknown additional opacity source, in the form of a power--law, to match the overall spectral shape because they do not include dust in solving for chemical equilibrium,  or account for the opacity of known dust species (note that this is inconsistent with their adopted atmospheric structure). They derive $T_{\\rm eff}$ for four dwarfs in common with us: for BRI 0021-0214 and Kelu-1 their values are higher than our spectral synthesis values by 100~K; for Denis-P J 1228-1547AB and Denis-P J 0205-1159AB their values are 200~K and 700~K lower.  Their value of 1200~K for Denis-P J 0205-1159AB does not seem likely, as we would then expect to see methane features  in the H-- and K--band regions, which are not seen.  \\cite{bas00} use Ames--Cond NextGen model fits to rubidium and cesium lines and derive effective temperatures   within 100~K of our spectral synthesis values for Kelu-1, Denis-P J 1058-1548 and Denis-P J 1228-1547AB; they derive a value 150~K lower than ours for the coolest object in common, Denis-P J 0205-1159AB, and a value 300~K higher for the M9.5 dwarf LP 944-20. The Ames-Cond models include the formation of dust but neglect its opacity (i.e. assume the dust has settled below the photosphere), which corresponds to a limiting case which complements that assumed here of the dust in equilibrium with the gas phase. It is expected that dust clouds will sink lower in the atmosphere with decreasing effective temperature (in the absence of replenishment of material in the upper photosphere), hence the Cond models are likely to be more correct than the Dusty models for the latest L--dwarfs. The Basri et al. lower temperature of 1750~K for Denis-P J 0205-1159AB is likely to be more accurate than that derived here, but the opposite is true at hotter temperatures and so our lower $T_{\\rm eff}$ for LP 944-20 is likely to be more accurate than their value.  Given the discrepancies seen in the model fits to the observed spectral energy distributions for the coolest objects in our sample, it would appear that gravitational settling of dust becomes important at $T_{\\rm eff}\\leq1800$~K.  Based on species condensation arguments, \\cite{k00} argue for an effective temperature around 2000~K for L0 spectral types, which is in agreement with this work.  They also advocate (based on luminosity) $T_{\\rm eff}\\sim$1300~K for the L8 dwarf (by the Kirkpatrick scheme) GL 584C. However the discovery of three dwarfs optically redder than L8 and showing both CO and CH$_4$ in their spectra (\\cite{l00}) suggests that these new later--type objects have $T_{\\rm eff}\\sim$1300~K and by implication L8 spectral types have $T_{\\rm eff}>$1300~K.  A hotter temperature scale also is supported by the preliminary model fits to L--band methane features seen in mid--L dwarfs reported by \\cite{noll00}. However all of these results can be estimates only as they either rely on the correlation between the appearance or disappearance of various species, which depends on the temperature of the line--forming region and not on the effective temperature, or they rely on what are acknowledged to be very preliminary atmospheric models in this temperature range.  In conclusion, the best current estimates are that the M dwarfs cover a range in $T_{\\rm eff}$ of 3700---2100~K, the L dwarfs 2000---$\\sim$1500~K and the known methane (T) dwarfs $\\sim$1300---800~K.  A definitive temperature scale for the L dwarfs will not be possible until model atmospheres which include gravitational settling are available; such work is in progress by the modellers of this group and others.  We note that any object with a spectral type L5 or later  must be substellar based on evolutionary models, and late--M to early--L dwarfs may be substellar depending on their age.  The discovery of binary L--dwarf systems offers a chance to further improve our knowledge of the fundamental parameters for this recently discovered low--mass population of our Galaxy."
    },
    "0010/astro-ph0010497_arXiv.txt": {
        "abstract": "We investigate the density profiles and evolution of weakly self-interacting cold dark matter haloes using a numerical code based on the collisional Boltzmann equation. This approach is alternative to N-body techniques in following the dynamical evolution of haloes in the cosmological context and taking into account particle self-interaction. The physical case with a cross section inversely proportional to the relative velocity of the colliding particles is modelled with an unprecedented resolution, spanning five orders of magnitude on the radius for each halo. The modelled haloes cover a mass range from dwarf galaxies to galaxy clusters. We find that for $\\sigma  v_{100} \\approx 10^{-24}$ cm$^2$/GeV, where $\\sigma$ is the cross section per unit mass and $v_{100}$ is the collision velocity in units of 100 km/s, soft cores in good agreement with observations on galactic as well as on galaxy cluster scales are obtained. Remarkably, the observed nearly invariance of the halo central density with mass is reproduced. ",
        "introduction": "The current Cosmology based on the growth of small fluctuations through gravitational instability in a universe with cold dark matter (CDM) defines a physically well-motivated theory, capable to explain most of the properties of the large-scale structures of the universe. In this scenario, where a large fraction of the matter is supposed non-dissipative, cold and collisionless, the dark haloes form hierarchically via gravitational collapse of primordial density fluctuations. The density profiles of these dark haloes have been studied systematically using N-body simulations (Navarro, Frenk $\\&$ 1996, 1997; hereafter NFW). It was found that a simple analytical density profile with one free parameter is able to describe the typical structure of the virialized CDM haloes. On its own, this free parameter, which can be the concentration, depends only on the halo mass.  The NFW density profile goes as $\\rho \\propto r^{-1}$ at small radii. According to recent high-resolution simulations, the inner density profile of the haloes is even steeper than $r^{-1}$ ($\\rho \\propto r^{-1.5}$) (Moore et al. 1999; Jing \\& Suto 2000). This prediction of the CDM scenario is in potential conflict with observations: the halo inner density profiles inferred from the HI rotation curves of dwarf and low surface brightness (LSB) galaxies and from a mass map constructed by strong lensing techniques for the galaxy cluster CL0024+1654 seem to be much shallower than $r^{-1}$ (Moore 1994; Flores \\& Primack 1995; Burkert 1995; de Blok \\& McGaugh 1997; Tyson et al. 1998). Recently, the rotation curves of some of the dwarf and LSB galaxies were re-obtained using high-resolution $H\\alpha$ observations, resulting the inner density profiles more concentrated than those obtained from the HI observations, but on average still shallower than $r^{-1}$ and with a large scatter (Swaters et al. 2000; van den Bosch \\& Swaters 2000; Dalcanton \\& Bernstein 2000). Besides, as Firmani et al. 2000b have shown, the halo inner density profiles of most of the observed LSB galaxies could be shallower than reported if the baryon contraction is taken into account (see also Firmani \\& Avila-Reese 2000).  An interesting property to be confirmed with more observational studies, in particular for galaxy clusters, is that the central density of dark matter haloes is roughly independent of the halo mass with a value around 0.02 M$_{\\odot}$ pc$^{-3}$ (Firmani et al. 2000a,b). Not only the predictions of the CDM scenario, but also of alternative scenarios as the warm dark matter one are unable to predict such a scale invariance of the central density of dark haloes (Avila-Reese et al. 1998; Col\\'{\\i}n et al. 2000).  In a recent burst of activity, several alternatives to the well motivated CDM scenario were proposed in order to overcome the difficulties mentioned above and others as the excess of satellites and a little angular momentum in the galaxy disc predicted by numerical simulations. These {\\it alternative} theories try to suppress the excess of central mass of the galactic haloes and may be summarized into two classes: 1) a manipulation of the power spectrum with a suppression of the small scales as an intrinsic property of a dark matter: Avila-Reese et al. 1998; Moore et al. 1999; Firmani et al. 2000a,b; Peebles 2000; Col\\'{\\i}n et al. 2000. These works make different assumptions concerning thermal energy at the beginning of the mass aggregation history (MAH). 2) Assuming for dark matter particles a different nature from collisionless: self-interacting (Spergel $\\&$ Steinhardt 2000); repulsive (Goodman 2000); decaying (Cen 2000); with a limiting phase-space density (Hogan $\\&$ Dalcanton 2000) and annihilating (Kaplinghat et al. 2000). Intriguingly, only the assumption of a self-interacting or annihilating cold dark matter seems to be able to reproduce the observed central density  scale invariance ranging from galaxy to galaxy cluster size.  In a previous work, adopting the self-interacting CDM scenario suggested by Spergel \\& Steinhardt, we have  proposed a thermodynamical model able to describe the gravothermal expansion of the core in the haloes (Firmani et al. 2000a); with this model, the scale invariance of the halo central densities was predicted. The NFW density profile shows a velocity dispersion raising outwards with radius (Cole $\\&$ Lacey 1996; Fukushige $\\&$ Makino 1997). Therefore, if particles are self-interacting, the inward heat transfer induces a thermalization process able to avoid the formation of a cuspy core. The influence of {\\it weak} self-interaction on the structure of the CDM haloes has been recently explored using cosmological numerical simulations (Yoshida et al. 2000; Dav\\'e et al. 2000). These works, where constant cross sections were used, have led to interesting results, that we will discuss in comparison with ours, later.  A cross section inversely proportional to the particle velocity dispersion seems to be better motivated by observations (Firmani et al. 2000a,b; Wyithe et al. 2000). In this paper we focus on detailed dynamical predictions of the density profiles of self-interacting spherical symmetric dark matter haloes ranging from galaxy to galaxy cluster mass. The cosmological model we assume is flat with vacuum density $\\Omega_{\\Lambda}=0.7$ and expansion rate $H_0=65$  km  s$^{-1}$ Mpc$^{-1}$. ",
        "conclusions": "We have calculated the evolution of the density profiles of self-interacting CDM haloes using a numerical approach based on the collisional Boltzmann equation. This study presents two original points: ({\\it i}) we were able to explore the halo density profiles covering about five orders of magnitude in the radius and for haloes from galaxy to galaxy cluster scales; ({\\it ii}) we used a self-interacting cross section inversely proportional to the collision velocity of the particles. Analysing the results obtained with our code we conclude: \\begin{itemize} \\item [(1)] The interior of dark haloes may be deeply affected by self-interaction. A modest cross section value ($\\sigma v_{100} \\approx 10^{-24}$ cm$^2$ GeV$^{-1}$) is enough to produce soft cores. \\item [(2)] Collisions between dark particles induce a thermalization process with the consequent appearance of a soft core. Thermal equilibrium in the core is established with a few collisions in a Hubble time ($\\approx 4$). \\item [(3)] Soft cores appear already at the very early dynamical evolution ($z \\ge 40$); with time the central density decreases and the core size increases. \\item [(4)] The observed scale invariance of the central density from galactic to galaxy cluster scales is reproduced only if the cross section is inversely proportional to the collision velocity. \\item [(5)] The core radius predicted by the model increases proportional to the halo maximum rotation velocity in agreement with the observed trend. \\item [(6)] The observed central density of CL0024+1654 is consistent with assuming a self-interacting dark matter cross section given by: $\\sigma v_{100} \\approx  10^{-24}$ cm$^2$ GeV$^{-1}$. \\end{itemize} The code is limited in analysing dark halo properties which depart from spherical symmetry. Further studies concerning dark halo triaxiality and the evolution of substructures have been planned using N-body techniques."
    },
    "0010/astro-ph0010342_arXiv.txt": {
        "abstract": "The division of galaxies into ``barred'' (SB) and ``normal'' (S) spirals is a fundamental aspect of the Hubble galaxy classification system. This ``tuning fork'' view was revised by de Vaucouleurs, whose classification volume recognized apparent ``bar strength'' (SA, SAB, SB) as a continuous property of galaxies called the ``family''. However, the SA, SAB, and SB families are purely visual judgments that can have little bearing on the actual bar strength in a given galaxy. Until very recently, published bar judgments were based exclusively on blue light images, where internal extinction or star formation can either mask a bar completely or give the false impression of a bar in a nonbarred galaxy.  Near-infrared camera arrays, which principally trace the old stellar population in both normal and barred  galaxies, now facilitate a quantification of bar strength in terms of their gravitational potentials and force fields.  In this paper, we show that the maximum value, $Q_b$, of the ratio of the tangential force to the mean axisymmetric radial force in a barred disk galaxy is a quantitative measure of the strength of a bar. $Q_b$ does not measure bar ellipticity or bar shape, but rather depends on the actual forcing due to the bar embedded in its disk.   We show that a wide range of true bar strengths characterizes the category ``SB'', while de Vaucouleurs category ``SAB'' corresponds to a much narrower range of bar strengths. We present $Q_b$ values for 36 galaxies, and we incorporate our bar classes into a dust-penetrated classification system for spiral galaxies. ",
        "introduction": "The presence of a bar in a disk galaxy implies a non-axisymmetric gravitational field.  The high frequency of occurrence of bars (over 65 percent, de Vaucouleurs 1963; Eskridge et al. 2000) and the fact that bars principally consist of an old stellar population \\citep{dev55,dvv59,fre89,elm85} implies that bars are fundamental components in the distribution of mass in galaxies.  Bars are believed to be the ``engines'' driving a wide variety of secular evolution processes in galaxy dynamics \\citep{pfe96}. Bar driven secular evolution appears to make significant changes in galaxy structure over a Hubble time. The main components found in barred galaxies include bulges, disks, lenses, and inner and outer rings -- and of course the bar itself \\citep{san61,dev59,kor79,sel93,but95,bco96}.  The absence or presence of a bar led \\citet{hub26} to develop two separate prongs to his classification tuning fork: the sequence of normal (unbarred) spirals Sa, Sb, and Sc, paralleled by a sequence of barred spirals SBa, SBb and SBc. \\citet{dev59} recognized that galaxies such as NGC 5236 (M83) showed a bar morphology intermediate between that of a normal and a barred spiral. He introduced the notation SA for unbarred spirals so that he could use the combined notation SAB for transitional cases like M83.  In this paper, we recognize a full continuum of ``bar strengths'', as does de Vaucouleurs, but we propose a numerical quantification of bar strength based on the gravitational forcing of the bar itself, not on visual appearance. From a sample of 36 galaxies, we recognize seven bar strength classes: bar class 0 galaxies, which are normal spiral galaxies without any bar; bar class 1 and 2 galaxies, which show ovals and weak bars (which de Vaucouleurs would have classified in his SAB class); and bar class 2-6 galaxies, which encompass all galaxies classified as SB by Hubble and de Vaucouleurs.  Our numerical quantification of bar strength is derived from the non-axisymmetric force field of a galaxy inferred from the near-infrared light distribution. Near-infrared $H$-band (effective wavelength 1.6$\\mu m$) and $K$-band (effective wavelength 2.2 $\\mu m$) images beautifully reveal the old stellar population or `backbone' of spiral galaxies \\citep{fro96,blo94,blo99}. The near-infrared light comes principally from old giant and supergiant stars \\citep{fro96}. The extinction at $H$ and $K$ is only 0.1-0.2 times that in visual light, so that dust has only a minimal effect on the inferred potentials. A rich duality of spiral structure has been found from studies of optical and near-infrared images; a spiral galaxy may present two completely different morphologies when examined optically and in the near-infrared \\citep{elm99,blo99}.  In the optical, dust often hides bars, as in the Milky Way. Seventy percent of spirals classified in the {\\it Carnegie Atlas of Galaxies} \\citep{san94}, based on Hubble bins, are classified as unbarred. This fraction drops to 27\\% when these galaxies are imaged in the near-infrared \\citep{esk99}. The remaining 43 percent show ovals or bars in the dust-penetrated regime. This high percentage of bars in the near-infrared agrees with the findings of \\citet{sei98}, who found that 90\\% of a sample of 45 galaxies showed some evidence of a bar in the $K$-band. ",
        "conclusions": "We have outlined the derivation of a fully quantitative bar strength class for spiral galaxies, based on the the maximum value of the amplitude of the tangential force to the mean radial force. Although the method is still based on light fluxes, since we use images to infer gravitational potentials, it provides a more direct handle on bar strength than any other light-based methods so far applied. Regardless of what the bar may visually look like, the ratio map will show a pattern of four maxima/minima that can be isolated fairly automatically to give a robust measure of bar strength. The method is thus free from uncertainties connected with defining bars or of full bulge/disk decompositions. It can be applied quickly and efficiently to many galaxies, and only in some cases is it necessary to give special treatment to the bulge if it suffers too much deprojection stretch. Bulge shape, dark matter, and star formation have little impact on the bar strength class for our sample galaxies. The most important effects are vertical scale height and, for highly inclined galaxies, deprojection uncertainties.  The method has much room for refinement. For example, bulges can be decomposed to eliminate deprojection stretch and their influence on the ratio maps more accurately accounted for. Improved orientation parameters can be used or derived for galaxies as data become available. Appropriate vertical scaleheights could be inferred more reliably from detailed near-infrared surface photometry and the type dependence of $h_R/h_z$. The validity of the constant $M/L$ assumption could be tested by measuring $J-K$ color distributions. Finally, deprojection uncertainties go beyond simple bulge deprojection stretch to the fundamental uncertainty of deprojecting any galaxy, even if a bulge is not present. All of these are issues which will be considered in a more detailed forthcoming study.  The addition of our bar strength parameter to the dust-penetrated classification scheme of \\citet{blo99} now gives a full dynamical appreciation of the range of evolved stellar disks with bars (Figure~\\ref{tuning}).  We thank A. Quillen, F.Combes, A. Kalnajs, and S. M. G. Hughes for programs that were used in this study. We also thank A. Quillen, F.Combes, B.Elmegreen, K. Freeman, P. van der Kruit, A. Kalnajs, R. de Grijs, R.I. Thompson and especially the (anonymous) referee for helpful and timely advice and comments which improved this study. D. Elmegreen, I. Puerari, P.Grosb{\\o}l, J. Knapen, K. C. Freeman, O. K. Park, and T. Jarrett very graciously loaned us some images which we used to test the bar strength method. We thank D. A. Crocker for her assistance in producing the ``tuning fork'' diagram in Figure~\\ref{tuning}. We gratefully acknowledge the support of the Anglo-American Chairman's Fund, which also made the visit to South Africa of one of us (RB) possible. We are indebted to CEO Mrs. M. Keeton and the Board of Trustees."
    },
    "0010/astro-ph0010483_arXiv.txt": {
        "abstract": "We present multiwaveband photometric and optical spectropolarimetric observations of the $R=15.9$ narrow emission line galaxy R117\\_A which lies on the edge of the error circle of the \\ROSAT X-ray source R117 (from \\citeNP{McHardy98}).  The overall spectral energy distribution of the galaxy is well modelled by a combination of a normal spiral galaxy and a moderate-strength burst of star formation.  The far infra-red and radio emission is extended along the major axis of the galaxy, indicating an extended starburst.  On positional grounds, the galaxy is a good candidate for the identification of R117 and the observed X-ray flux is very close to what would be expected from a starburst of the observed far infra-red and radio fluxes.  Although an obscured high redshift QSO cannot be entirely ruled out as contributing some fraction of the X-ray flux, we find no candidates to $K=20.8$ within the X-ray errorbox and so conclude that R117\\_A is responsible for a large fraction, if not all, of the X-ray emission from R117.  Searches for indicators of an obscured AGN in R117\\_A have so far proven negative; deep spectropolarimetric observations show no signs of broad lines to a limit of one per cent and, for the observed far infra-red and radio emission, we would expect a ten times greater X-ray flux if the overall emission were powered by an AGN.  We therefore conclude that the X-ray emission from R117 is dominated by starburst emission from the galaxy R117\\_A. ",
        "introduction": "Despite extensive investigation, there remains much uncertainty regarding the origin of the X-ray emission from narrow emission line galaxies (NELGs) found in deep X-ray surveys (\\citeNP{CCRSS1}; \\citeNP{McHardy98}; \\citeNP{Schmidt98}; \\nocite{Roche95} \\citeANP{Roche96} 1995, 1996).  There are several possible emission mechanisms, where the two most plausible are starburst activity or the presence of an active galactic nucleus (AGN).  A burst of intense star-forming activity creates large numbers of hot OB stars, which are strong X-ray emitters, and such galaxies have been proposed as a significant contributor to the X-ray Background (XRB) Radiation (\\eg~\\citeNP{GP90}).  It is also generally agreed that obscured AGN can account for the majority of the energy in the XRB (\\nocite{M94}Madau, Ghisellini \\& Fabian 1994; \\citeNP{C95}).  AGN have intrinsically soft spectra, but photo-electric absorption by gas surrounding the AGN has the effect of preferentially removing lower energy photons, thereby hardening the observed spectra to agree with the spectrum of the XRB.  Results from deep \\ROSAT X-ray surveys show that the number of X-ray sources at faint fluxes continues to rise faster than the contribution from QSOs and clusters (\\citeNP{Ioannis96}; \\citeNP{McHardy98}).  Through accurate positions from the \\ROSAT {\\em HRI} \\cite{Has98}, many of these faint X-ray sources appear to be unambiguously identified with galaxies showing narrow emission lines in their optical spectra.  It is difficult to determine the dominant power source from the usual line ratio diagnostics \\cite{Osterbrock89} since many of these objects lie on the border between AGN and starburst, and it is also likely that the two phenomena may coexist in some objects (\\citeNP{Iwasawa97}; \\citeNP{McHardy98}).  Here we present multiwavelength imaging and photometry of R117, a \\ROSAT source from the UK Deep Survey of \\citeANP{McHardy98} (1998), identified with a NELG at $z=0.061$.  These observations include spectropolarimetry, to search for broad emission lines in polarized light as a sign of AGN activity.  In addition, submillimetre (sub-mm) photometry aims to detect thermal emission from dust associated with the source of the X-ray flux, whether due to starburst activity within the galaxy or the presence of an obscured AGN.  The available data point towards a pure starburst origin to the soft X-ray emission in R117, which implies that star-forming galaxies may make a significant contribution to the soft X-ray background radiation. ",
        "conclusions": "\\label{sec:concl}  Here we have presented a multiwavelength study of the properties of the \\ROSAT X-ray source R117, from the radio to the X-ray.  Having ascertained that the $R=15.9$ galaxy, R117\\_A, is the most likely source of the X-ray emission, we have used a variety of techniques in order to determine the dominant process behind this activity.  The SED of the galaxy is well modelled by a spiral galaxy with a moderate burst of star-formation.  Optical photometry and morphology are consistent with R117\\_A being a spiral galaxy, with a central bar structure or nuclear star-forming region.  This galaxy counterpart to the X-ray source is highly plausible on positional grounds, and in the absence of any strong alternatives.  The ratios between the X-ray, FIR and radio fluxes are consistent with this emission being due to a starburst, as if an AGN were the dominant source of the power in this object, the X-ray flux would be ten times larger.  Spectropolarimetric observations of R117\\_A place strong limits on the presence of any broad components to the optical emission lines which would be expected from AGN activity.  The starburst nature of R117 is also consistent with both the detection of weak radio emission and the radio-to-submm spectral index.  From the multiwavelength photometric data, combined with the lack of any broad emission line component in polarized light, we therefore conclude that starburst activity is the dominant source of the {\\em soft} X-ray flux detected by \\ROSAT from the galaxy R117.  The broader implications of these observations concern the identification and classification of X-ray sources found in deep X-ray surveys.  In particular, galaxy identifications have been criticised as being misidentifications in \\ROSAT X-ray surveys and it has been claimed that there is no starburst contribution to deep X-ray surveys (e.g. \\citeNP{Lehmann2000}).  Here we show that, in at least one case, a starburst galaxy is by far the most likely identification of a faint \\ROSAT X-ray source.  We also note that \\citeANP{Fiore2000} (2000) find one definite starburst galaxy in a sample of a dozen \\Chandra sources.  Although we agree with the hypothesis that obscured AGN are probably major contributors to the XRB, and that high luminosity NELGs are probably mostly powered by an AGN, we conclude that low luminosity X-ray emission from starburst galaxies cannot be entirely neglected as contributing to the X-ray background, particularly at softer energies."
    },
    "0010/astro-ph0010160_arXiv.txt": {
        "abstract": "We report on the first attempts at searching for CO in red low surface brightness galaxies, and the first detection of molecular gas in a low surface brightness ($\\mu_B(0)_{obs}>23$ mag arcsec$^{-2}$) galaxy.  Using the IRAM 30m telescope, CO(1-0) and CO(2-1) lines were searched for in four galaxies -- P06-1, P05-5, C05-3, \\& C04-2. In three of the galaxies no CO was detected, to T$_{MB}\\sim$1.8mK (at the 3$\\sigma$ level). In the fourth galaxy, P06-1, both lines were detected. Comparing our findings with previous studies shows P06-1 to have a molecular-to-atomic mass ratio considerably lower than is predicted using theoretical models based on high surface brightness galaxy studies. This indicates the N(H$_2$)/$\\int{T(CO)dv}$ conversion factor for low surface brightness galaxies may currently be consistently underestimated by a factor of 3 - 20. ",
        "introduction": "Recently it has become clear that there exists a large population of low surface brightness (LSB) galaxies which previously had escaped detection (e.g. \\cite{ibm88}; \\cite{schom92}; \\cite{mds99}; O'Neil, Bothun, \\& Cornell 1997a; \\cite{ohb00}). The importance of LSB galaxies in the local universe is emphasized by studies showing that the majority of the galaxies in the local universe are LSB galaxies. Realizing then that most galaxies are optically diffuse, it becomes important to understand these systems if we wish to understand galaxy formation and evolution as a whole. This study is made difficult, though, by the lack of data on LSB galaxies.  In particular,  in spite of the importance of LSB galaxies there is at the present time little information about the molecular ISM in these systems.  To date four studies have been undertaken to look for CO in low surface brightness systems.  The first two studies -- \\cite{schom90} and \\cite{dv98} -- focused on galaxies with $(B - V)\\:\\leq\\:0.6$ and achieved 3$\\sigma$ detection limits of T$_{MB}$=8.4 -- 24.3 mK (with the NOAO 12m telescope and the JCMT). The next two studies (Knezek 1993; Braine, Herpin, \\& Radford 2000) were of giant LSB systems.  In all four cases no CO was detected. A fifth, very recent, study by Matthews (2000) is also well worthy of mention. This study was designed to look for CO in late-type, edge-on galaxies. CO was detected for a sample of galaxies, a subset of which, if de-projected from their edge-on orientation, may be LSB systems.  The recent discovery of red ($B-V\\geq 1$) LSB galaxies (\\cite{oneil97}) has brought the idea of searching for molecular gas in LSB systems back to the forefront.  As red colors should be indicative of reasonably high metallicities and therefore should insure against CO depletion (for example, as is seen in low metallicity dwarf galaxies), it seemed worthwhile to make a fifth attempt at CO detection in these diffuse systems. ",
        "conclusions": "Assuming a Salpeter-like IMF, near-IR observations by \\cite{bell99} of red LSB galaxies indicate they should have metallicities in the $\\sim$\\Zsol\\ range.  When combined with a theoretical study of the molecular ISM in LSB galaxies (MSM), this indicates the CO intensity of the red LSB galaxies should be above 1.0 K km s$^{-1}$ (for the J=1--0 transition). (It should be noted that the \\cite{bell99} paper accurately predicts the values found by YK89 in their observations of HSB spiral galaxies.) The MSM paper also provides a prediction of the observed \\HHI\\ for \\Zsol\\ galaxies ranging from 1 -- 3, depending on the structure of the galaxies' ISM. For the three non-detections in our sample, both the observed $\\int{T(CO)dv}$ and \\HHI\\ fall well short of the predicted value.  Our one detection, though, matches the predicted $\\int{T(CO)dv}$ value well but has a \\HHI\\ value a factor of 6 -- 20 too small.  A number of factors may be contributing to the discrepancy between our observe values (and upper limits) and those predicted by both MSM and \\cite{bell99}: \\begin{enumerate}  \\item If the CO extent is considerably larger than the beam size, our observed values could be underestimated. It is doubtful, though, that this idea plays much of a role. Even if the observed $\\int{T(CO)dv}$ value of P06-1 is half the actual value, this could only increase M$_{H_2}$/M$_{HI}$ by a factor of 2, and would most likely increase it by considerably less. Additionally, the three non-detections have optical (d$_{27}$) sizes of less than 36\\arcsec, insuring that most of each galaxy would fit within the beam.  \\item If the metallicity of the LSB galaxies is considerably lower than that predicted by \\cite{bell99}, the galaxies will have fewer dust grains to drive molecule formation, resulting in lower than average molecular content.  Additionally, low metallicity results in a low UV attenuation. This implies that the H$_2$ (which is strongly self-shielding) would not be much affected by the lower metallicity, while the CO would more readily be destroyed (i.e.  Schombert \\etal\\ 1990). Thus X, the H$_2$ to CO ratio, would increase.  For our observed galaxies,the theoretical predictions of MSM show that decreasing the metallicities to Z=0.3\\Zsol\\ will both increase X by a factor of $\\sim$2 and decrease \\HHI\\ by 5.  Simultaneously, $\\int{T(CO)dv}$ is reduced  to 0.1 -- 0.3 K \\kms\\ (depending on the galaxies' central surface brightness). With these values the upper limits on $\\int{T(CO)dv}$ for C04-2 and C05-3 could readily fall onto the values predicted by MSM for \\HHI\\ and could come considerably closer to their predicted $\\int{T(CO)dv}$ values.  The upper limits of P05-5, though, are still short of the predicted \\HHI\\ value by a factor of 3.  Attempting to match P06-1 into MSM's predictions through lowering its metallicity also has problems.  As the predicted and observed values of $\\int{T(CO)dv}$ match, lowering the galaxy's metallicity to 0.3\\Zsol\\ can allow the predicted and observed \\HHI\\ values to match, but the observed $\\int{T(CO)dv}$ value is then too high by a factor of 10.  \\item The third factor which should be considered is the homogeneity of the ISM.  MSM show that the galaxies with `clumpier' ISM have higher values of X than their homogeneous counterparts, as the denser regions are better shielded from the background interstellar radiation field.  \\item The last item to consider in these calculations is the efficiency of the galaxies in converting atomic hydrogen into molecular hydrogen. If the velocity dispersion of LSB galaxies is higher than in their high surface brightness counterparts, the formation and growth of molecular clouds could be adversely affected.  This would result in lower molecular content in LSB systems. This idea is supported by the fact that the one LSB galaxy which has been detected in CO also has the highest HI gas mass (and presumably the highest total mass) of any LSB galaxy in this study. \\end{enumerate}  As can be seen from the above arguments, no one parameter can readily account for both the observed $\\int{T(CO)dv}$ values/limits and for the low \\HHI\\ values which are observed both in this project and in previous searches for CO in LSB galaxies.  If MSM's  prediction for $\\int{T(CO)dv}$ is correct, then we are handed a dilemma -- either the four galaxies observed in this project are homogeneous, in which case one of the galaxies (P06-1) has a metallicity 10 times that of the other three galaxies, in spite of their similar colors, or the galaxies have a clumpy structure and their metallicities are similar and all are a factor of 3 -- 10 times fainter than that predicted by \\cite{bell99}.  In the latter case the models of MSM predict that P06-1 has considerably more high density regions than the other three homogeneous galaxies.  Assuming the MSM values for $\\int{T(CO)dv}$ are correct, at least to within a factor of $\\sim$3 (to account for possible variations in the HI$\\rightarrow$H$_2$ conversion rate), we have to face the issue that the previously used value of N(H$_2$)/$\\int{T(CO)dv}$ is low by a factor of 3 -- 20.  due to the large number of uncertainties in X, it is quite likely that this interpretation is correct, at least to some degree.  It should be emphasized, though, that such a high value of X may be indicative of a different evolutionary path or processes in LSB galaxies.  Clearly there are still a large number of questions yet to be answered about these systems before a true understanding of their stellar structure and evolution can be found.  Perhaps the most important issue is a determination of the metallicities of the red LSB galaxies so that more accurate comparisons between these objects and the models of MSM can be obtained.  Regardless, though, two facts seem to be clear from our CO observations.  First, the observed CO flux intensities of LSB galaxies are well below predictions made which assume LSB and HSB galaxies to evolve in an identical fashion.  Second, even the higher CO/H$_2$ conversion factor used by Schombert \\etal\\ (1990) and de Blok \\etal\\ (1998) is most likely too low by a factor of at least two, resulting in higher than previously believed limits on M$_{H_2}$ in LSB galaxies."
    },
    "0010/astro-ph0010026_arXiv.txt": {
        "abstract": "We explain the H$\\alpha$ emission of the cometary knots in the Helix Nebula (NGC 7293) with an analytical model that describes the emission of the head of the globules as a photoevaporated flow produced by the incident ionizing radiation of the central star. We compare these models with the H$\\alpha$ emission obtained from the HST ({\\it Hubble Space Telescope}) archival images of the Helix Nebula. From a comparison of the H$\\alpha$ emission with the predictions of  the analytical model we obtain a rate of ionizing  photons from the central star of about $\\rm 5 \\times 10^ {45} \\ s^{-1}$, which is consistent with estimates based on the total H$\\beta$ flux of the nebula. We also model the tails of the cometary knots as a photoevaporated wind from a neutral shadow region produced by the  diffuse ionizing photon field of the nebula. A comparison with the HST  images allows us to obtain a direct determination of the value of the diffuse ionizing flux. We compare the ratio of diffuse to direct stellar flux as a function of radius inside an HII region with those obtained from the observational data through the analytical tail and head wind model. The agreement of this model with the values determined from the observations of the knots is excellent. ",
        "introduction": "The small scale structure of the Planetary Nebula (PN) known as the Helix Nebula (NGC 7293, PK 36-57$^\\circ$1) is characterized by many thousands of small knots. These knots (which were first reported  by Vorontsov--Velyaminov 1968) have a cometary shape, with their tails pointing away from the central source. Groundbased work (Meaburn et al. 1992, 1996, 1998) and Wide Field and Planetary Camera HST observations (O'Dell \\& Handron 1996; O'Dell \\& Burkert 1997; Burkert \\& O'Dell 1998) revealed a multitude of spatially resolved knots, about 3500, with highly symmetric appearance. The knots have also been detected in CO (Huggins et al.~1992) and, at least outside the main nebula, in C~I (Young et al.~1997).  This PN is one of the closest to us with a parallax distance of 213~pc (Harris et al. 1997). Other methods have been applied to determine the distance to this PN, giving  values ranging from  120 to 400 ~pc. Optical images show a complicated morphology characterized by a helical structure in H$\\alpha$ and [N~II] and a more elliptical shape in [O~III]. The deprojection of these images has proven to be difficult, with suggestions of both elliptical/toroidal shapes and bipolar ones (Meaburn \\& White~1982) being made. O'Dell (1998) has  suggested that the ionized region is a disk.  The central star is very hot and has a low luminosity, indicating that it is well down the cooling track of its post-AGB evolution. The temperature of the central star of the Helix has been measured by the H$\\beta$ Zanstra method to be $\\simeq 1.2 \\times 10^5\\, \\rm {K}$ (G\\'orny, Stasi\\'nska \\& Tylenda 1997).  The high effective temperature combined with the fairly low luminosity and the large size of the PN , about 0.5~pc  in  diameter at a  distance of 213~pc,  leads to a complicated ionization structure with co-existing regions of high and low ionization character (O'Dell 1998; Henry, Kwitter \\& Dufour 1999). The low luminosity of the star also implies a very low density wind, making it virtually undetectable. However, with its expected high velocity ($v_\\infty \\simeq 6000$~km~s$^{-1}$ if radiation driven) it could still have a dynamical effect on the PN structure.  The measured densities in the nebula are low, about  60~cm$^{-3}$ (O'Dell 1998), in contrast with the densities derived for the neutral/molecular globules, $n_{\\rm knot}\\sim 10^6$~cm$^{-3}$ (O'Dell \\& Handron~1996). This leads to the curious result that a substantial fraction of the mass of the PN is in the form of these neutral globules. The aspherical shape of the PN is also found in the distribution of the knots and in the large scale CO emission. Meaburn et al.~(1998) have shown that the knots follow the same velocity distribution as the CO, although at a lower typical expansion velocity. Young et al.~(1999) found the deprojected expansion velocities of the knots and the CO ring to be 19 and 29 $\\rm km~s^{-1}$, respectively.  Some other authors have treated the photoevaporation of clumps in an ionizing radiation field. Bertoldi \\& McKee (1990) developed an approximate analytic theory of the evolution of a photoevaporating cloud exposed to the ionizing radiation of a newly formed star, finding an equilibrium cometary cloud configuration. Johnstone, Hollenbach \\& Bally (1998) modelled the photoevaporation of dense clumps of gas by an external source of ultraviolet radiation including thermal and dynamical effects. Mellema et al. (1998) study the evolution of dense neutral clumps located in the outer parts of a planetary nebula (like  the Helix nebula).  \\begin{figure*} \\begin{center} \\epsfig{file=Fig1.eps,width=5.8cm} \\end{center} \\figcaption{Schematic diagram of the cometary head. $F_h$ is the incident flux at the surface of the neutral knot, and $R_h$ and $n_h$ are the radius and the density of the neutral knot (respectively). \\label{fig1}} \\end{figure*} ",
        "conclusions": "In this paper we propose an analytical model for the heads of the cometary globules in the Helix Nebula, in which the emission is assumed to come from a flow which is being photoevaporated from the surface of a neutral clump. Using the H$\\alpha$ emission obtained from HST images of the Helix Nebula (extracted from the HST archive, see also O'Dell \\& Handron 1996; O'Dell \\& Burkert 1997) and the predictions from our analytic model, we can calculate the ionizing stellar photon flux at the positions of the knots.  We find that the knot brightnesses are fully consistent with the ionizing photon luminosity of ${S_{\\star}} = 5.25\\times 10^{45} ~\\rm s^{-1}$ for the central star, deduced from the total H$\\beta$ flux of the nebula. This value, however, is somewhat uncertain as in the present work we have made no attempt to substract the [N II] 6583 flux from  the f656n filter. Given the rather strong [N II] emission of some of the Helix Knots, our H$\\alpha$ flux determinations could in some cases lie up to $\\approx$30 % The angular size of the knots has a value of $<R_h> = (0.68 \\pm 0.07) ~''$ (where $R_h$ is the radius of the neutral clump forming the knot, see section 3.1). From our small sample of 26 knots, we find no correlation between $R_h$ and distance from the central star (see Figure~\\ref{fig4}).  The photoevaporation model also predicts masses for the knots, which come out fully consistent with observationally derived masses. Their evaporation time is between several thousands to 10$^4$ years, dependent on the temperature of the neutral gas in the knots.  In this paper we also present a model for the cometary tails as a flow which is being photoevaporated from a neutral, cylindrical ``shadow region'' behind the neutral clumps which form the knots. This flow is photoevaporated by the impinging diffuse ionizing field, formed in the surrounding nebula. As we have done for the heads of the cometary globules, we compare the emission of the cometary tails with the predictions from our model in order to quantitatively determine the value of the diffuse ionizing photon flux at the positions of the globules.  Knowing the direct stellar radiation and the diffuse flux at different distances from the central star (i.e., at the positions of the cometary globules), we can then calculate the diffuse-to-direct ionizing flux ratio ($F_d/F_*$) as a function of distance from the central star. We compare the data with simple models for the diffuse ionizing field (Henney 2000) and conclude that the diffuse field is intermediate in magnitude between that expected from a filled and shell-like geometry, which may be due to the disk-like nature of the Helix nebula."
    },
    "0010/astro-ph0010095_arXiv.txt": {
        "abstract": "We have obtained new CCD {\\it BV\\/} photometry of the little-studied southern Galactic globular cluster NGC\\,5986, including light curves of 5 of its RR~Lyrae variables. The cluster's red giant branch bump is detected for the first time at $V = 16.47 \\pm 0.03$.  We derive a reddening and true distance modulus of $E(B-V)$~=~0.29~$\\pm$~0.02 and $(m-M)_{0}$~=~15.15~$\\pm$~0.10, respectively. The cluster's color-magnitude diagram reveals a mostly blue horizontal branch, like that of M13 or M2, and quite unlike M3; yet all of these clusters have nearly identical metallicities ([Fe/H]$_{CG97}$ = $-1.35$). We show that the RR~Lyrae variables in NGC\\,5986 are about 0.2 mag brighter on average than those in M3, an important exception to the often-employed, universal $M_V$(RR)--[Fe/H] relation. Finally, we note that NGC\\,5986 contains two luminous stars with spectral types A-F, which are likely to be post-asymptotic-giant-branch (PAGB) objects.  The $V$-band luminosity function (LF) of such yellow PAGB stars is a promising standard candle. We suggest that the LF is sharply peaked at $M_V$(PAGB)~=~$-3.28 \\pm 0.07$. ",
        "introduction": "Population~II stars evolving off the asymptotic giant branch (AGB) and passing through spectral types F and A are excellent candidates for a new extragalactic standard candle (Bond 1997, 2001). These ``yellow'' post-AGB (PAGB) stars are the visually brightest members of old populations. They should have a narrow luminosity function (LF), because essentially a single main-sequence turnoff mass is feeding the PAGB region of the HR diagram. They are also easily recognized because of their enormous Balmer jumps. A zero-point calibration for the luminosities of Population~II yellow PAGB stars may be set with those in Galactic globular clusters (GCs).  The aim of this paper is to obtain the color-magnitude diagram (CMD) and distance of the little-studied southern GC, NGC\\,5986.  This cluster is remarkable because it contains two candidate A-F type PAGB stars, discovered some years ago during a photographic grism survey (Bond 1977). Membership of the two PAGB stars in the cluster is confirmed by their radial velocities and by Str\\\"omgren photometry showing very low surface gravities (Bond 2001). This hitherto obscure cluster thus may be destined to fill a role comparable to that of the handful of galaxies that have produced more than one Type~Ia supernova. Since the cluster contains two PAGB stars, questions of cluster distance and reddening drop out; hence their observed $V$ magnitudes will provide direct evidence on the width of the Population~II yellow PAGB LF.   The CMD of NGC\\,5986 was first studied by Harris, Racine, \\& DeRoux~(1976) with photographic plates. A search for variable stars in the cluster was made by Liller \\& Lichten (1978), also with photographic plates. We present new CCD {\\it BV\\/} photometry of NGC\\,5986, including new light curves of the cluster's RR~Lyrae variables. Comparisons of our modern CCD photometry are made with the older studies.  We provide new estimates of the cluster's reddening and distance, and we derive the luminosities of the two candidate PAGB stars.  In addition to our study of the PAGB stars, it is also interesting to compare the CMD and the RR~Lyrae stars of NGC\\,5986 to those of other Galactic GCs. Such comparisons remain an active area of research because they are relevant to the interrelated problems of the Oosterhoff dichotomy (Oosterhoff 1939) and the ``second-parameter'' effect (Sandage \\& Wildey 1967), with implications for the formation of the Galaxy (e.g., Lee, Demarque, \\& Zinn 1990, 1994). For example, a new, third Oosterhoff class may be emerging (Pritzl et al.~2000), which is not easily understood in the context of canonical stellar evolution theory.  Also, new exceptions to a universal luminosity-metallicity relation for RR~Lyrae stars have been found in both metal-rich (Layden et al.~1999; Pritzl et al.~2000) and metal-poor (Lee \\& Carney 1999; Clement \\& Shelton 1999) clusters. We discuss NGC\\,5986 in the context of these recent developments. ",
        "conclusions": "We have presented a new $BV$ color-magnitude diagram of NGC\\,5986, as well as new $BV$ light curves for 5 of the cluster's RR~Lyrae variables.  We have detected the RGB-bump for the first time on this cluster's giant branch.  Our analysis of the color-magnitude diagram yields new estimates for the reddening and true distance modulus: $E(B-V) = 0.29 \\pm 0.02$ and $(m-M)_0 = 15.15 \\pm 0.10$, respectively.  In comparison with the globular cluster M3, which has the same metallicity as NGC\\,5986, we find that the RR~Lyrae stars in the latter are about 0.2 mag brighter. This is an important exception to a universal $M_V$(RR)--[Fe/H] relation in a metal-poor cluster.  Our $BV$ photometry of the two candidate yellow PAGB stars in NGC\\,5986 suggests that the $V$-band LF of these stars is sharply peaked at $M_V$(PAGB) = $-3.28 \\pm 0.07$."
    },
    "0010/astro-ph0010576_arXiv.txt": {
        "abstract": "We revisit the idea that density-wave wakes of planets drive accretion in protostellar disks.  The effects of many small planets can be represented as a viscosity if the wakes damp locally, but the viscosity is proportional to the damping length.  Damping occurs mainly by shocks even for earth-mass planets.  The excitation of the wake follows from standard linear theory including the torque cutoff. We use this as input to an approximate but quantitative nonlinear theory based on Burger's equation for the subsequent propagation and shock. Shock damping is indeed local but weakly so.  If all metals in a minimum-mass solar nebula are invested in planets of a few earth masses each, dimensionless viscosities [alpha] of order dex(-4) to dex(-3) result.  We compare this with observational constraints.  Such small planets would have escaped detection in radial-velocity surveys and could be ubiquitous.  If so, then the similarity of the observed lifetime of T Tauri disks to the theoretical timescale for assembling a rocky planet may be fate rather than coincidence. ",
        "introduction": "The disks of young low-mass protostars (``protostellar disks'') are observed to accrete at rates $\\sim 10^{-8}M_\\odot~\\mbox{yr}^{-1}$ \\citep*{HCGD}. A comparable average accretion rate is implied by the disk masses inferred from dust emission ($\\sim 10^{-2\\pm 1}~M_\\odot$, \\citealp{OB95}) when combined with the maximum ages of T Tauri stars that show evidence for disks ($\\sim 10^{6-7}~\\rm{yr}$, \\citealt{SSECS}), or with radiochemical estimates for the lifetime of the protostellar nebula \\citep{PC94}.  Indeed, the evidence that young protostars are surrounded by viscous accretion disks has only strengthened in the quarter century since \\citet{LP74} proposed it.  While the effective viscosity of the hotter accretion disks surrounding degenerate stars and black holes is probably due to magnetohydrodynamic (MHD) turbulence \\citep[cf.][]{BH98}, protoplanetary disks may be too cold, too weakly ionized, and too resistive to sustain MHD turbulence \\citep{BB94,Jin96,HS98} except perhaps at small radii, in surface layers \\citep{Gam96}, or during FU Orionis-type outbursts \\citep{HKH93}.  In view of several physical uncertainties in the ionization balance \\citep[cf.][]{Gam96}, and the possibly subtle nature of the conductivity itself \\citep{Wardle}, this conclusion is only tentative.  But it is probably fair to say that the best reason for supposing that the viscosity of protoplanetary disks is magnetic remains the lack of a theoretically attractive alternative. Convection \\citep{LP80} and nonlinear shear instability \\citep{Zahn,Dubrulle} have been proposed, but these mechanisms appear to fail because of the very strongly stabilizing influence of a positive radial gradient in specific angular momentum \\citep{RG92,SB96,BHS}.  \\citet{Larson89} suggested that density waves might provide the efffective viscosity mechanism in protostellar disks.  As he made clear, wave transport is rather different from viscous transport. In the standard theory of thin accretion diks, viscosity ($\\nu$) plays two roles. Firstly, it provides an outward angular-momentum flux ($F_J$), a torque exerted by the part of the disk interior to a given radius on the part exterior. Secondly, it heats the disk by dissipating the differential rotation. In steady state, $F_J$ balances the inward flux of angular momentum carried by the accreting matter, while the dissipation balances the loss of its orbital energy.  Even if the disk is not steady, provided only that it is truly viscous, then \\begin{displaymath} F_J = - \\nu\\,\\times\\,2\\pi\\Sigma r^3\\frac{d\\Omega}{dr} \\end{displaymath} where $\\Sigma$ is the surface density and $\\Omega$ the angular velocity. Also, the dissipation rate per unit area is related to $F_J$ by \\begin{equation}\\label{QJrel} \\dot E= -F_J\\frac{d\\Omega}{dr}. \\end{equation} Any dissipative mechanism such that $F_J\\propto d\\Omega/dr$ and eq.~(\\ref{QJrel}) hold can be described by an effective viscosity $\\nu_{\\rm eff}$; one expects this to be the case for dissipative mechanisms that are sufficiently local.  Since density waves are not local, one may deduce a different value of $\\nu_{\\rm eff}$ from $F_J$ than from $\\dot E$.  In particular, inviscid linear theory \\citep[henceforth GT80]{GT80} predicts that a planet embedded in a gas disk excites density waves at the inner and outer Lindblad resonances of its (circular) orbit; these waves carry $F_J>0$ but $\\dot E=0$ by assumption, so that $\\nu_{\\dot E}=0$ but $\\nu_{J}>0$. Actually, the theory implicitly assumes some damping at large distance from the resonances, else the waves would reflect from the edges of the disk and produce zero net flux.  If one is concerned with the torques on the disk only, then the damping length is unimportant as long as it is large. But without dissipation, the waves cause no secular changes to the orbits of the gas even at the Lindblad resonances \\citep{GN89}.  By demanding that eq.~(\\ref{QJrel}) be satisfied, \\citet{Larson90} found that steady spiral shocks produce \\begin{equation}\\label{alphaLarson} \\alpha\\sim 0.013\\left[(c/V)^3+0.08(c/V)^2\\right]^{1/2} \\end{equation} {\\it independently of the source of excitation of the waves.} Here $c$ is the sound speed and $V=r\\Omega$ the orbital velocity. For $c/V=0.03$, perhaps characteristic of the later phases of the protosolar nebula at 1~AU, this gives $\\alpha\\approx 10^{-4}$. Larson's result depends upon a number of additional assumptions, some inspired by earlier work by \\citet{Sp87} on self-similar shocks: the waves were taken to form two spiral arms separated by $180^\\circ$ in azimuth; and because they would be very tightly wrapped since $c\\ll V$, \\citet{Larson90} approximated the wave profile by analyzing a train of perfectly axisymmetric nonlinear waves. Most importantly, at least by contrast with our own work reported below, no source of excitation for the waves exists within the range of radii where eq.~(\\ref{alphaLarson}) applies.  Larson argued, however, that a bisymmetric spiral shock of the required strength could be launched from the $2:1$ resonance of a jovian-mass planet, and he deduced an accretion timescale $\\sim 10^{6-7}~\\mbox{yr}$ in accord with observational requirements.  Jupiter-mass planets have been discovered during the past few years by exquisitely sensitive radial-velocity surveys \\citep{MCM}.  Most of these planets have orbital periods shorter than that of Mercury.  They are detected around $4-5\\%$ of the stars surveyed, mostly G, F, and K dwarfs.  There are indications that stars of higher-than-solar metallicity are preferred \\citep{Gon97,SIM,GLTR}.  Because of observational selection effects, it is probably too early to derive the incidence of giant planets in long-period orbits ($\\sim 10~{\\rm yr}$).  If such planets are as rare as those on short-period orbits, however, then we must seek another explanation for the observed accretion rates and inferred lifetimes of most protostellar disks. Nevertheless, it remains an intriguing notion that planet formation is not just constrained by disk lifetimes, but may actually determine them.  Our paper therefore inquires whether accretion could be driven by planets of roughly terrestrial mass, which radial-velocity surveys would not have detected.  Other things being equal, the torque exerted on the disk by a planet is proportional to the square of the planet's mass ($M_p$), so the influence of $M_p\\sim 10^{-5}M_\\odot$ might be thought negligible compared with that of $M_p\\sim 10^{-3} M_\\odot$. Just because its local effects are so strong, however, the larger planet will almost certainly open a gap in the disk, whereas the smaller one will not; since the torque per unit radius (smoothed over resonances) decreases rapidly with distance from the planet (GT80), the shorter range of its interaction may partly compensate for the terrestrial planet's smaller mass.  At solar abundance there could be tens of earth-mass planets in the disk, each too small to have captured a significantly massive atmosphere.  We assume that the planets are distributed through the disk in proportion to the local surface density of the gas, neglecting orbital migration and radial changes in the timescale of planet formation.  Unlike Larson, therefore, we have a distributed source of excitation for density waves, so that eq.~(\\ref{alphaLarson}) need not apply (although in fact we obtain a comparable value of $\\alpha$ in the minimum-mass solar nebula).  As discussed, no sensible effective viscosity can be derived for density waves without considering their damping. \\S2 shows for a generic damping mechanism in a thin disk that if the waves are absorbed over a distance small compared to the disk radius, but possibly large compared to its thickness, and if the local damping rate depends mainly on distance from the planet, then eq.~(\\ref{QJrel}) is satisfied so that a consistent value of $\\nu_{\\rm eff}$ or $\\alpha$ can in fact be defined.  However, the viscosity is directly proportional to the damping length. After briefly considering other mechanisms in \\S3, we conclude that shock damping is probably most important even at $M_p\\sim M_\\oplus$. In \\S4, we calculate the excitation of the wake from linear theory, which is valid for sufficiently small $M_p$ [eq.~(\\ref{Mlimit}]. In \\S5 and the Appendix, we show that the subsequent nonlinear propagation of the wake away from the planet can be reduced to Burger's equation using some transformations and some approximations that should be accurate for sufficiently low-mass planets in sufficiently thin disks. We describe quantitatively the formation of the shock and its subsequent decay with increasing distance from the planet.  Most importantly, we obtain formulae for the shock-damping length, and hence for $\\alpha_{\\rm eff}$, as functions of $M_p$ and local disk parameters. \\S6 summarizes our main results and discusses their implications for disks resembling a standard minimum-mass solar nebula. ",
        "conclusions": "\\label{disc}   We have shown that density-wave wakes excited by planets in a nonselfgravitating protostellar disk serve as an effective viscosity mechanism, provided that the waves are damped over a distance smaller than the orbital radius of the planet.  The effective viscosity is proportional to the damping length, which is most likely to be limited by shocks unless the mass per planet $M_p\\ll M_\\oplus$.  If a fixed total mass is available to form planets, then $\\alpha_{\\rm eff}\\propto M_p^{3/5}$. Eq.~(\\ref{alpha_eff}) suggests that $\\alpha_{\\rm eff}\\sim 10^{-4}$ in the inner solar system where $M_p\\sim M_\\oplus$, and $\\alpha_{\\rm eff}\\sim 3\\times 10^{-4}$ at $r\\sim 10~AU$ if $M_p\\sim 10~M_\\oplus$, which is characteristic of the rock+ice \\emph{cores} of the giant planets, though very much less than the total mass of Jupiter or Saturn \\citep{Gu99}.  Our estimates for $\\alpha_{\\rm eff}$ may seem optimistic because they assume that all of the rocks and ices are efficiently converted into planets. On the other hand, $\\alpha_{\\rm eff}$ could be even larger if $M_p$ is augmented by gas.  In that case, $\\alpha_{\\rm eff}\\propto M_p^{8/3}$ because the mass in planets is not limited to the metals. There are two restrictions on the amount of gas we can allow for. The first is technical: we have calculated the excitation of the density waves from linear theory, which cannot be justified when $M_p$ exceeds the limit (\\ref{Mlimit}). This could be overcome by nonlinear two-dimensional numerical calculations \\citep[\\emph{e.g.~}][]{KP96}. The other restriction is more substantial: once $M_p$ is large enough to open a gap in the disk, the dissipation rate probably scales more slowly than $M_p^{8/3}$.  Several estimates exist for the minimum mass required to open a gap, $M_{\\rm gap}$.  \\citet{LP93} and \\citet{BCLNP} argue that in a completely inviscid disk, $M_{\\rm gap}\\approx M_1$ as given by eq.~(\\ref{Mlimit}) because of radial gas pressure gradients; recall that at this mass, the Roche lobe (or Hill sphere) of the planet is $\\approx h$. \\citet{Wa97} argues for $M_{\\rm gap}\\approx 0.14\\pi\\Sigma h r$, or $M_{\\rm gap}\\approx 0.8 (r/AU)^{3/4}\\,M_\\oplus$ in the fiducial MMSN (\\ref{MMSNSig})-(\\ref{MMSNc}).  This results from comparing the time to open a gap with the time to migrate across it, assuming that planetary wake damps immediately at the resonances where it is excited.  Presumably $M_{\\rm gap}$ should be increased to allow for the finite damping length that we have computed, but it would still be proportional to $\\Sigma$, unlike eq.~(\\ref{Mlimit}).  $M_{\\rm gap}$ is expected to be larger in a viscous disk than in an inviscid one because the torque exerted by the planet can be balanced by a viscous torque. \\citet{BCLNP} cite \\begin{displaymath} \\frac{M_{\\rm gap}}{M_*}=\\frac{40\\nu}{\\Omega r^2} \\approx 40\\alpha\\left(\\frac{h}{r}\\right)^2, \\end{displaymath} which is again independent of $\\Sigma$ (if $\\alpha$ is). In the MMSN, this becomes \\begin{displaymath} M_{\\rm gap}\\approx 2.\\left(\\frac{\\alpha}{10^{-4}}\\right) \\left(\\frac{r}{AU}\\right)^{1/2}~M_\\oplus. \\end{displaymath} In our case, it seems likely that the shocks created by one planet will resist the opening of a gap next to an adjacent planet, so it is perhaps reasonable to use the value (\\ref{alpha_eff}) for $\\alpha$ on the righthand side above.  In the discussion to follow, we shall follow \\citet{LP93} and adopt (\\ref{Mlimit}) as a provisional value for $M_{\\rm gap}$, but clearly the question deserves further study. Then $M_{\\rm gap}\\sim 45~M_\\oplus$ at $10~{\\rm AU}$ using eq.~(\\ref{MMSNc}).  If the planets at $10~{\\rm AU}$ accrete enough gas to reach this limit (Saturn exceeds it by a factor $\\sim 2$), then $\\alpha_{\\rm eff}\\approx 3\\times 10^{-3}$.  In short, values for $\\alpha_{\\rm eff}$ in the range from $10^{-4}$ up to at least $10^{-3}$ seem plausible for the density-wave mechanism in a minimum-mass solar nebula.  The predicted accretion timescale is \\begin{displaymath} t_{\\rm acc}= \\alpha^{-1}\\left(\\frac{r}{h}\\right)^2\\Omega^{-1} \\approx 5\\times 10^6 \\left(\\frac{\\alpha}{10^{-3}}\\right)^{-1} \\left(\\frac{r}{30~{\\rm AU}}\\right) ~\\mbox{yr}, \\end{displaymath} in reasonable agreement with the disk lifetimes cited in \\S 1.  By comparing observed sizes, ages, and accretion rates of T Tauri disks with self-similar models, \\cite{HCGD} estimate $\\alpha\\sim 10^{-2}$, which may be too large to be explained by planetary wakes, especially if one must rely on the rocky core masses only.  The strongest lower bounds on $\\alpha$ assume that the disks started out much more compact than observed.  In view of the wide scatter and uncertainties in the observationally derived parameters, the questionable relevance of self-similar models, and the theoretical uncertainties just discussed, it is unclear whether \\citeauthor{HCGD}'s result is a major difficulty for us. Another concern is that planets are likely to migrate inwards with respect to the gas \\citep{Wa97}, perhaps enhancing the viscosity of the inner disk but diminishing that of the outer parts where most of the gas probably resides.  Still another is that our mechanism operates only after planet growth, which standard models place late in the lifetime of the gaseous nebula, if not afterwards \\citep{PHP93,INE00}, whereas \\citeauthor{HCGD} find \\emph{larger} accretion rates in younger disks on average.  All of these objections have merit.  Nevertheless, the density-wave mechanism is interesting from several points of view.  First, it provides a minimum effective viscosity when other mechanisms are in abeyance.  For example, MHD turbulence may dominate only during FU Orionis-like outbursts when the temperature and ionization of the disk rises \\citep{Gam98}; self-gravity may dominate in young, massive, cool disks \\citep{LP87}; but neither is likely to be effective in the late, thermally passive phases when it is believed that planets form. Second, calculable models for the effective viscosity are so difficult to come by that it is worthwhile to examine any plausible candidate, even if others seem more likely to dominate (but cannot be calculated).  Third, the formalism offered in \\S 5 and the Appendix for the development and damping of the shock, though only approximate, is expected to be reasonably accurate for weak shocks and simple enough to use that it may find other applications.\\footnote{Burger's equation can be solved analytically given a closed form for the initial wave profile \\citep[cf.][]{Whitham}.} Finally, there may be other nonaxisymmetric structures in protostellar disks of greater mass than planets, for example vortices \\citep{AW95,GL00,Na00}.  Any such structure would create a density-wave wake."
    },
    "0010/astro-ph0010606_arXiv.txt": {
        "abstract": "{ We present the first results on the hard X-ray continuum image (up to 15~keV) of the supernova remnant Cas~A measured with the EPIC cameras onboard XMM-Newton. The data indicate that the hard X-ray tail, observed previously, that extends to energies above 100~keV does not originate in localised regions, like the bright X-ray knots and filaments or the primary blast wave, but is spread over the whole remnant with a rather flat hardness ratio of the 8$-$10 and 10$-$15~keV energy bands.  This result does not support an interpretation of the hard X-radiation as synchrotron emission produced in the primary shock, in which case a limb brightened shell of hard X-ray emission close to the primary shock front is expected.  In fact a weak rim of emission near the primary shock front is discernable in the hardest X-ray image but it contains only a few percent of the hard X-ray emissivity.  The equivalent width of the Fe-K line blend varies by more than an order of magnitude over the remnant, it is hard to explain this as Fe-emission from the reverse shock heated ejecta given the ejecta temperature and the age of the remnant.  The uniquely high wavelength-dispersive RGS-spectrometer has allowed, for the first time, to extract monochromatic images in several highly ionised element species with high spectral resolution. We present here a preliminary result on the measurement of the \\ion{O}{viii} Ly-$\\alpha$ and Ly-$\\beta$ brightness distribution and brightness ratios.  The large observed decrease of the Ly-$\\alpha$/Ly-$\\beta$ ratio going from the N to the SE can be explained by small-scale (10~\\arcsec) variations in the $N_{\\rm H}$ column over the remnant and the potential presence of resonance scattering of the \\ion{O}{viii} Ly-$\\alpha$ photons in the limb brightened shell. ",
        "introduction": "The young galactic supernova remnant (SNR) Cassiopeia A (Cas A) is widely believed to be the result of the core collapse of a massive star, probably an early type Wolf-Rayet star (Fesen et al.  \\cite{fesen}).  Cas A is classified as an oxygen rich remnant since optical spectroscopic observations (Chevalier \\& Kirshner \\cite{chevalier}) show the supernova ejecta (in the form of fast moving knots) to contain mostly oxygen and oxygen burning products such as sulphur, argon and calcium.  At all wavelengths Cas A has the appearance of a broken shell with a radius varying between 1.6\\arcmin\\ to 2.5\\arcmin.  One of the outstanding problems in the study of Cas A is the origin of a recently discovered hard X-ray continuum in the spectrum of this remnant.  From a theoretical point of view both synchrotron emission from shock accelerated electrons and non-thermal Bremsstrahlung from electrons accelerated from the tail of the thermal emission could be possible explanations for the hard continuum.  In addition, high resolution spatially resolved spectroscopy of the remnant in the X-ray domain is a powerful tool to study in detail the distribution and physical properties of the (reverse) shock heated plasma.  The bandwidth (up to 15 keV) of XMM-Newton and the presence of high wavelength-dispersive spectrometers (RGS) offers an unique opportunity to address these questions.  In this respect XMM-Newton strongly complements Chandra, and this paper gives a first highlight of this capability. ",
        "conclusions": ""
    },
    "0010/astro-ph0010430_arXiv.txt": {
        "abstract": "We use the observed optical-UV and X-ray emission spectrum of Cen X-4 during quiescence to constrain models for the accretion flow in this system. We argue that the optical-UV emission is not due to an optically-thick quiescent accretion disk, nor due to synchrotron emission from an Advection-Dominated Accretion Flow (ADAF). Emission from the bright spot could account for the observed optical-UV component if the mass transfer rate in Cen X-4 is $\\gsim 2 \\times 10^{16}$~g~s$^{-1}$. Although the presence of an ADAF around the neutron star leads to Compton upscattering of the soft X-ray photons radiated from the stellar surface, we find that this process alone cannot account for the power law component seen in the quiescent X-ray spectrum of Cen X-4 and other X-ray transients containing neutron stars; this result is independent of whether the source of soft photons is incandescent thermal emission or accretion-powered emission.  We conclude that, in models which invoke the presence of an ADAF and a propeller effect for the quiescence of X-ray transients containing neutron stars, the intrinsic emission from the ADAF must contribute very little to the optical-UV and X-ray emission observed. If these ADAF+propeller models are correct, the X-ray power law component observed must arise from regions where the gas impacts the neutron star surface. Variability studies could greatly help clarify the role of the various emission mechanisms involved. ",
        "introduction": "Soft X-ray Transients (SXTs) are close binary systems in which a low-mass secondary (either a main-sequence star or a subgiant) transfers mass via Roche-lobe overflow onto a black hole (BH) or neutron star (NS) primary (see reviews by Tanaka \\& Lewin 1995; van Paradijs \\& McClintock 1995; White, Nagase \\& Parmar 1995).  SXTs spend most of their lifetimes in a low luminosity quiescent state, but occasionally undergo dramatic outbursts during which both the optical and X-ray emission increase by several orders of magnitude (e.g. Chen, Shrader \\& Livio 1997; Kuulkers 1998). SXTs provide us with some of the best stellar-mass black hole candidates known to date, thanks to lower limits set on the mass of the accretor via the observation during quiescence of Doppler-shifted lines in the spectrum of the secondary (see, e.g., McClintock 1998).  While observations (see, e.g., Tanaka \\& Shibazaki 1996) indicate that SXTs accrete matter via a standard thin disk (Shakura \\& Sunyaev 1973) during outburst, the situation appears more complex in quiescence (see, e.g., Lasota 1996).  Narayan, Barret \\& McClintock (1997; see also Narayan, McClintock \\& Yi 1996) proposed that accretion in quiescent BH SXTs proceeds via a two-component accretion flow, consisting of an inner hot advection-dominated accretion flow (ADAF; see Narayan, Mahadevan \\& Quataert 1998 for a review of ADAFs) surrounded by an outer thin accretion disk.    According to these models, the unusually low luminosities of BH SXTs in quiescence is due to the defining property of ADAFs: the bulk of the viscously dissipated energy is stored in the gas and advected with the flow into the black hole (Ichimaru 1977; Rees et al. 1982; Narayan \\& Yi 1994, 1995; Abramowicz et al. 1995; Narayan et al. 1996, 1997a). By contrast, in NS SXTs all the advected energy is expected to be radiated from the neutron star surface, resulting in a much higher radiative efficiency of the accretion flow even in the presence of an ADAF (Narayan \\& Yi 1995). This distinction between black hole and neutron star systems motivated a comparison of the outburst amplitudes of BH SXTs and NS SXTs as a function of their maximum luminosities by Narayan, Garcia \\& McClintock (1997) and Garcia et al.  (1998). These authors showed that the observations reveal systematically lower relative luminosities in BH SXTs and argued that this constitutes a confirmation of the presence of an event horizon in BH SXTs.   Menou et al. (1999) developed ADAF+thin disk accretion models for quiescent BH and NS SXTs that account for the luminosity differences between the two classes of systems (see also Lasota 2000). However, the relatively low quiescent luminosities of NS SXTs can be reconciled with the model predictions only if an efficient ``propeller effect'' (Illarionov \\& Sunyaev 1975) operates during quiescence, so that the magnetosphere of the rapidly rotating neutron star prevents much of the accreting material from reaching the neutron star surface (see also Asai et al. 1998; Zhang, Yu \\& Zhang 1998).   An alternative explanation for the quiescent X-ray emission of NS SXTs has been proposed by Brown, Bildsten and Rutledge (1998). These authors argue that nuclear reactions in the crust of neutron stars, triggered during the outbursts of NS SXTs, efficiently heat up the NS cores in these systems. The energy deposited during an outburst is reemitted during quiescence at a rate sufficient to power the observed quiescent emission (see also Rutledge et al. 1999).  Recently, Campana \\& Stella (2000) also proposed that the relativistic wind of the NS (in a pulsar regime) in quiescent SXTs could be responsible for part of the X-ray emission observed; the X-rays are produced by the impact of the wind on the gas stream flowing from the companion star.    So far, the quiescent emission of NS SXTs has been mainly discussed in terms of their luminosity, whereas it is likely that spectral information will help discriminate between the proposed emission mechanisms discussed above (see, e.g., Rutledge et al. 2000). In this paper, we construct models for the broadband emission spectrum of quiescent NS SXTs, with a particular emphasis on Cen X-4. {Our main goal is to test the ADAF+propeller scenario proposed by Menou et al. (1999), but some of our results also have implications for the incandescence scenario put forward by Brown et al. (1998).} An important motivation for this work is the recent finding by McClintock \\& Remillard (2000) that Cen X-4 (a NS SXT) and A0620-00 (a BH SXT), despite similar orbital periods, show significantly different quiescent optical-UV emission. This result could be an indication of the different nature of the accreting compact object in the two systems.   In \\S2, we recall important observational properties of the NS SXT Cen X-4, which are later used to constrain our models. In \\S3, we discuss several candidates for the quiescent optical-UV emission of Cen X-4. We then turn to models of the X-ray spectrum in \\S4.  Finally, we discuss consequences, limitations and possible extensions of this work in \\S5. ",
        "conclusions": "We constructed models of the accretion flow in the NS SXT Cen X-4 during quiescence. Many of our conclusions can in principle be generalized to other NS SXTs. We do not find a satisfying explanation for the observed optical-UV and X-ray emission spectrum of Cen X-4 within the framework of ADAF+propeller accretion models.  We argued that the bright spot can power the quiescent optical-UV emission of Cen X-4 if the mass transfer rate of the system is $\\gsim 2 \\times 10^{16}$~g~s$^{-1}$. This is larger than the estimates based on outburst fluences and therefore suggests, if correct, that a fair amount of ``invisible'' accretion occurs in quiescence, which could be due to an efficient propeller acting during this phase. Note that we did not consider the possibility of reprocessing of X-rays by the accretion disk for powering the optical-UV flux because the X-ray luminosity, $L_X \\simeq 2-3 \\times 10^{32}$~ergs~s$^{-1}$, appears insufficient to explain the strong optical-UV component, with $L_{\\rm opt-UV} \\simeq 0.8 \\times 10^{32}$~ergs~s$^{-1}$.    When ruling out emission from a quiescent disk, we neglected possible deviations from blackbody emission in the innermost regions of the disk. Most of the observed emission actually originates from the outer regions ($R^2 T_{\\rm eff}^4 \\propto \\dot M^-_{\\rm crit}/R \\propto R^{1.67}$; see Eqs.~[1] and~[2]) which are optically thick according to the Disk Instability Model (DIM). For this reason, we do not expect possible complications for the emission in the inner regions of the disk to be sufficient to explain the spectral discrepancy shown in Fig.~\\ref{fig:dimdisk}. Conceivably, the presence of a disk chromosphere could induce spectral deviations from blackbody-type emission but more detailed modeling is required to address this question. Note that the broad Mg~II line in the spectrum of Cen X-4, with an integrated energy $L_{\\rm MgII} \\simeq 2.8 \\times 10^{30}$~erg~s$^{-1}$ (McClintock \\& Remillard 2000), does suggest emission from an optically-thin disk chromosphere. The single-peaked shape can be explained by the superposition of two double-peaked doublet components separated by $7.2 \\AA$, thereby supporting a disk origin for this line.  We argued that Compton upscattering of soft X-ray photons emitted at the surface of a NS by an ADAF is not sufficient to explain the power law component inferred in the quiescent X-ray spectrum of Cen X-4 and other NS SXTs. Instead, most of the flux in the power law component could be associated with the accretion-powered emission in higher density regions close to the NS surface (see, e.g., Shapiro \\& Salpeter 1975; Turolla et al. 1994; Popham \\& Sunyaev 2000; Medvedev \\& Narayan 2001; we note, however, than none of these models can be directly applied to the ADAF+propeller scenario because of discrepancies in the accretion rate or the geometry assumed).   Compton upscattering by a surrounding ADAF could also have applied to the scenario of Brown et al. (1998) if, for instance, the propeller effect is efficient enough in quiescence to prevent all the mass accreted via the ADAF to reach the NS surface, so that the accretion-powered luminosity becomes negligible compared to the NS incandescent emission. However, the difficulties with this scenario are similar to the case of accretion-powered emission: Compton upscattering in the low-density ADAF is insufficient to produce the observed X-ray power law emission. Therefore, the origin of the power law component in the incandescence scenario remains unclear.    Variability studies could therefore play an important role in clarifying which mechanisms operate in quiescent NS SXTs. For instance, variability in the BB component would rule out the scenario of Brown et al. (1998). Similarly, if coherent X-ray pulsations were to be observed in a quiescent NS SXT, they would occur in both the BB and the power law component according to the scenario presented in \\S5, while they should not occur for the BB component according to the scenario of Brown et al. (1998). Campana \\& Stella (2000) have discussed additional variability characteristics expected in their pulsar scenario.    It is somewhat embarrassing that no obvious signature of accretion via an ADAF exists in the model proposed for the quiescent emission spectrum of Cen X-4. The ADAF remains unseen in optical-UV, perhaps because of the truncation by the magnetosphere; similarly, its X-ray emission must be dominated by accretion-powered emission at the surface of the NS, which has a much higher radiative efficiency than the ADAF. Consequently, studying the effect of possible winds from the ADAF (which would only reduce the intrinsic ADAF emission; Blandford \\& Begelman 1998; Quataert \\& Narayan 1999) is essentially irrelevant in the present context. This also means that it is impossible, if there is indeed an ADAF and a propeller acting in Cen X-4 during quiescence, to constrain the efficiency of the propeller relative to that of the winds at removing mass from the accretion flow.    Finally, we note that our study does not offer a clear explanation for the qualitative spectral difference between A0620-00 and Cen X-4 in the optical-UV. Although the presence of a magnetosphere in the case of Cen X-4 is expected to reduce the synchrotron ADAF emission relative to A0620, and therefore lead to spectral differences, this does not explain the significantly hotter UV emission of Cen X-4."
    },
    "0010/astro-ph0010110_arXiv.txt": {
        "abstract": "s{ We consider a sample of about 50 distant galaxy clusters at $z>0.15$ ($<z>\\sim0.3$), each cluster having at least 10 galaxies with available redshift in the literature.  We select member galaxies, analyze the velocity dispersion profiles, and evaluate in a homogeneous way cluster velocity dispersions and virial masses.  We apply the same procedures already recently applied on a sample of nearby clusters ($z<0.15$) in order to properly analyze the possible dynamical evolution of galaxy clusters.  We do not find any significant difference between nearby and distant clusters.  In particular, we consider the galaxy spatial distribution, the shape of the velocity dispersion profile, and the relations between velocity dispersion and X--ray luminosity and temperature.  Our results imply little dynamical evolution in the range of redshift spanned by our cluster sample, and suggest that the typical redshift of cluster formation is higher than that of the sample we analyze.} ",
        "introduction": "The knowledge of the properties of galaxy clusters plays an important role in the study of large scale structure formation. In particular, one can use their number density as a function of their mass (\"mass function\") to constrain cosmological parameters. Several recent works attempt to estimate the value of the matter density parameter $\\Omega_m$ from the evolution of cluster mass function (e.g., Carlberg {\\it et al.}~\\cite{carl97}; Borgani {\\it et al.}~\\cite{borg99}).   However, the estimate of cluster masses is not an easy task.  Both the virial theorem applied to positions and velocities of cluster member galaxies and the dynamical analysis of hot X-ray emitting gas assume that clusters are systems in dynamical equilibrium, while the mass estimates derived from gravitational lensing phenomena require a good knowledge of cluster geometry.  As for the analysis of the internal dynamics of nearby clusters (at redshift $z\\lesssim 0.15$), recent conclusions are based on very large samples (Fadda {\\it et al.}~\\cite{fad96}; Mazure {\\it et al.}~\\cite{maz96}; Girardi {\\it et al.}~\\cite{gir98b}). In particular, Girardi {\\it et al.}~\\cite{gir98b} showed that virial masses based on member galaxies are quite consistent with those obtained from X-ray analysis suggesting that most clusters are not very far from dynamical equilibrium.  As for distant clusters, most results come from the analysis of the 16 clusters at intermediate redshifts, $0.18<z<0.55$ of CNOC (Canadian Network for Observational Cosmology; Yee {\\it et al.}~\\cite{yee96}) which represents a remarkably homogeneous sample.  In particular, as found in nearby clusters, Lewis {\\it et al.}~\\cite{lew99} claim for consistency between masses coming from optical and X--ray data. However, the difficulties of obtaining many redshifts in distant clusters have prevented from building larger samples. Rather, several works, concerning one or a small number of clusters, and using different techniques of analysis, can be found in the literature.  The availability of a variety of techniques, already applied to nearby clusters, suggests their application to distant clusters.  We thus ensure the homogeneity of our results over a large range of cosmological distances.  A homogeneous analysis is in fact fundamental for the understanding of the evolution of cluster properties. We present the preliminary results of our analysis of a sample of about 50 clusters with $z>0.15$ where we use the same techniques already used by Girardi {\\it et al.}~\\cite{gir98b} (cf. also Fadda {\\it et al.}~\\cite{fad96}) on a sample of 170 nearby clusters (at $z<0.15$, data from ENACS -- ESO Nearby Abell Cluster Survey, Katgert {\\it et al.}~\\cite{kat98} -- and other literature).  Unless otherwise stated, we give errors at the 68\\% confidence level (hereafter c.l.).  A Hubble constant of 100 $h$ \\ks $Mpc^{-1}$ and a deceleration parameter of $q_0=0.5$ are used throughout. ",
        "conclusions": "In order to properly analyze the possible dynamical evolution of galaxy clusters we apply the same procedures already applied on a sample of nearby clusters (170 clusters at $z<0.15$ from ENACS and other literature, Girardi {\\it et al.}~\\cite{gir98b}, cf. also Fadda {\\it et al.}~\\cite{fad96}) to a corresponding sample of distant clusters.  We consider a sample of 51 distant galaxy clusters at $z>0.15$ ($<z>\\sim0.3$), each cluster having at least 10 galaxies with available redshift in the literature.  A part from three cluster field showing two overlapping peaks in their velocity distribution and so large uncertainties in their dynamics, 45 fields show only one peak in their velocity distribution and three fields show two separable peaks for a total of 51 well defined cluster systems. These 51 systems are those used in the comparison with nearby clusters (160 well defined systems).  We select member galaxies, analyze the velocity dispersion profiles, and evaluate in a homogeneous way cluster velocity dispersions and virial masses.  As a main general result, we do not find any significant evidence for dynamical evolution of galaxy clusters. More in detail, our results can be summarized as follows.  \\begin{itemize}  \\item The galaxy spatial distribution is similar to that of nearby clusters, i.e. the fit to a King--like profile gives a two-dimensional slope of $\\alpha=0.7$ and a very small core radius of $R_c=0.05$ \\hh.  \\item When data are good enough, the integrated velocity dispersion profiles of distant clusters show a behavior similar to those of nearby clusters, i.e.  they show strongly increasing or decreasing behaviors in the central cluster regions, but are flattening out in the external regions suggesting that in such regions, they are no longer affected by velocity anisotropies.  \\item The average velocity dispersion profile can be explained by a model with isotropic orbits, which well describe also nearby clusters. Possible evidences for more radial orbits are not statistically significant.  \\item There is no evidence of evolution in both $L_{bol,X}$--$\\sigma_v$ and $\\sigma_v$--$T_X$ relations, thus in agreement with previous results (Mushotzky \\& Scharf~\\cite{mus97}; Borgani {\\it et al.}~\\cite{borg99}).  \\end{itemize}  Moreover, on average, within the large scatter of present data, we find no significant evidence of discrepancies between our virial mass estimates and those from X--ray and gravitational lensing data, thus suggesting that distant clusters are not far from global dynamical equilibrium (cf. also Allen~\\cite{all98}; Lewis {\\it et al.}~\\cite{lew99}).  We conclude that the typical redshift of cluster formation is higher than that of our sample in agreement with previous suggestions (e.g., Schindler~\\cite{sch99}; Mushotzky~\\cite{mus00}).  In particular, we agree with preliminary results by Adami {\\it et al.}~\\cite{ada99}, who applied the same techniques used for the nearby ENACS clusters on 15 distant clusters, ($<z>\\sim0.4$) from the Palomar Distant Cluster Survey (Postman {\\it et al.}~\\cite{pos96}).  Although some clusters at very high redshift, e.g. $z>0.8$, are already known (e.g., Gioia {\\it et al.}~\\cite{gio99}; Rosati {\\it et al.}~\\cite{ros99}), the construction of a large cluster sample useful for studying internal dynamics will require a strong observational effort. We stress how both the poor number of galaxies and the small spatial extension of some clusters can affect the robustness of their resulting properties."
    },
    "0010/astro-ph0010056_arXiv.txt": {
        "abstract": "We use high resolution numerical simulations of the formation of cold dark matter halos to simulate the background of decay products from neutralino annihilation, such as gamma-rays or neutrinos.  Halos are non-spherical, have steep singular density profiles and contain many thousands of surviving dark matter substructure clumps. This leads to several unique signatures in the gamma-ray background that may be confirmed or rejected by the next generation of gamma-ray experiments. Most importantly, the diffuse background is enhanced by over two orders of magnitude due to annihilation within substructure halos. The largest dark substructures are easily visibly above the background and may account for the unidentified EGRET sources. A deep strip survey of the gamma-ray background would allow the shape of the Galactic halo to be quantified. ",
        "introduction": "\\label{s:Intro}  Determining the nature of dark matter is of fundamental importance to both Astronomy and Particle Physics. Both theory and observational data currently favour a universe with a matter density that is dominated by non-baryonic particles. Many candidates have been proposed: some are known to exist, others are more speculative (e.g., Ref. ~\\cite{ellis98} and references therein).  Structure formation in a universe dominated by cold dark matter (CDM) has been extensively tested against observations and the model has proven highly successful at reproducing the large scale properties and distributions of galaxies ~\\cite{davis85,baugh}. On the non-linear scales of galactic halos it remains to be confirmed wether the model can successfully reproduce the observational data ~\\cite{moore94,moore99a,Klypin99}.  Direct detection in the laboratory is the ultimate technique for verifying the existence of dark matter particles (see, Ref. ~\\cite{collar99}). However, even the most popular candidate for dark matter, the neutralino, has a cross section that spans many orders of magnitude and the current laboratory searches are only just becoming sensitive to the cosmologically interesting parameter range. Presently, Astronomical observations provide the best insights into the nature of the dark matter, furthermore direct detection relies on the existence of a smooth component of dark matter.  Within the next few years indirect detection of neutralinos will provide interesting constraints on their possible cross-section and masses.  Neutralino-neutralino annihilation produces observable photons (as well as a host of other particles) that may be observed as a diffuse gamma-ray background from the halo surrounding the Milky Way as discussed in Refs. ~\\cite{gunn78,stecker78,silkS84,turner86,silkB87,bouquet89,lake90}, and more recently, in Refs. ~\\cite{berg2,berg98a,baltz99,gondolo99a,baltz00}.  Renewed interest in these predictions has recently arisen because of an unexplained component of diffuse high energy photons in the Egret data (e.g. Ref.~\\cite{dixon98}), and also the possibility of an excess from the center of the Galaxy itself ~\\cite{Mayer98}, unexpected clumpy emission and the unresolved ``discrete sources'' ~\\cite{hartman}. Progress in this area will result from several new and sensitive gamma-ray surveys such as GLAST ~\\cite{glast} and VERITAS ~\\cite{veritas}.  The efficiency of the annihilation process is strongly dependent on the local density and the cross-section of the neutralino.  Many authors have calculated the expected flux from the Galactic halo using simple models for the expected mass distribution of neutralinos within the Galaxy (see, e.g., Refs.~\\cite{berg98a,gondolo99a}) or from its satellites ~\\cite{lake90,baltz00}.  Advances in computational Cosmology have lead to several recent breakthroughs that have direct relevance to the detection of dark matter. In particular, the numerical resolution that can be achieved using parallel computational techniques is now sufficient to study the internal structure of dark matter halos  that form within a cosmological context. The results of these simulations have important implications for indirect (and direct) detection of dark matter candidates. Most significantly for particle-particle annihilation, we are now confident that the central density profile of CDM halos follows a singular power law down to small scales ~\\cite{carlberg94,navarro96,moore98a,ghigna98}. Thus we may expect a point like source of mono-chromatic gamma-rays emanating from the center of the Milky Way, where the annihilation rate will be very high.  A second fundamental prediction of the CDM model is that previous generations of the merging hierarchy survive within halos ~\\cite{ghigna98}.  Halos that accrete into larger systems may be tidally stripped of most of their mass, however their dense central cores survive and continue to orbit within the parent halos.  This may present some problems for the CDM model since the predicted number of satellites within the Milky Way's halo is 50--100 times as many as observed ~\\cite{moore99a}. If the CDM model is correct, then only a fraction of these satellites must have formed stars and most of the substructure remains as dark objects within the Galactic halo.  The possibility of an enhanced gamma-ray background from dark matter substructure was explored by Bergstr\\\"om {\\it et al.}~\\cite{berg99}, who made simple assumptions as to the mean density and abundance of such clumps.  We can now use the high resolution N-body simulations to directly measure these quantities.  The simulations also allow us to study the influence of the halo shapes on the diffuse gamma-ray background and the intensity of the central halo emission that arises from the singular dark matter density profiles.  This paper is organized as follows. In Section~\\ref{s:estimating} we explore the gamma-ray background that results from the smoothly distributed component of dark matter using both analytical and simulated halos. In Section~\\ref{s:subs} we focus on the substructure within halos. Our conclusions are summarised in Section~\\ref{s:concs}. ",
        "conclusions": "\\label{s:concs}  Numerical simulations that follow the growth of structure within a universe dominated by neutralinos (cold dark matter) have achieved a resolution that allows their global structure and internal structure to be quantified.  The density profiles, shapes of dark matter halos, abundance and properties of dark matter substructure, all play an important role in determining the absolute surface brightness of observable products from dark matter annihilation.  We have used the results from the highest resolution simulations ever performed of CDM halos to examine the expected all-sky distribution of gamma-rays from neutralino annihilation. Substructure can boost the expected flux significantly over that originating from a smooth dark matter halo.  Thus, gamma-ray observations, such as EGRET data, may already have the potential of constraining a large part of the parameter range of the neutralino cross-sections.  The distinguishing shapes of CDM halos and the unique spatial and flux distribution of point sources from substructure within the Galactic halo should allow a unique identification of observational data with dark matter."
    },
    "0010/astro-ph0010260_arXiv.txt": {
        "abstract": "We report on a large X-ray flare from the Lindroos binary system HD\\,560 during a pointed {\\em ROSAT} PSPC observation. HD\\,560 is composed of a B9 and a G5 type star. The late-type companions in bound Lindroos binaries are Post T Tauri star (PTTS) candidates, i.e. pre-main sequence stars on their final approach to the main sequence. Strong (magnetic) activity is, therefore, expected for the G star, in contrast to the late-B type component for which no mechanism of X-ray production is known. The system is unresolved in the {\\em ROSAT} image.  During outburst the count rate has changed by at least a factor of 2, and presumably more (the maximum was not observed). The peak luminosity measured is $\\sim 10^{31}\\,{\\rm erg/s}$, comparable to the largest values derived for flares on pre-main sequence T Tauri stars. A two-temperature model for thermal emission from a hot, optically thin plasma shows an increase of the higher temperature as a result of coronal heating during the flare. The lower temperatures do not change during outburst. ",
        "introduction": "T Tauri stars (TTS) are young, late-type stars contracting to the main sequence (MS). In their evolution pre-main sequence stars pass through a phase before reaching the MS, defined as Post T Tauri stars (PTTS). Herbig~(1978) first defined PTTS as young stars more evolved than Classical T Tauri stars (CTTS) but still contracting to the MS.  PTTS are difficult to detect because they do not show extreme properties which make them easy to identify: unlike CTTS, they do not show IR or UV excesses and the H$\\alpha$ line is not longer seen in emission. Therefore, their identification relies on the presence of the Li I (6708\\AA) absorption line and the Ca II chromospheric lines in their spectra as well as on their X-ray emission.  Murphy (1969) was the first to propose that PTTS could be found in binary systems composed of early-type primaries and late-type secondaries, given that the MS lifetime of high-mass stars is comparable to the contraction timescale of late-type stars to the MS.  Gahm et al. (1983) and Lindroos (1985) carried out photometric and spectroscopic observations of 253 binary systems and Lindroos (1986) was able to isolate a sample of 78 visual binary systems with likely PTTS secondaries.  The X-ray emission from these PTTS candidates as observed by {\\em ROSAT} has been studied by Schmitt et al. (1993) and Hu\\'elamo et al. (2000). The derived X-ray luminosities  for most of the late-type stars range between $\\lg{L_{\\rm x}} = 28 - 31\\,[{\\rm erg/s}]$. These values are comparable to those reported for T Tauri stars stars (Neuh\\\"auser et al. 1995), showing that PTTS although somewhat more evolved are still strong X-ray emitters.  PMS late-type stars are also known to emit X-ray flares (see e.g. Feigelson \\& DeCampli 1981, Montmerle et al. 1983, Stelzer et al. 2000).  Therefore, we have studied the X-ray lightcurves of all Lindroos PTTS candidates observed by the {\\em ROSAT} satellite (see Hu\\'elamo et al. 2000 for further details) in order to search for X-ray variability due to activity episodes.  As a result of our study of the {\\em ROSAT} PSPC and HRI lightcurves of the Lindroos systems, we have detected a large flare from HD\\,560, a binary system composed of a B9 and a G5 star. A previous flaring episode from HD\\,560 observed with {\\em EXOSAT} was reported by Tagliaferri et al. (1988). The {\\em EXOSAT} Medium Energy experiment provided no spatial resolution, however. Therefore, it was unclear whether to ascribe the event to HD\\,560 or to the nearby Seyfert type I galaxy, QSO\\,0007+107, located at $\\alpha = 00^{\\rm h}10^{\\rm m}31.01^{\\rm s}$ and $\\delta = 10^\\circ58^\\prime 29.5^{\\prime\\prime}$. While Pounds \\& Turner~(1987) attributed the flare to the galaxy, Tagliaferri et al. (1988) argued in favor of HD\\,560. The observation of another flare with the {\\em ROSAT} PSPC whose spatial resolution is sufficient to identify the flaring source with HD\\,560, supports the arguments put forth by Tagliaferri et al. (1988). In this paper we provide an analysis of the {\\em ROSAT} data of HD\\,560, and discuss the flare event. ",
        "conclusions": "We have presented a large X-ray flare from the Lindroos binary system HD\\,560 observed with the {\\em ROSAT} PSPC. Earlier observations of a flare by {\\em EXOSAT} had been reported, but left the identification with HD\\,560 ambiguous due to the lack of spatial resolution of the Medium Energy experiment.  The maximum X-ray luminosity during the {\\em ROSAT} flare corresponds to $\\sim 10^{31}$\\,erg/s, a value typical for young late-type stars. We have computed the bolometric luminosity for the G and the B star in HD\\,560 from the V magnitudes (Lindroos 1985) and the bolometric correction listed by Schmidt-Kaler (1982). Using the observed quiescent emission we derive the ratio of X-ray to bolometric luminosity \\lgLrat for primary and secondary: $(\\lgLrat)_{\\rm prim} = -4.81$ and $(\\lgLrat)_{\\rm sec} = -2.76$. The latter value agrees well with the `canonical' value of $-3$ for late-type stars, while the value for the primary is neither typical for late-type nor for early-type stars. Therefore, supposedly, the flare has erupted on the G5 component of the system. However, the binary is unresolved, and it can not be ruled out that the flare originated on the B9 star. Whether or not late-B type stars emit X-rays is an unsolved problem.  We have analysed X-ray spectra for both the quiescent and the flare phase, and find acceptable fits for both phases with a two-temperature Raymond-Smith model. During the flare the higher temperature showed an increase with respect to the quiescent stage, while the lower temperature remained unaffected. A one-temperature model with free abundance provides fits of comparable quality. However, the rather similar temperature derived for the flare and the quiescent state with the one-temperature model is difficult to reconcile with coronal heating expected during flare outburst."
    },
    "0010/astro-ph0010506_arXiv.txt": {
        "abstract": " ",
        "introduction": " ",
        "conclusions": "\\subsection{The Abundances of \\NH3 in the Galaxies} The abundance of \\NH3 in Maffei 2 relative to H$_2$ ([\\NH3]/[H$_2$]) is calculated from their column densities. As the column density of H$_2$, we employed 1.4 $\\times$ 10$^{22}$ cm$^{-2}$ (H\\\"uttemeister et al. 1995). The obtained abundance is 1 $\\times$ 10$^{-7}$, which is higher than the corresponding values of 2.7 $\\times$ 10$^{-8}$ in NGC 253 and 1.9 $\\times$ 10$^{-8}$ in IC 342 (Takano et al. 2000). Maffei 2 has the highest abundance of \\NH3 among galaxies where \\NH3 is detected. \\par Abundances of \\NH3 in our Galactic sources strongly depend on regions. The abundances are between 1$\\times$10$^{-8}$ and 10$^{-4}$ as follows. In our Galactic center Sgr B2 the abundances are reported to be (1 -- 10)$\\times$10$^{-8}$ (Irvine et al. 1987) and 8$\\times$10$^{- 8}$ -- 10$^{-4}$ (H\\\"uttemeister et al. 1993). In the cyanopolyyne peak of TMC-1, which is a quiescent young dark cloud where ammonia formation is not yet prominent, the abundance is 2$\\times$10$^{-8}$ (Irvine et al. 1987). In star forming regions abundances are typically more than 10$^{-7}$. For example the abundance in the Orion ridge is 2$\\times$10$^{-7}$  (Irvine et al. 1987). The abundance in Maffei 2 is similar to this value. \\par The abundance in Maffei 2 is much higher than the upper limit of 1.4 $\\times$ 10$^{-9}$ in M82 (at T$_{rot}$ = 30 K, Takano et al. 2000), though star formation activity and amount of molecular gas of Maffei 2 are significantly lower than those in M82; peculiarity of the molecular abundance in M82 is further supported by the present result. \\subsection{Ortho-to-para Abundance Ratios of \\NH3} There are ortho-\\NH3 ($K$ = 3n, n is integer) and para-\\NH3 ($K$ $\\not=$ 3n) due to the three equivalent hydrogen nuclei with spin 1/2. No dipole and collisional transitions are allowed between the ortho and para states. The spin statistical weight is 4 : 2 for ortho : para (e.g. Townes, Schawlow 1975). On the other hand, the number of the para levels is almost two times larger than that of the ortho levels, and consequently the ortho-to-para abundance ratio approaches to 1.0, if \\NH3 is produced and equilibrated under high-temperature conditions ($\\gtsim$ 40 K). On the other hand, the ratio becomes quite large if \\NH3 is produced and equilibrated under low-temperature conditions (e.g. more than 10 at 5 K), because the lowest level, (0,0), belongs to ortho. The dependence of ortho-to-para abundance ratios of \\NH3 on temperatures, which determine the distribution between ortho and para, was calculated and shown in figure 3 of Takano et al. (2000). \\par The ortho-to-para abundance ratios obtained in Maffei 2 are about 2.6 for all velocity components. The ratio at the total velocity component is 2.6$^{+1.1}_{-0.8}$, which corresponds to a temperature of distribution between ortho- and para-\\NH3 of 13$^{+5}_{-3}$ K. If this temperature can be  interpreted as a formation temperature of \\NH3, it was produced on low-temperature grain and subsequently evaporated to gas-phase. \\par Studies of \\NH3 abundance in the aspect of an ortho-to-para abundance ratio have been limited. In our Galaxy, Umemoto et al. (1999) recently obtained ratios of 1.3 -- 1.7, which are significantly higher than the value of high-temperature limit of 1.0, in the outflow lobe of a star forming region L1157. In external galaxies, ratios have been obtained only in NGC 253 (Takano et al. 2000). The values are quite different depending on the velocity components; $\\sim$ 1 for the relatively red-shifted component and more than $\\sim$ 14 for the relatively blue-shifted component. The ratio obtained in Maffei 2 is not so large as obtained in the relatively red-shifted component in NGC 253. \\subsection{Double Peak Spectra of \\NH3} As shown in figure 1, the lineshapes of the \\NH3 lines in Maffei 2 show double peak. Such double peak has already been reported at the CO $J$ = 1 -- 0 (Sargent et al. 1985; Weliachew et al. 1988), $J$ = 2 -- 1 (Sargent et al. 1985), and $J$ = 3 -- 2 (Hurt et al. 1993b; Mauersberger et al. 1999) transitions. In addition, the H$_2$CO 2$_{11}$ -- 1$_{10}$ and 2$_{12}$ -- 1$_{11}$ transitions (H\\\"uttemeister et al. 1997), and HNC $J$ = 1 -- 0 (H\\\"uttemeister et al. 1995) transition also show the double peak. Ishiguro et al. (1989) found a molecular ring with a diameter of about 500 pc $\\times$ 240 pc in the bar. Gas is deficient inside of the ring. These double peaks can be interpreted to originate from this ring structure. For CO and \\NH3, the velocity component at $\\sim$ 10 km s$^{-1}$ is clearly stronger than that at $\\sim$ --90 km s$^{-1}$. This may be related to the small scale asymmetries in the molecular bar (Ishiguro et al. 1989) and to the large scale asymmetry of the arms found by the infrared observation (Hurt et al. 1993a). \\par \\vspace{1pc}\\par We thank all members of the 45-m telescope in Nobeyama for support of our observations. We are grateful to H. Spinrad for informing us the coordinate before publication. We thank P. Schilke for his support in the preliminary survey of \\NH3 with the Effelsberg 100-m telescope."
    },
    "0010/astro-ph0010489_arXiv.txt": {
        "abstract": "We report on nine years of timing observations of PSR~J2019+2425, a millisecond pulsar in a wide 76.5~day orbit with a white dwarf.  We measure a significant change over time of the projected semi-major axis of the orbit, ${\\dot{x}/x}=1.3\\pm0.2\\times 10^{-15}$\\,s$^{-1}$, where $x\\equiv(a_1\\sin i)/c$. We attribute this to the proper motion of the binary.  This constrains the inclination angle to $i<72^\\circ$, with a median likelihood value of $63^\\circ$. A similar limit on inclination angle arises from the lack of a detectable Shapiro delay signal.  These limits on inclination angle, combined with a model of the evolution of the system, imply that the neutron star mass is at most 1.51\\,$M_\\odot$; the median likelihood value is 1.33\\,$M_\\odot$. In addition to these timing results, we present a polarization profile of this source.  Fits of the linear polarization position angle to the rotating vector model indicate the magnetic axis is close to alignment with the rotation axis, $\\alpha<30^\\circ$. ",
        "introduction": "\\begin{figure}[b] {\\footnotesize $^1$Present address: NRAO, Box 2, Green Bank, WV 24944} \\end{figure} \\addtocounter{footnote}{1}  Neutron star masses measured in radio pulsar binary systems are consistent with a remarkably small range, \\mbox{$m=1.35\\pm 0.04$\\,M$_\\odot$} (Thorsett \\& Chakrabarty 1999)\\nocite{tc99}.  The best such measurements, those with uncertainty less than 5\\%, are of pulsars in tightly bound, highly eccentric neutron star--neutron star binaries.  In such systems, relativistic phenomena detected in radio pulse timing experiments allow the direct, high precision measurement of the masses of the component stars.  By contrast, measurement of masses of pulsars in neutron star--white dwarf systems tend to have lower precision, or to be statistical in nature.  Measurement of masses in these systems are of interest because their evolution is substantially different from that of neutron star--neutron star binaries.  In particular, neutron star--white dwarf systems go through an extended period of mass transfer, during which the secondary loses several tenths of a solar mass of matter, at least some of which is accreted onto the neutron star, making the system visible as a low mass X-ray binary.  Thus, one might expect the neutron stars in these systems to be more massive than those in neutron star--neutron star binaries.  In this paper, we describe pulse timing observations of PSR~J2019+2425, a millisecond pulsar in a 76.5~day orbit with a white dwarf.  Our observations constrain the ({\\it a priori} unknown) inclination angle of the orbit. By combining this constraint with a theory of orbital evolution, we determine an upper limit of the mass of the pulsar.  We have previously reported on this pulsar in Nice \\& Taylor (1995)\\nocite{nt95}.  The present work represents a tripling of the time span of the observations.  In \\S\\ref{sec:obs} we describe the data acquisition.  In \\S\\ref{sec:timing} we present an analysis of the pulse arrival times. Implications for the neutron star mass are given in \\S\\ref{sec:mass}. Polarimetry of the pulsar is discussed in \\S\\ref{sec:polar}.  \\bigskip ",
        "conclusions": "Using two different methods, our timing observations of PSR~J2019+2425 constrain the orbital inclination to be $i\\lesssim 72^\\circ$.  Combined with a model for evolution of the system, this limits the neutron star mass to $m_1<1.51$\\,M$_\\odot$, with the median likelihood $m_1=1.33$\\,M$_\\odot$ for a population~II secondary.  For a population~I secondary the mass values would be lower.  The use of $\\dot{x}$ to place an upper limit on $m_1$ based on a measurement of $\\dot{x}$ has been used for one other pulsar.  PSR~J0437$-$4715 has $\\dot{x}=8\\times 10^{-12}$, $x=3.367$\\,lt-s, $\\mu=141$\\,mas/yr, which gives a limit $i<43^\\circ$ \\cite{sbm+97}.  When combined with the $P_b-m_2$ relation of Rappaport {\\it et al.}, this gives $m_1<1.51$\\,M$_\\odot$ \\cite{tc99}, coincidentally the same value as we find for PSR~J2019+2425.  Our finding adds to the growing body of evidence that neutron stars in neutron-star--white dwarf orbits are not much more massive than those in neutron star--neutron star binaries.  (See Thorsett \\& Chakrabarty 1999 for a comprehensive review of pulsar mass measurements\\nocite{tc99}).  It is somewhat surprising that the neutron stars in these systems are not more massive, as the secondaries in these systems must lose several tenths of a solar mass as they evolve towards white dwarfs.  Much of this mass could, in principle, be accreted onto the neutron stars during the low-mass X-ray binary (LMXB) phase. To keep the neutron star mass low, most of the transferred mass must leave the system, perhaps being ejected from the vicinity of the neutron star via the ``propeller effect'' \\cite{is75}. Taam, King \\& Ritter (2000)\\nocite{tkr00} suggest that wide ($P_{\\rm b}$ $\\gtrsim$ 2 days) LMXBs are likely to be transient, with accretion occurring in short, super-Eddington outbursts that expel most of the transferred mass in a wind.  Low neutron star masses such as that of PSR~J2019+2425 lend strong support to this scenario."
    },
    "0010/astro-ph0010395_arXiv.txt": {
        "abstract": "We extend a previous bispectrum analysis of the COBE-DMR 4 year maps,  allowing for the presence of correlations between all possible angular scales. We find that the non-Gaussian signal found earlier for bispectrum components including adjacent modes does not extend to triplets of modes with larger separations. Indeed  for all separations $\\Delta\\ell >1$ we find that the COBE-DMR data is very Gaussian. The implication seems to be that the previously detected non-Gaussian scale-scale correlation falls off very quickly with mode separation. ",
        "introduction": "\\label{introduction}  The issue of whether or not the primordial fluctuations are Gaussian is of fundamental importance  in theories of structure formation. Gaussian large-scale primordial fluctuations are believed to be the hallmark of the simplest inflationary scenarios. In these theories vacuum quantum fluctuation, with Gaussian statistics, are stretched by a period of exponential expansion to give rise to the large scale fluctuations we observe today. In contrast, topological defect theories, which are also possible candidates for structure formation \\cite{string1,string2}, are believed to produce non-Gaussian density fields.  Non-Gaussianity is also believed to be associated with some non-minimal inflationary models \\cite{Salopek:1992jq,Linde:1997gt,Contaldi:1999jr}, such as those generating isocurvature fluctuations \\cite{peebles}.  Several recent \\cite{fmg,pvl98,magueijo1} papers have shown indications of non-Gaussianity in the cosmic microwave background (CMB) temperature fluctuations, as measured by the COBE-DMR experiment. The first of these detections\\\\ \\cite{fmg} was later found to be due to an experimental systematic \\cite{banday}, and the second \\cite{pvl98} to an error of method \\cite{belen}. The third claim \\cite{magueijo1}, however, remains unassailable. In that work a bispectrum analysis was carried out, using adjacent scales $\\{\\ell-1,\\ell,\\ell+1\\}$ as components of the bispectrum. A significant deviation from Gaussianity was found, with a high confidence level.  An obvious question is whether this non-Gaussianity extends to components correlating scales with a larger separation, say $\\{\\ell-\\Delta\\ell,\\ell,\\ell+\\Delta\\ell\\}$, with $\\Delta\\ell >1$. The purpose of this paper is to answer this question. As we shall see the non-Gaussian signal found in \\cite{magueijo1} does not appear to extend to larger separations $\\Delta\\ell$. ",
        "conclusions": "The comparison between $P(J^{ \\Delta \\ell}_\\ell)$ and the observed $J^{ \\Delta \\ell}_\\ell$ for various $\\Delta \\ell$ is displayed in figs. ~\\ref{fig1} - ~\\ref{fig4}.  The observed values agree well with the probability distributions from the Monte Carlo distributions: implying that as far as $\\Delta\\ell>1$ bispectrum components are concerned the data supports Gaussianity.  To quantify just how well the observations agree with Gaussianity, we do a goodness of fit statistic on the data.  We calculate the $\\chi^2$ value for the different $\\ell$ separations, and compare with the respective distributions $F(\\chi^2)$. The rationale behind considering separate $\\chi^2$ was explained in \\cite{magueijo1}\\footnote{This is to be contrasted with more dubious philosophy towards combining separate tests expressed in \\cite{teg,kog}.}, and for the distributions $P(^{ \\Delta \\ell}_\\ell)$ found (which are very close to Gaussians) it makes sense to use the standard $\\chi^2$: \\begin{equation} \\chi^2 = \\frac{1}{N}{\\sum_{i=1}^N} \\frac{\\left( x_i-\\mu_i\\right)}{\\sigma_i^2} \\end{equation} where $N$ is the number of different observables and  $\\mu_i$ and $\\sigma_i^2$ are the expectation value and the variance of the $i$th observable respectively. This reduces to the expression proposed in \\cite{fmg} in our case. The $F(\\chi^2)$ are then calculated through further Monte Carlo simulations, so that possible correlations between the different $\\ell$ may be taken into account.  The $\\chi^2$ distributions and corresponding observed values are shown in fig ~\\ref{fig5}.  As we can see $\\chi^2\\approx 1$ for all separations $\\Delta\\ell >1$, with deviations well within the spread expected from $F(\\chi^2)$. This is to be contrasted with the results obtained for $\\Delta\\ell=1$, that is bispectrum components measuring correlations between adjacent modes. For these $\\chi^2\\ll 1$, meaning that the observed $J_\\ell$ do not display the scatter around zero expected from a Gaussian distribution. We now found that this distinctive non-Gaussian signal does not extend to bispectrum components with  $\\Delta\\ell>1$. The implication seems to be that the inter-mode correlations suggested by the findings of \\cite{magueijo1} fall off quickly with the scale separation, something which conforms to theoretical prejudice.  In a recent paper \\cite{kog} a variety of components of the bispectrum was studied in the context of a semi-analytical texture model. Such a model is at the very least oversimplified, and probably has little to do with real-life textures. In any case that work stresses the importance of analysing off-diagonal components, as we have done in this paper.      \\begin{figure} \\begin{center} \\psfig{file=lsep2.ps,width=12cm} \\end{center} \\caption{The vertical thick dashed lines represent the values of the observed $J^{ \\Delta \\ell = 2}_\\ell$, for different values of $\\ell$. The solid lines are the corresponding probability distribution functions of the $J^{ \\Delta \\ell=4}_\\ell$s for a Gaussian sky with extended galactic cut and DMR noise, as inferred from 10000 realizations.} \\label{fig1} \\end{figure} \\begin{figure} \\begin{center} \\psfig{file=lsep3.ps,width=12cm} \\end{center} \\caption{The vertical thick dashed lines represent the values of the observed $J^{ \\Delta \\ell = 3}_\\ell$, for different values of $\\ell$. The solid lines are the corresponding probability distribution functions of the $J^{ \\Delta \\ell=4}_\\ell$s for a Gaussian sky with extended galactic cut and DMR noise, as inferred from 10000 realizations.} \\end{figure}  \\begin{figure} \\begin{center} \\psfig{file=lsep4.ps,width=12cm} \\end{center} \\caption{The vertical thick dashed lines represent the values of the observed $J^{ \\Delta \\ell = 4}_\\ell$, for different values of $\\ell$. The solid lines are the corresponding probability distribution functions of the $J^{ \\Delta \\ell=4}_\\ell$s for a Gaussian sky with extended galactic cut and DMR noise, as inferred from 10000 realizations.} \\end{figure}  \\begin{figure} \\begin{center} \\begin{tabular}{c} \\psfig{file=lsep5.ps,width=8cm} \\\\ \\psfig{file=lsep6.ps,width=8cm} \\end{tabular} \\end{center} \\caption{The vertical thick dashed lines represent the values of the observed $J^{ \\Delta \\ell}_\\ell$s, $\\Delta \\ell = 5$, and $6$ for different values of $\\ell$.  The solid lines are the corresponding probability distribution functions of the $J^{ \\Delta \\ell}_\\ell$s for a Gaussian sky with extended galactic cut and DMR noise, as inferred from 10000 realizations.} \\label{fig4} \\end{figure}  \\begin{figure} \\begin{center} \\begin{tabular}{cc} \\psfig{file=chi2.ps,width=6cm} & \\psfig{file=chi2a.ps,width=6cm} \\\\ \\psfig{file=chi2b.ps,width=6cm} & \\psfig{file=chi2c.ps,width=6cm} \\\\  \\end{tabular} \\end{center} \\caption{The dashed lines are observed values of $\\chi^2$ from the COBE data for different $\\ell$ separations $\\Delta \\ell = 2,3,4$ and $5$.  The solid lines are corresponding probability distribution functions for $\\chi^2$ for a Gaussian sky with extended galactic cut and DMR noise, as inferred from 10000 realisations.} \\label{fig5} \\end{figure}"
    },
    "0010/astro-ph0010348_arXiv.txt": {
        "abstract": "In some circumstances, the mass accretion rate $\\Mdot_\\star$ onto a compact star may depend not only on external boundary conditions, but also on the radius $R_\\star$ of the star.  Writing the dependence as $\\Mdot_\\star\\propto R_\\star^p$, we estimate $p$ for transient binary systems in a quiescent state.  We use the observed luminosities $L$ of accreting neutron stars ($R_\\star\\sim 10^6~{\\rm cm}$) in soft X-ray transients and white dwarfs ($R_\\star\\sim 10^9~{\\rm cm}$) in similar cataclysmic variables, and estimate $\\Mdot_\\star$ in each system through the relation $L\\approx GM_\\star\\Mdot_\\star/R_\\star$, where $M_\\star$ is the mass of the star.  From the available data we infer that $p\\sim 0.9\\pm 0.5$.  This radial dependence is consistent with radiatively inefficient accretion flows that either are convective or lose mass via a wind. ",
        "introduction": "For steady accretion onto a compact object, it is often assumed that the mass accretion rate onto the central star $\\Mdot_\\star$ is determined only by boundary conditions at some large radius where the gas is initially introduced into the accretion flow.  This assumption is valid for a number of idealized accretion solutions: (i) spherical accretion (Bondi 1952), (ii) steady accretion via a thin disk (Shakura \\& Sunyaev 1973; Novikov \\& Thorne 1973), and (iii) steady accretion via an advection-dominated accretion flow (ADAF; Ichimaru 1977; Narayan \\& Yi 1994; 1995a,b; Abramowicz et al. 1995; Chen et al. 1995).  Recently, it has become clear that under some circumstances $\\dot M_\\star$ may depend not only on outer boundary conditions but also on the radius of the central star $R_\\star$.  A power-law dependence has been proposed: \\begin{equation} \\Mdot_\\star \\propto R_\\star^{p}. \\end{equation} Gruzinov (1998), for instance, argued that spherical accretion in the presence of strong conduction behaves according to equation (1), with $p$ of order unity.  In the case of a thin accretion disk, it has been known for a number of years that large changes in the opacity of the accreting gas as a result of hydrogen ionization lead to a thermal instability (see Cannizzo 1993 for a review).  This instability causes the well-known transient behavior of accreting white dwarfs (WDs) in cataclysmic variables (CVs), and accreting neutron stars (NSs) and black holes in soft X-ray transients (SXTs).  If accretion in the quiescent state of transient binaries occurs via a pure thin disk (which is in some doubt, see below), then $\\Mdot_\\star$ is expected to scale according to equation (1) with a large value of $p\\sim2.7$ (Ludwig, Meyer-Hofmeister \\& Ritter 1994; Hameury et al. 1998; Menou et al. 1999a).  In the case of advection-dominated accretion, the radius of the star can again be important in some circumstances.  For instance, if there are strong outflows, as seems likely (Narayan \\& Yi 1994, 1995a), the mass accretion rate is expected to behave as in equation (1). Blandford \\& Begelman (1999) suggested that ADAFs with outflows, which they named ADIOS, may have values of $p$ in the range $0-1$. Igumenshchev \\& Abramowicz (1999, 2000) found ADIOS-like flows in numerical simulations when they used large values of the viscosity parameter $\\alpha>0.3$.  Their simulations suggest that $p\\sim0.3-0.5$; this estimate is, however, very approximate since the results from the simulations are not consistent with a power-law behavior.  Narayan \\& Yi (1994, 1995a) argued that advective flows should, in addition to having outflows, also be convectively unstable. Igumenshchev \\& Abramowicz (1999, 2000) found well-developed convection in numerical simulations when they used small values of $\\alpha<0.1$.  These convective flows correspond to a new form of quasi-spherical accretion (Stone, Pringle \\& Begelman 1999), which is called a convection-dominated accretion flow or CDAF.  Both the numerical simulations and analytical work (Narayan, Igumenshchev \\& Abramowicz 2000; Quataert \\& Gruzinov 2000; Igumenshchev, Abramowicz \\& Narayan 2000) show that CDAFs follow equation (1) with $p\\sim1$.  In this {\\it Letter} we are concerned with transient binaries such as CVs and SXTs.  Narayan, McClintock \\& Yi (1996) and Lasota, Narayan \\& Yi (1996) argued that quiescent SXTs cannot have pure thin disks; a similar argument had been made earlier by Meyer \\& Meyer-Hofmeister (1994) for CVs.  Narayan et al. (1996) suggested that SXTs, and by extension CVs, have composite accretion flows in which a thin disk is present at large radii,  beyond $\\sim 10^9-10^{10}$ cm, and an ADAF is present at smaller radii.  In view of the recent work described above, it is possible that, rather than an ADAF, the inner flow might consist of an ADIOS or a CDAF.  We thus have four different possibilities for the accretion flow in quiescent CVs and SXTs, each with a different prediction for $p$; (i) pure thin disk ($p=2.7$), (ii) ADAF ($p=0$), (iii) ADIOS ($p\\sim0.3-0.5$), and (iv) CDAF ($p\\sim1$).  Theoretical analysis is presently incapable of discriminating among these four possibilities.  Here we use a novel empirical method to infer the value of $p$.  We consider CVs and SXTs with similar orbital periods (in the range $\\sim2-20$ hours), which ensures that the outer boundary conditions of their accretion flows are similar.  The mass transfer rate from the companion star in a semi-detached binary system depends primarily on the binary period, and is fairly insensitive to the mass of the compact star (Menou et al. 1999a).  This is especially true for systems with orbital periods of a fraction of a day or shorter, in which mass transfer is driven primarily by gravitational radiation losses.  Thus, for the CVs and SXTs considered here, the mass supply rate is likely to be similar.  On the other hand, the radius of the accreting star in the systems we consider spans a wide range, from $\\sim 10^9~{\\rm cm}$ for WDs in CVs to $\\sim 10^6~{\\rm cm}$ for NSs in SXTs.  For accretion onto a star with a hard surface (as distinct from a black hole, see Narayan, Garcia \\& McClintock 1997), the accretion luminosity $L$ has a simple dependence on $R_\\star$ and $\\Mdot_\\star$: \\begin{equation} L\\approx  {GM_\\star\\Mdot_\\star \\over R_{\\star}}~, \\label{eq:1} \\end{equation} where $M_\\star$ is the mass of the accreting star. Therefore, given the observed $L$, and estimates of $M_\\star$ and $R_\\star$, we may infer the value of $\\Mdot_\\star$.  Since we select systems with similar outer boundary conditions, but with a wide range of $R_\\star$, we should be able to estimate the dependence of $\\Mdot_\\star$ on $R_\\star$, and thus the value of $p$.  Note that correction factors of order unity are unimportant in this analysis, because the value of the gravitational potential $GM_\\star/R_\\star c^2$, or equivalently $L/\\Mdot_\\star c^2$, differs by three orders of magnitude between NSs and WDs.  Before proceeding, we note that Medvedev \\& Narayan (2000) recently obtained a new hot quasi-spherical accretion solution which applies under some circumstances to a rotating star with a surface.  The luminosity of this solution does not scale according to equation (1), but depends primarily on the radius and the spin of the star.  The estimate of $p$ derived in this {\\it Letter} would not be valid if accretion in the CVs or NS SXTs we consider is described by this solution.  In \\S 2 we present the available data on the luminosities of WDs (\\S 2.1) and NSs (\\S 2.2) with similar binary periods and companions. In \\S 3, we use the data to infer the best-fit value of $p$. Finally, we summarize our conclusions and discuss their implications in \\S 4. ",
        "conclusions": "The accretion luminosities of NSs and WDs in quiescence with similar binary companions are comparable (see Table 1). This surprising result implies that the accretion rate at the surface of a WD, namely $\\Mdot_\\star(10^9~{\\rm cm})$, is larger by three orders of magnitude than that on the surface of a NS, $\\Mdot_\\star(10^6~{\\rm cm})$.  In both cases the radius of the hard surface where the accretion flow terminates is much smaller than the outer boundary of the accretion flow, and hence the flow should exhibit a similar behaviour on scales larger than its inner edge.  Figure 1 implies that the mass accretion rate varies with radius roughly as $\\Mdot_\\star \\propto R_\\star^{0.9\\pm 0.5}$. This behaviour is remarkably similar to that predicted for CDAFs (Narayan et al. 2000; Quataert \\& Gruzinov 2000; Igumenshchev et al. 2000).  The data are marginally consistent with an ADIOS model (Blandford \\& Begelman 1999), for which Igumenshchev \\& Abramowicz (1999, 2000) found $p\\sim0.3-0.5$.  The data do not support the assumption of a constant $\\Mdot_\\star$, as in a pure ADAF model (e.g. Narayan et al. 1996), or a very steep scaling ($p\\sim2.7$) as in a pure quiescent thin disk (e.g. Menou et al. 1999a).  Unfortunately, the quantitative results are subject to large uncertainties due to the small number statistics of the relevant systems.  They could also be affected by a number of systematic effects.  One potential systematic effect is probably not important, namely we do not believe that there is a serious bias in the sample of systems we have used.  Our sample of CVs is an unbiased sample of binaries with orbital periods in the range of interest.  Also, virtually all known NS SXTs have been detected in X-rays in quiescence; therefore, the sample of NS SXTs in Table 1 is likely to be complete and unbiased (except for the one system we did not include because its period is not known, see \\S2.2).  One potentially important systematic effect is related to the fact that the luminosities in Table 1 may not represent just the accretion luminosities in quiescence, but may also include a cooling component which originates from the heat deposited in the accretor during an earlier high $\\Mdot$ (outburst) state.  If some of the quiescent emission in CVs is due to the cooling of the WD (Sion 1999), then $p$ would be smaller than the value we obtained; in principle, if the accretion luminosity in quiescent CVs is very much smaller than the observed luminosity, $p$ could be as small as zero, as required for a pure ADAF model.  The luminosity of a cooling WD should decay with time in a predictable fashion, so this scenario could be tested with careful observations of the CVs listed in Table 1.  On the other hand, if the quiescent luminosities of NS SXTs are primarily due to NS cooling (as proposed by Brown et al. 1998), then $p$ would be larger than our estimated value, perhaps as large as the 2.7 predicted for a pure thin disk.  For the reasons given in \\S2.2, we consider this unlikely.  A systematic effect with an opposite sign involves the possibility that we have underestimated the bolometric luminosities of the systems under consideration by ignoring the emission at unobserved photon frequencies.  This is probably not a serious problem, as CVs appear to radiate primarily in optical, UV and EUV, where there are adequate observations, and NS SXTs emit primarily in the X-ray band (see \\S2.1 and \\S2.2).  Finally, it is possible that magnetic fields of NSs or WDs reduce the mass accretion rate onto the star, thereby giving an artificially low estimate of $\\Mdot_\\star$.  This is more likely in NS SXTs than in CVs (Menou et al. 1999b).  If the effect is important, then $p$ would be smaller than our estimate.  A key aspect of our analysis is the relation between the mass accretion rate and the luminosity given in equation (2).  This formula is valid so long as the accreting star has a hard surface.  If the accretor is a black hole, and if accretion occurs via a radiatively inefficient flow, such as an ADAF or an ADIOS or a CDAF (but not a thin disk), then the accretion luminosity is expected to be significantly less than that given in equation (2).  This is because the energy advected with the accretion flow can disappear through the event horizon (Narayan \\& Yi 1995b; Narayan et al. 1997).  The quiescent luminosities of BH SXTs do appear to be significantly lower than the luminosities of NS SXTs.  This is consistent with the presence of an event horizon in black hole SXTs (Narayan et al. 1997; Menou et al. 1999b)."
    },
    "0010/astro-ph0010381_arXiv.txt": {
        "abstract": "The energy spectrum of cosmic-ray antiprotons ($\\bar{p}$'s) has been measured by BESS successively in 1993, 1995, 1997 and 1998. In total, 848 $\\bar{p}$'s were clearly identified in energy range 0.18 to 4.20 GeV. From these successive measurements of the $\\bar{p}$ spectrum at various solar activity, we discuss about the effect of the solar modulation and the origin of cosmic-ray $\\bar{p}$'s. Measured $\\bar{p}/p$ ratios were nearly identical during this perfiod, and were consistent with a prediction taking the charge dependent solar modulation into account.  \\keyword Cosmic-ray, Antiproton, Measurements, Solar modulation} \\PACS 98.70.Sa, 95.85.Ry ",
        "introduction": "The origin of cosmic-ray antiprotons ($\\bar{p}$'s) is still holding a fundamental question even after their spectrum has been measured precisely in our previous work \\cite{BESS97}. In recent solar-minimum period, we have collected 458 $\\bar{p}$'s in the energy range between 0.18 and 3.56 GeV and observed a clear peak in the spectrum around 2 GeV, a generic feature of ``secondary'' $\\bar{p}$'s which are produced by the interaction of Galactic high-energy cosmic-rays with the interstellar medium. It has become evident that the dominant component of cosmic-ray $\\bar{p}$'s is the secondary $\\bar{p}$'s. While the energy spectrum and absolute flux are well reproduced by several theoretical calculations \\cite{BERG,BIEB}, there remains some diversity among those calculations in the low energy region. They seem to arise from the different treatment of tertiary interactions of $\\bar{p}$'s with the interstellar medium, $\\bar{p}$ production mechanism such as nuclear sub-threshold effects, galactic propagation and solar modulations. Moreover, although the data have still large statistical errors at low energies, the observed flux appears to be higher than those calculations in the low energy region below 1 GeV where the secondary flux decreases sharply. Therefore, we can not rule out the primary origins, such as annihilation of neutralinos \\cite{BERG,SILK,MI96} or evaporation of primordial black holes \\cite{HA75,MA96}.  The cosmic-ray particles are decelerated in entering the heliosphere by the expanding solar magnetic disturbances, and their energy spectra observed at 1 A.U. should be deformed. The flux variation is expected to be small for secondary $\\bar{p}$'s at low energies because of their relatively hard spectrum, while any softer component would exhibits a large variation. It has been pointed out that additional soft component from a novel source can be singled out by measuring the $\\bar{p}$ spectra at different solar periods \\cite{MI96}. It has been recently suggested that the solar magnetic polarity and the charge dependent effect cannot be neglected and the modulation follows a 22-year cycle rather than an 11-year cycle \\cite{BIEB}. Actually, measurements of $e^{-}/{\\rm He^{2+}}$ ratio \\cite{e_he_ratio} and $e^{-}/p^{+}$ ratio \\cite{e_p_ratio} variations over a long period appear to support this charge dependence. Since protons and antiprotons are different only in their charge sign, the simultaneous measurements of both particles are most suitable for studying the charge dependence of solar modulation.  We report here a new measurement of cosmic-ray $\\bar{p}$ spectrum in the energy range between 0.18 and 4.20 GeV, based on 384 events detected by the BESS spectrometer in 1998 flight. In total, 848 $\\bar{p}$ events has been accumulated from the measurement in successful four flights in 1993, 1995, 1997 and 1998. The spectrum measured in 1998 is compared with the previous results \\cite{BESS97,YOMO,IMAX,CAPRICE,MA98,MASS91}, in order to observe changes with solar activity. ",
        "conclusions": "Fig.\\ref{fig:pbflux} shows the BESS 1998 $\\bar{p}$ spectrum, together with the previous measurements \\cite{BESS97,YOMO,IMAX,CAPRICE,MA98,MASS91}. By using 1998 flight data, we have measured the $\\bar{p}$ spectrum in a wider energy region of 0.18 to 4.20 GeV and detected again a clear peak around 2 GeV as we did in 1997. As recently pointed out \\cite{MI96}, increase of solar activity suppresses any soft primary $\\bar{p}$ component while only modestly affects the secondary $\\bar{p}$ spectrum. The solar activity at the time of the BESS 1998 flight was greater than the 1995 and 1997 flights \\cite{NEUT}. Therefore, the shape of the 1998 spectrum should be closer to the shape of the spectrum composed only of secondary $\\bar{p}$, even if the primary $\\bar{p}$'s existed. Shown also in Fig.\\ref{fig:pbflux} is a recent theoretical calculation \\cite{FRB} for the secondary $\\bar{p}$ at the BESS 1998 flight (solar modulation parameter $\\phi_{98}$=610MV, which was determined from BESS 1998 proton spectrum \\cite{SAN98}). In the whole energy region, this calculation agrees well with our spectrum at 86\\% confidence level. This implies that secondary $\\bar{p}$ is the dominant component of the cosmic-ray $\\bar{p}$'s and the model used for the calculation is basically correct. The expected spectrum at BESS 1997 was also calculated in the same way by using the solar modulation parameter  $\\phi_{97}$=500MV as represented by the dashed curve in Fig.\\ref{fig:pbflux}. It appears to be insensitive to the change of the modulation parameter. This is because the interstellar $\\bar{p}$ flux steeply drops toward the low energy in contrast to the interstellar proton spectrum \\cite{LAB97}. In fact, the measured spectra are nearly constant in the peak region. However, in the low energy region (0.18 -- 0.78 GeV), the agreement between the 1997 spectrum and the calculation is less consistent. Moreover, the 1995 spectrum shows larger deviation from the calculation in spite of similar modulation parameter ($\\phi_{95}$=540MV) as the BESS 1997. This might be due to statistical fluctuation, or might suggest a contribution of low-energy primary $\\bar{p}$ from novel source such as annihilation of neutralinos \\cite{BERG,SILK,MI96} or the evaporating primordial black holes \\cite{HA75,MA96}. Because $\\bar{p}$'s from these ``primary'' sources, if they exist, are expected to be prominent at low energies and to exhibit large solar modulations \\cite{MI96}.  In order to study the solar modulation, the $\\bar{p}/p$ ratio is more adequate, which is expected to clarify the charge-dependence of the solar modulation \\cite{BIEB}. In Fig. \\ref{fig:ppratio} the BESS 1998 $\\bar{p}/p$ ratio shows almost similar spectrum as the spectrum at the solar minimum (the 1997 data), especially around 2 GeV where we have better statistics. The lowest energy point of our data might suffer some uncertainties in the proton flux due to large subtraction of atmospheric secondary protons, which is significant in the low energy region \\cite{SAN98}. Despite the general advantage of taking $\\bar{p}/p$ ratio to cancel various systematic errors, such as live-time and geometrical acceptance, it should be noted that the most significant systematic errors in the analysis come from the uncertainties in the atmospheric $\\bar{p}$ calculation and in the interaction losses, which can not be canceled even in the form of the ratio. Taking the charge-dependence of the solar modulation into account \\cite{BIEB}, (i) the $\\bar{p}/p$ ratio is expected to be nearly identical during the positive Sun's polarity phase; and (ii) when the polarity switches into the negative, the $\\bar{p}/p$ ratio shall rapidly increase as represented by the dotted curve in Fig. \\ref{fig:ppratio}. The BESS 1993, 1995, 1997 and 1998 spectra, measured in the positive polarity phase, were nearly identical as shown in Fig. \\ref{fig:ppratio} and consistent with the prediction. It can be more clearly understood in the form of the ratios of $\\bar{p}/p$ ratio as shown in Fig. \\ref{fig:ratio}. The ratios of $\\bar{p}/p$ ratio normalized by the BESS 1997 data are consistently distributed around the unity within the statistical fluctuation.  Our measurements on the $\\bar{p}/p$ ratio are consistent with the prediction taking the charge dependent solar modulation into account. However, it is too early to conclude that the evidence for the charge dependence is exhibited. We are planning further annual flights in the following solar-maximum period in which the polarity switches into the negative. The observation of the increase would be a definite evidence for the importance of the charge dependent effect in the solar modulation. At the same time, measurements of the $\\bar{p}$ spectrum through the solar-maximum period itself have a crucial importance to determine the origin of the low energy  $\\bar{p}$'s. As previously mentioned, any soft $\\bar{p}$ component from the primary sources would be suppressed as the solar activity increases, while the secondary $\\bar{p}$ spectrum is affected modestly. Therefore, to observe variations in the low energy region would help us to draw a firm conclusion on whether we are seeing the primary $\\bar{p}$'s from novel sources.  Sincere thanks are given to NASA/GSFC/WFF balloon office and NSBF for balloon launch and recovery of the payload in Canada campaign. The authors deeply thank ISAS and KEK for their continuous support and encouragement  for the BESS experiment. They would specially thank Prof. T. Sumiyoshi, Dr. R. Suda and Mr. T. Ooba for their kind guidance and cooperation to develop Aerogel Cherenkov Counters for the flight in 1998. The analysis was performed by using the computing facilities at ICEPP, Univ. of Tokyo.  This experiment was supported by Grant-in-Aid for Scientific Research in Monbusho, Heiwa Nakajima Foundation in Japan, and by NASA in the USA."
    },
    "0010/astro-ph0010038_arXiv.txt": {
        "abstract": "We have used the Keck I telescope  to resolve  at  mid-infrared wavelengths the dust emission from HD 56126 (IRAS 07134+1005), a post-Asymptotic Giant Branch carbon star with a detached dust shell.  The gross morphology of the image can be explained by a strong wind which started to die about 1500 years ago.  If the star had an effective temperature, $T_{*}$, near 2600 K when it was losing ${\\sim}$3 ${\\times}$ 10$^{-5}$ M$_{\\odot}$ yr$^{-1}$, then during the past 1500 yr, the average value of $dT_{*}/dt$ has been +2.2 K yr$^{-1}$.  With such a time variation of the effective temperature,  the 11.7 ${\\mu}$m image can be approximately reproduced if the mass loss rate varied as $T_{*}^{-8.26}$, as proposed in recent models for dust-driven winds. Since the mass loss rate appears to be very sensitive to the effective temperature of the star, we speculate that the observed deviations from spherical symmetry of the dust shell can be explained by plausible variations in the surface temperature of the mass-losing star caused by rotation and/or magnetic fields. ",
        "introduction": "Mass loss with rates near ${\\sim}$ 10$^{-5}$ M$_{\\odot}$ yr$^{-1}$ during the Asymptotic Giant Branch (AGB) phase is important both in the evolution of the mass-losing star and in replenishing the interstellar medium (see, for example, Habing 1996).  After a star evolves off the AGB, its effective temperature rises, its radius shrinks and the wind dies (Vassiliadis \\& Wood 1993, Blocker 1995, Schroder, Winters \\& Sedlmayr 1999).  Stars slightly beyond the end of their AGB evolution can be identified by having only relatively cold dust since they are no longer losing much mass.  Therefore, they exhibit a large infrared excess but with  dust  temperatures typically less than 200 K (see, for example, Kwok 1993).  The detached dust shell can be extended over the sky and thus provide the opportunity to examine the history of the star's mass loss.  Meixner et al. (1997, 1999) and Dayal et al. (1998) have acquired mid-infrared images and resolved more than 10 post-AGB stars or ``transition objects\". Even in the best-studied cases, the angular diameter of the dust emission is  not much larger than the characteristic $0{\\farcs}9$ angular resolution available at the 4m-class telescopes used for these previous studies. With the 10m Keck telescope, it is possible to achieve significantly higher angular resolution (see, for example, Jura \\& Werner 1999, Jura, Chen \\& Werner 2000) and thus develop more exacting tests of models for dying winds.  Here, we report the result of using Keck to obtain mid-IR images of HD 56126 (m$_{V}$ = 8.27 mag, spectral type F5I), a post-AGB carbon star with a nebula displaying one of the largest angular sizes in the sample studied by Meixner et al. and extended optical reflection nebulosity as well (Ueta, Meixner \\& Bobrowsky 2000). This star is also a  ``21 ${\\mu}$m\" source (see, for example Kwok, Volk \\& Hrivnak 1999), a feature attributed to  TiC (von Helden et al. 2000). In  Section 2 we present our observations while  in Section 3, we describe our estimates for the physical properties of the circumstellar envelope and the central star.  In section 4, we present a model to explain the gross features of the system while  in Section 5, we discuss models to explain deviations from a spherically symmetric wind. In Section 6 we present our conclusions. ",
        "conclusions": "We have obtained high-resolution mid-IR images of the detached dust shell around the carbon star HD 56126 and propose the following:  1.  The gross features of the region beyond a radius of 1${\\farcs}5$ can be reproduced by an approximately spherical wind from a star which had a  mass loss rate of ${\\sim}$3 ${\\times}$ 10$^{-5}$ M$_{\\odot}$  yr$^{-1}$ while it was on the AGB. After the star left the AGB, its effective temperature increased with an average rate of +2.2 K yr$^{-1}$ and the mass loss rate varied as $T_{*}^{-8.26}$, as expected in models for dust-driven winds.  2.  The extreme sensitivity of the mass loss rate to the effective temperature of the star allows us to tentatively propose that both the ellipticity in the outer envelope and the inner blobs can be explained by plausible temperature variations on the surface of the star which might have resulted from rotation and/or large scale magnetic fields.  We thank M. Meixner for a particularly thoughtful referee's report and M. Morris for his comments. This work has been partly supported both at UCLA by NASA and at the Jet Propulsion Laboratory, California Institute of Technology, under an agreement with NASA."
    },
    "0010/astro-ph0010512_arXiv.txt": {
        "abstract": "We discuss the influence of the finite extension and the geometry of clusters of galaxies, as well as of the polytropic temperature profile, on the determination of the Hubble constant. ",
        "introduction": "The SZ effect (Sunyaev-Zel'dovich 1972) offers the possibility to put important constraints on the cosmological models. Combining the temperature change in the cosmic microwave background due to the SZ effect and the X-ray emission observations, the angular distance to galaxy clusters, and consequently the Hubble constant $H_o$, can be derived.\\\\ The SZ effect is difficult to measure, since systematic errors can be important.  For example, Inagaki et al. (1995) analysed the reliability of the Hubble constant measurement based on the SZ effect. Cooray (1998) showed that projection effects of clusters can lead incidence on the calculations of the Hubble constant and the gas mass fraction, and Hughes \\& Birkinshaw (1998) as well as Sulkanen (1999) pointed out, that galaxy cluster shapes can produce systematic errors on the measured value of $H_o$. \\\\ The aim of this contribution is to investigate the influence of extension, shape and temperature profile of the cluster gas distribution on the inferred value of $H_o$. In Section 2 we recall the calculations for the determination of the Hubble constant for a non-spherical geometry and finite extension of the galaxy cluster, and in Section 3 we present a quantitative discussion on the incidence of these effects on the value of the Hubble constant. In Section 4 we give a short outlook. ",
        "conclusions": ""
    },
    "0010/astro-ph0010274_arXiv.txt": {
        "abstract": "This  paper employs  direct spectral fitting  of  individual  orbital  data in order to measure  rapid  X-ray iron K$_{\\alpha}$ line and continuum spectral slope variations in Seyfert 1 galaxies with unprecedented  temporal resolution.  Application of this technique to  a long \\xte\\  observation of \\mcg\\ indicates  that the line flux does vary on short ($\\ls$1~d) timescales, but that these variations are {\\it not} correlated with  changes in the continuum flux or slope. These rapid  variations indicate that  the line does  indeed originate close  to the  black hole,  confirming predictions  based on  its very broad  profile.   However,  the  lack  of  a  correlation  with  the continuum presents problems for models in which the line variations are driven by  those in the continuum, modified  only by light-travel time effects.  Instead, it may  be that the line responds according to a physical  process with  a different time  scale, such  as ionization instabilities  in the  disk,  or  perhaps that  the  geometry and physical picture is  more complex than implied by the simplest disk--corona models.  These data  also indicate that  the slope of the  underlying power-law continuum ($\\Gamma$) shows strong variability and is tightly correlated with the continuum flux in the sense  that the spectrum steepens as the source brightens. All  of these results  have been  checked with  extensive simulations, which also indicated that  a spurious correlation between $\\Gamma$ and Compton reflection fraction ($R$)  will result if these quantities are measured from the same spectra.  This casts serious doubts on previous claims of such a $ \\Gamma - R $ correlation. ",
        "introduction": "\\label{intro}  X-ray  observations probe  the  innermost regions  of Active  Galactic Nuclei (AGN) and potentially offer a direct probe the immediate environment of the central,  supermassive black holes thought to power AGN  (e.g., Mushotzky, Done  \\& Pounds  1993).  The  currently favored model  for  the  central  regions  comprises a  hot  ($\\sim  10^{8}  - 10^{9}$~K) optically-thin corona lying  above and below the surface of an optically thick, geometrically thin accretion disk (e.g., Haardt \\& Maraschi 1993).  Ultraviolet photons from the disk are inverse-Compton scattered to X-ray energies  (`up-scattered') as they pass through the corona.  The  X-ray source, in  turn, illuminates the  accretion disk, with  photons  either absorbed  and  thermalized  within  the disk  or `reflected,'  producing a  characteristic reflection  hump  and strong iron  K$_{\\alpha}$ fluorescence  (Lightman  \\& White  1988; George  \\& Fabian 1991).  \\ginga\\ observations showed  these features to be common  to Seyfert 1 nuclei and  provided strong support  for this model (Pounds  \\et 1990; Nandra  \\& Pounds  1994).  However,  the  \\asca\\ mission provided  the   first  spectrally  resolved   iron  K$_{\\alpha}$  line profiles, showing that  many Seyfert 1  galaxies to  have a broad,  redshifted iron line profiles  (Tanaka \\et 1995; Nandra  \\et 1997b), best  explained by a combination  of Doppler and  gravitational redshifts from  the inner  regions of  an accretion disk around  a black hole  (Fabian \\et 1995).  These  provide some of  the best  evidence in  support of the  black hole  model (see reviews by Fabian \\et 2000 and Nandra 2000).  Such a compact region {\\it  should} respond very rapidly to changes in the  illuminating flux  (Stella 1990;  Matt \\&  Perola  1992).  Simple models of X-ray illuminated accretion disks, in which the reprocessing time  scale  is  very  short,  predict  the  line  should  respond  to variations in the  continuum with a delay of  order the light-crossing time ($\\sim 1000 M_{7}$~s at  20$r_{g}$ from a black hole of  mass $10^{7} M_{7}$\\Msun, where $r_{g} = GM/c^{2}$).  The  response of the  line to the continuum could then be used  to deduce the geometry and absolute size scales of the inner accretion disk.  This could in principle provide a more accurate method for estimating the mass of the central black hole and,  when  combined with accurate  line  profiles,  even  reveal its  spin (Reynolds \\et 1999; Reynolds 2000).  In the  case of \\mcg, the  profile is so  broad that most of  the line flux  appears to  originate within  $\\sim 20r_{g}$  of the  black hole (Tanaka  \\et 1995;  Iwasawa  \\et 1996;  Guainazzi  \\et 1999).   Recent \\chandra\\ observations do not show any strong, narrow component to the iron line (Lee \\et 2000b), supporting the  view that, in this object  at least, most of  the line flux originates  close to the black hole.   Attempts to measure  the expected continuum  flux driven variability of  the iron line (Iwasawa  \\et 1996, 1999;  Lee \\et 2000a) have led to  suggestions that the line may  be variable, but also provide indications that  the simple picture  above may not fully  explain the behavior.  The situation  in other Seyfert  1s is even less clear.   For instance, Nandra \\et  (1999) found  that in NGC~3516,  another Seyfert~1  with a broad iron line,  the core of the line  profile (6.0--6.4~keV) appeared to track the  continuum  but  the  broad   wings  of  the  line  varied  in  an uncorrelated fashion.   In a long (30~d)  observation, NGC~7469 showed marginal evidence for iron line variability correlated with X-ray flux (Nandra \\et 2000).  Complex variability  has also been reported in the iron lines of NGC~4051 (Wang \\et 1999) and NGC~7314 (Yaqoob \\et 1996). Weaver,  Gelbord  \\&  Yaqoob   (2000) suggest  that  iron  line variability is a common property in Seyfert 1s based on an analysis of archival \\asca\\ data. On the  other  hand,  Chiang  \\et  (2000)  found  no  evidence  for  line variability  in a  simultaneous \\xte/\\asca\\  observation  of NGC~5548. However, recent \\chandra\\ observations  of NGC~5548 revealed a strong, narrow  component to  the  iron line,  indicating  that a  significant fraction of the  line flux originates in distant  material (Yaqoob \\et 2000).  A  constant, narrow  line is indeed  predicted  by AGN unification schemes  (Krolik, Madau \\& \\.{Z}ycki 1994; Ghisellini, Haardt \\& Matt 1994).  In order to determine if  Seyfert~1 iron lines do indeed vary rapidly, we have  designed a new  analytical technique specifically  to measure rapid changes in the X-ray  spectrum by direct spectral fitting to the individual orbital data.  This has then been applied to \\mcg, an ideal subject  for this  approach because  it is  a bright,  highly variable Seyfert 1 with an unambiguously broad iron line and a very long (8~d), quasi-continuous \\xte\\ observation.  The rest  of the paper is  organized as follows.  In  the next section the  observations and  data reduction  procedures are  described.  The spectral analysis is discussed in \\S~\\ref{analysis} and the results of this   analysis   are    tested   using   extensive   simulations   in \\S~\\ref{simulations}.   The   implications  of  the   results  of  are discussed in  \\S~\\ref{discussion}, followed by a brief  summary of the main results in \\S~\\ref{conclusions}. ",
        "conclusions": "\\label{conclusions}  This  paper  presents  a  new  method  of  analyzing  Seyfert~1  X-ray variability  data, designed  to measure  rapid spectral  variations by fitting data from individual orbits.  This was applied to a long (8~d) \\xte\\ observation  of the bright,  strongly variable Seyfert  1 galaxy \\mcg,  and the  conclusions  were tested  with extensive  simulations. This provided  the clearest evidence to  date that the  iron line flux ($F_{K\\alpha}$) of  a Seyfert~1 varies on  relatively short ($\\ls$1~d) timescales.   However, the variations were {\\it  not} correlated with or simply  related to the continuum  variations.  As the  line is very broad, indicating that  the bulk of the flux is  produced close to the black hole,  the lack of correlation contradicts  the simplest picture in which the line variations  are driven by variations in a point-like continuum  source, modulated by  light-travel effects.   Although many modifications to  this standard  picture are possible,  one reasonable solution is  that the  line  variations  are dominated   by   some   other   process,  for   instance,   ionization instabilities, in which case the dominant timescale would be that for the  system to  return to equilibrium.   Another  is that there is something wrong with the simple disk-corona picture described above.   In  any event  this  result  indicates  that the  physics  of reprocessing  in the central  engine is  more complex  than previously thought.  The slope  of the underlying  X-ray power-law ($\\Gamma$) shows strong variability  and is  tightly correlated  with the  source flux  in the sense that the  power-law steepens as the source  gets brighter.  This behaviour  indicates that the  properties of  the  corona are rapidly variable and, if Compton-cooling is the dominant process in determining the X-ray spectral slope,  may be driven  by variations in the  unobserved EUV  band. The limited band-pass  of these data  and intrinsic bias  in the  simultaneous measurement  of $\\Gamma$  and  $R$ appear to conspire to produce a spurious  correlation between these two quantities.  This strongly supports the conclusions  of previous works (e.g., Nandra \\et 2000) that  indicates that  the correlation  claimed in  Zdziarski \\et (1999) may be an artifact of the analysis.  The unexpected result  concerning the iron line variability  hint at much more complex  behavior than  expected in Seyfert  1s.  \\xmm,  with its unprecedented combination of  high throughput and spectral resolution, is ideally  suited to study  rapid iron line variability.   We eagerly await the  results of  the long observations  of this and other Seyfert 1s  that will soon be made by \\xmm."
    },
    "0010/astro-ph0010568_arXiv.txt": {
        "abstract": "A sample of J--type carbon stars was searched for OH maser emission. The new detection of three OH lines towards two silicate carbon stars is reported. In V778\\,Cyg, previously known as the main--lines (1665 and 1667\\,MHz) maser source, the satellite 1612\\,MHz emission was discovered while in NSV\\,2814 the main OH lines were detected. The presence of OH maser lines confirms the former suggestion that oxygen--rich material is located in the vicinity ($\\approx$\\,$10^{15-16}$\\,cm) of silicate carbon stars. ",
        "introduction": "\\label{intr}  The silicate emission features at about 10 and 18\\,$\\mu$m are characteristic for oxygen--rich dust envelopes. However, these features were also discovered in some optically classified carbon stars (Little-Marenin\\,\\cite{littlemarenin86}; Willems \\& de Jong\\,\\cite{willems86}), named later silicate carbon stars. Furthermore, many silicate carbon stars are recognized as $^{13}$\\,C-rich (J-type) carbon stars (Lambert et al.\\,\\cite{lambert90}; Lloyd Evans\\,\\cite{lloydevans90} and references therein). The most recent compilation of 22 known and suspected silicate carbon stars is presented by Chen et al.\\,(\\cite{chen99}). Note also the discovery of the first extragalactic silicate carbon star (IRAS\\,04496$-$6958) in the Large Magellanic Cloud (Trams et al.\\,\\cite{trams99}). The detection of silicate emission from some (silicate) carbon stars suggests that their relatively close surroundings contain oxygen--based dust, in spite of their photospheric chemical composition which shows C/O\\,$>$\\,1. An additional argument for the existence of oxygen--rich material in the vicinity of these stars comes from the detection of water and hydroxyl maser lines towards some of them (see Little-Marenin et al.\\,\\cite{littlemarenin94} and Engels\\,\\cite{engels94} and references therein). To date H$_2$O masers were found in four (J--type) silicate carbon stars and only one of them (V778\\,Cyg) exibits OH maser emission in the main--lines. In addition 1612\\,MHz maser emission was detected in FJF\\,270 (te Lintel Hekkert\\,\\cite{telintelhekkert91}). Here we report the results of a high sensitivity search for OH emission towards 14 J--type carbon stars including 7 known silicate carbon stars. ",
        "conclusions": "Our program stars are listed in Table\\,1 with the designation from the General Catalogue of Cool Galactic Carbon Stars (CCGCS, Stephenson\\,\\cite{stephenson89}) given in the first column. Known silicate carbon stars are marked in boldface. Previously detected H$_2$O and/or OH maser lines are marked by a `+' sign while our new detections are depicted by a `++' sign. OH masers were found in NSV\\,2814 (main--lines) and in V778\\,Cyg (satellite line). Note that the 1612\\,MHz maser line towards V778\\,Cyg has not been detected either by Barnbaum et al.\\,(\\cite{barnbaum91}) nor by Little-Marenin et al.\\,(\\cite{littlemarenin94}).  \\begin{table}[ht] \\caption{List of observed stars.} \\centering \\begin{tabular}{l l c c c c}\\hline CCGCS & other name & OH 1612 & OH 1665 & OH 1667 & H$_2$O \\\\ \\hline {\\bf 1158} & {\\bf NSV 2814}  & $-$ &  ++ &  ++  &  + \\\\ {\\bf 1653} & {\\bf BM Gem} & $-$ & $-$ & $-$ & $-$ \\\\ 1682 & & $-$ & $-$ & $-$ & + \\\\ 1698 & & $-$ & $-$ & $-$ & + \\\\ 1921 & & $-$ & $-$ & $-$ & $-$\\\\ 2028 & & $-$ & $-$ & $-$ & $-$\\\\ 2156 & MT Hya & $-$ & $-$ & $-$ & $-$\\\\ 3066 & HD 100764 & $-$ & $-$ & $-$ & $-$\\\\ {\\bf 3935} & {\\bf FJF 270} & + & $-$ & $-$ & $-$ \\\\ {\\bf 4222} & {\\bf NC 83} & $-$ & $-$ & $-$ & $-$\\\\ {\\bf 4923} & {\\bf V778 Cyg} & ++ &  + &  +  &   +  \\\\ {\\bf 5526} & {\\bf MQ Cyg} & $-$ & $-$ & $-$ & $-$\\\\ 5548 & V1415 Cyg & $-$ & $-$ & $-$ & + \\\\ {\\bf 5848} & {\\bf EU And} & $-$ & $-$ & $-$ & + \\\\ \\hline \\end{tabular} \\end{table} \\begin{figure}[ht] \\vskip3.15in \\hskip-0.2in \\special{psfile=figure1.ps hscale=37  vscale=37} \\special{psfile=figure2.ps voffset=66 hoffset=175 hscale=37  vscale=37} \\vskip-1.2in \\narrowcaption{OH maser spectra of V778\\,Cyg (left) and NSV\\,2814 (right). Transition frequencies in MHz are shown.} \\end{figure}  The spectra are shown in Fig.\\,1 where, for the sake of comparison, re--observed OH main--lines from V778\\,Cyg are presented. In V778\\,Cyg the 1667\\,MHz feature at $-$16.0\\,km\\,s$^{-1}$ has a peak flux density of 0.30\\,Jy, which is a factor of 1.5 higher than the 1665\\,MHz peak of 0.19\\,Jy at $-$16.9\\,km\\,s$^{-1}$. Within the measurement errors the intensities and radial velocities of these lines are the same as observed 12 years ago (Little--Marenin et al.\\,\\cite{littlemarenin88}). The 1612\\,MHz feature of 0.12\\,Jy peaks at $-$14.4\\,km\\,s$^{-1}$. This indicates that the 1612\\,MHz emission is slightly redshifted relatively to the OH main--lines but coincides with a $-$14\\,km\\,s$^{-1}$ component of the H$_2$O maser (Engels \\& Leinert\\,\\cite{engelsleinert94}).  In NSV\\,2814 the peak intensities of the 1665 and 1667\\,MHz are 0.24 and 0.27\\,Jy, respectively. The velocities of OH features at $-$23.5 and $-$22.8\\,km\\,s$^{-1}$ are very different from the H$_2$O maser velocity of $-$13\\,km\\,s$^{-1}$ detected about 7 years ago (Engels\\,\\cite{engels94}). Although temporal variations may affect the maser profiles, the difference in velocities of the OH and H$_2$O maser features of about 10\\,km\\,s$^{-1}$ is quite high for an AGB object. This may suggest that each maser is associated with a different object. Little-Marenin et al.\\,(\\cite{littlemarenin94}) noted that CCGCS\\,1158 may be incorrectly associated with the IRAS source. Furthermore, the intensity ratio of 1667/1665\\,MHz transitions of about 1.1 suggests optically thin LTE conditions, so that the masers may arise in the intervening interstellar medium. The case of NSV\\,2814 needs further careful studies.  In spite of almost 15 years of debate, the formation of silicate carbon stars still remains controversial. A stability of the silicate feature (Yamamura et al.\\,\\cite{yamamura00}) and OH main--lines (this work) in V778 Cyg makes the scenario of {\\it fast} transition from M--type AGB stars into carbon stars difficult to reconcile with the lifetime of silicate emission from an expanding detached shell (see e.g. Yamamura et al.\\,\\cite{yamamura00}). A plausible scenario may be some kind of long--lived reservoirs of oxygen--rich material in a binary system (e.g. Lloyd Evans\\,\\cite{lloydevans90} and Jura \\& Kahane\\,\\cite{jura99}) with (possibly) a main sequence secondary (e.g. Lambert et al.\\,\\cite{lambert90} and Yamamura et al.\\,\\cite{yamamura00}). Silicate carbon stars seem to be related to the {\\it normal} J--type carbon stars in the sense that both groups show H$_2$O maser emission (Engels\\,\\cite{engels94}) and that the mechanism responsible for their low $^{12}$C/$^{13}$C ratios might be the same (Ohnaka \\& Tsuji\\,\\cite{ohnaka99})."
    },
    "0010/astro-ph0010618_arXiv.txt": {
        "abstract": "Recently completed H$\\alpha$ surveys of large portions of the sky can be used to create maps of the free-free intensity distribution at high Galactic latitude that are independent of the spectral fits to the CMB data.  This provides an opportunity to test the accuracy of the spectral fitting procedures and to search for other sources of Galactic forground contamination that could be confused spectrally with the free-free, such as spinning dust grains.  The Wisconsin H$\\alpha$ Mapper (WHAM) survey has sampled the sky north of declination $-30\\deg$ at about one degree angular resolution and has revealed that, except for a few isolated regions of enhanced emission, $\\Delta$T$_{ff}$(30 GHz) $\\la 30 \\, \\mu$K at Galactic latitudes near 15$\\deg$, decreasing to $\\Delta$T$_{ff}$(30 GHz) $\\la 4 \\,\\mu$K at latitudes above 50$\\deg$. Also in progress are H$\\alpha$ surveys that sample the sky at higher angular resolution. ",
        "introduction": "Diffuse, low density ionized hydrogen at a temperature near $10^4$ K is wide spread throughout the disk and halo of the Milky Way (e.g., Reynolds 1991).  The thermal bremsstrahlung associated with this gas is expected to be one of the principal sources of foreground contamination to the CMB observations at mm wavelengths (Bennett et al 1992).  However, because Galactic synchrotron emission dominates the spectrum at lower frequencies and thermal emission from dust dominates at higher frequencies, the distribution of the free-free emission over the sky cannot be directly observed.  Fortunately, hydrogen recombination lines, and H$\\alpha$ in particular, provide a way to trace the ionized hydrogen and thus the free-free.  A thorough review of the implications of H$\\alpha$ observations for studies of the CMB has been provided by McCullough et al (1999).  This paper is a brief update with an emphasis on the recently completed Wisconsin H$\\alpha$ Mapper (WHAM) survey. ",
        "conclusions": "Thermal bremsstrahlung from the warm ionized component of the interstellar medium is one of the principal Galactic foregrounds contaminating observations of the cosmic microwave background (CMB).  Near the peak of the CMB, the intensity of the free-free may be comparable to that of the Galactic synchrotron and thermal dust emission; however, because it does not dominate the light from the sky at any wavelength, its angular distribution over the sky cannot be observed directly.  Within the last few years, surveys of diffuse interstellar H$\\alpha$ emission have been undertaken to explore the warm ionized interstellar gas.  At high Galactic latitudes, where interstellar extinction of the optical H$\\alpha$ is small, the relationship between the intensity of the H-recombination line and the free-free continuum at a given frequency is only a function of the electron temperature. Since the temperature of the gas is tightly constrained, these high latitude H$\\alpha$ surveys can provide accurate maps of the Galactic free-free emission, and thus play a potentially important role in testing the accuracy of the spectral fits to CMB foregrounds and exploring the degree to which spinning dust grains contribute to the foreground emission."
    },
    "0010/astro-ph0010104_arXiv.txt": {
        "abstract": "Observational evidence shows that gravitational lensing induces an angular correlation between the distribution of galaxies and much more distant QSOs. We use weak gravitational lensing theory to calculate this angular correlation, updating previous calculations and presenting new results exploring the dependence of the correlation on the large-scale structure. We study the dependence of the predictions on a variety of cosmological models, such as cold dark matter models, mixed dark matter models, and models based on quintessence. We also study the dependence on the assumptions made about the nature of the primordial fluctuation spectrum: adiabatic, isocurvature and power spectra motivated by the cosmic string scenario are investigated. Special attention is paid to the issue of galaxy biasing, which is fully incorporated. We show that different mass power spectra imply distinct predictions for the angular correlation, and therefore the angular correlation provides an extra source of information about cosmological parameters and mechanisms of structure formation. We compare our results with observational data and discuss their potential uses. In particular, it is suggested that the observational determination of the galaxy-QSO correlation may be used to give an independent measurement of the mass power spectrum. ",
        "introduction": "According to general relativity, light from distant objects such as quasars is affected by the large-scale distribution of the intervening masses, and, consequently, these act as gravitational lenses. A gravitational lens enlarges the solid angle of the source and conserves its surface brightness. So at the same time it brings to view faint objects (magnification bias) and dilutes their population density. These two antagonistic effects can generate either correlation or anti-correlation between the angular distribution of background and foreground objects (see e.g. Schneider, Ehlers \\& Falco 1992).  Several studies have explored the correlation between QSO's and galaxies at various wavelengths. For example, Bartelmann \\& Schneider (1994) found a statistically significant correlation between 1Jy sources and IRAS galaxies for quasars at high redshifts. Also, Ben\\'{\\i}tez \\& Mart\\'{\\i}nez-Gonz\\'alez (1997) reported statistically significant correlations between COSMOS/UKST galaxies and PKS radio quasars, and Ben\\'{\\i}tez, Sanz \\& Mart\\'{\\i}nez-Gonz\\'alez (2001) found positive correlations using radio-loud quasars from the 1Jy and Half-Jansky samples. These results were obtained for small angles on the sky (of the order of 10 arcmin). Recently, Williams \\& Irwin (1998) investigated the correlation between QSO's from the LBQS catalog and galaxies from the APM survey and detected a significant positive angular correlation, ranging from scales of 10 arcmin to over a degree. For a more extended review of observational data see Ben\\'{\\i}tez et al. (2001), Bartelmann \\& Schneider (2001), and Norman \\& Willians (2000).  An analytical theory of the galaxy-QSO correlation due to weak gravitational lensing was introduced by Bartelmann (1995), Dolag \\& Bartelmann (1997), hereafter B95 and DB97, and Sanz, Mart\\'{\\i}nez-Gonz\\'alez \\& Ben\\'{\\i}tez (1997), having as base the work by Kaiser (1992). We will review the theory briefly in Section 2. The predicted correlations between galaxies and QSO's agree reasonably well with some observations for angles below 10 arcmin. However, Williams \\& Irwin (1998) claim that for larger angular scales theory and observations differ by one order of magnitude for standard cosmological models such as cold dark matter ($SCDM$) and cold dark matter with a cosmological constant ($\\Lambda CDM$).  Although this conclusion has to be tested with future observations, it is of importance to investigate the predictions of the galaxy-QSO correlation for a large class of cosmologies. The effect of bias, matter content and current expansion rate has to be investigated in detail. This will be the aim of this paper, where we keep the assumption that the weak lensing approximation is valid. Effects of higher order terms are not investigate here. These contributions where estimated recently by Williams (2000), and were shown to increase the amplitude by less than 10 per cent for nearby large-scale coherent structures. Our investigations show that the observations of correlations between QSOs and galaxies on large angular scales can provide independent information relevant to structure formation understanding, and are somewhat sensitive to the values of a set of cosmological parameters.  The paper is organized as follows. In Section 2 we review the theoretical formalism used. In Section 3 we present the cosmological models we consider in our investigation and study their predictions. The models include mixed dark matter and quintessence models, as well as isocurvature and cosmic string based models of structure formation. We analyze in detail the effect of several cosmological parameters on the galaxy-QSO correlation in a flat universe with cosmological constant and initial adiabatic fluctuations. We discuss our findings and some observational data in Section 4. ",
        "conclusions": "\\label{Discussion}  We studied in detail the role of several cosmological parameters and models of structure formation on the prediction of the angular correlation between galaxies and QSOs induced by weak gravitational lensing.  In a log-log plot $\\xi_{GQ}(\\phi)$ can be roughly characterized by an amplitude and a slope. The amplitude is closely related to the normalization of the mass power spectrum, and the slope to its shape. More power on small scales (large $k$) implies a steeper slope. This is a reflection of the fact that the amount of small scale structure is responsible for the galaxy-QSO correlation on small angles and larger structures are responsible for the correlation on larger angles. This simplicity of behavior results in a relatively large degeneracy of $\\xi_{GQ}(\\phi)$ as a function of the cosmological parameters and models analyzed. In brief, a higher value of $\\Omega_m$, $n$, or $h$ has nearly the same effect on $\\xi_{GQ}(\\phi)$ as a lower value of $\\Omega_{\\nu}$, or $\\Omega_b$. This effect is a higher correlation on small angles. Roughly, this represents a steeper slope for $\\xi_{GQ}(\\phi)$, which can be also achieved by shifting $\\omega_Q$ closer to zero, or by introducing an isocurvature or cosmic string component.  The main lesson given by the use of different theoretical mass power spectra is that the galaxy-QSO correlation at different angles contains information about the amount of structure at different scales.  Our calculations fully incorporate galaxy biasing. This requires the use of an empirical galaxy power spectrum. We choose to use the spectrum obtained from the APM survey, but this choice is not unique. Spectra from other surveys are available and could be used as well. These spectra do not all agree (see Einasto et al. 1999), allowing some freedom for the galaxy-QSO correlation prediction, because our $\\xi_{GQ}(\\phi)$ calculation can be seen as an integral over an effective mean spectrum $\\sqrt{P_{gal}(k)P_{\\delta}(k)}$. In a similar way, the use of a power spectrum of cluster of galaxies allows for a prediction of the correlation between galaxy clusters and QSOs. To illustrate this point we use the spectrum obtained from the Abell-ACO clusters of galaxies (Miller \\& Batuski 2001) as $P_{gal}(k)$. We show the effect of this alternative choice at Fig. \\ref{observations}, which gives a higher amplitude for $\\xi_{GQ}(\\phi)$, roughly given by a factor $(\\sigma_8)_{Abell-ACO}/(\\sigma_8)_{APM}\\approx 4$, where $(\\sigma_8)_{Abell-ACO}=3.2$ is the rms fluctuation for the Abell-ACO spectrum, and $(\\sigma_8)_{APM}=0.82$. The correlation curve found using the Abell-ACO spectrum is also slightly steeper than that using APM, because the first spectrum is flatter than the second at large $k$ [$P_{Abell-ACO}(k)\\propto k^{-1.2}$ and $P_{APM}(k)\\propto k^{-1.4}$].  We also plot in Fig. \\ref{observations} some points obtained from observations. A direct comparison among the different data sets is not possible, because they were obtained from different QSO and galaxy populations. For the same reason a strict comparison between these observational points and the theoretical predictions shown is not possible. That would require an individual calculation for each of the data sets, which would take in account the different redshift distributions for QSOs and galaxies, and also the particular power spectrum of the foreground population (galaxies or galaxy groups). This would be necessary because when selection criteria are applied to a foreground object catalog the effective power spectrum of the resulting subset is possibly different from the spectrum of the whole population. A rigorous analysis of the data would also require considerations about observational particularities such as the possible effects of dust and systematics. That is not our aim here. Nevertheless, a qualitative agreement in behavior between data and predictions can be glimpsed.  Another way to analyze the results from the several groups (avoiding the messiness of noise data) is to capture the essence of each data set by adjusting a power law to each one. We find the best fit of $\\xi_{GQ}(\\phi)/(s-1) = A (\\phi/1^{\\prime})^B$ to the data by the least squares method (see Table \\ref{fit-table}). To compare with the predictions shown in Fig. \\ref{observations} we can take the amplitude $A$ as being the value of $\\xi_{GQ}(\\phi)/(s-1)$ at $\\phi=1^{\\prime}$ (for the Abell-ACO result $A\\approx 0.04$ and for APM $A\\approx 0.008$), and power index $B$ as the average bi-log slope of $\\xi_{GQ}(\\phi)/(s-1)$, which gives $B\\approx -0.8$. Note however that for $\\phi {\\raise0.3ex\\hbox{$\\;<$\\kern-0.75em\\raise-1.1ex\\hbox{$\\sim\\;$}}} 1^{\\prime}$ the weak lensing assumption is not satisfied (strong lensing becomes important), so that the predictions for $\\xi_{GQ}(\\phi)$ are underestimated in this range.  Ben\\'{\\i}tez \\& Mart\\'{\\i}nez-Gonz\\'alez (1997) (BMG97) found the galaxy-QSO correlation for angles up to 15 arcmin using radio loud QSOs from the PKS catalog and also using optically selected QSOs from the LBQS catalog. For this last group there is a large dispersion of data and points with negative correlation on small angles were found; the overall result is a null correlation (dust and selection bias were mentioned as possible explanations). The large error bars obtained for the two quasar samples make these observations compatible with predictions. Ben\\'{\\i}tez et al. (2001) (BSMG00) obtained a somewhat higher correlation on the same anglular scales using quasars from the Half-Jansk and 1Jy samples.  Williams \\& Irwin (1998) (WI98) also used a subsample of the LBQS catalog to determine the galaxy-QSO correlation, but for larger angles (7 to 300 arcmin). For angles up to nearly 100 arcmin they find a much higher correlation than the predicted one, and for angles larger than that the correlation is dispersed around zero, being compatible with predictions. Croom \\& Shanks (1999) (CS99) reinterpreted the anti-correlation between faint QSOs and galaxy groups found by Boyle, Fong \\& Shanks (1988) as being due to gravitational lensing. A flat QSO number count ($s=0.78$) was used to explain the negative correlation, but a surprisingly high amplitude is obtained when $\\xi_{GQ}(\\phi)/(s-1)$ is calculated. This high amplitude would be in concordance with a higher normalization of the power spectrum of the galaxy groups. Other works also seek to find the observed galaxy-QSO correlation, but the general conclusion that one can reach is that up to now $\\xi_{GQ}(\\phi)$ is an observationally not very well determined quantity.  The use of much larger catalogs (e.g. the Sloan Digital Sky Survey with $\\approx 10^5$ QSOs, and the 2dF survey) may allow a more precise and accurate measurement of $\\xi_{GQ}(\\phi)$ which in turn may constitute an independent way for the determination of the mass power spectrum, and a valuable tool for investigating structure formation and large-scale structure."
    },
    "0010/astro-ph0010196_arXiv.txt": {
        "abstract": "We have collected all of the published photometry for \\object{GRB~990123} and \\object{GRB~990510}, the first two gamma-ray bursts where breaks were seen in the light curves of their optical afterglows, and determined the shapes of their light curves and the break times.  These parameters were used to investigate the physical mechanisms responsible for the breaks and the nature of the ambient medium that the bursts occurred in.  The light curve for \\object{GRB~990123} is best fit by a broken power law with a break $1.68 \\pm 0.19$ days after the burst, a slope of $\\alpha_1 = -1.12 \\pm 0.08$ before the break, and a slope of $\\alpha_2 = -1.69 \\pm 0.06$ after the break.  This is consistent with a collimated outflow with a fixed opening angle of $\\theta_0 \\approx 5\\degr$.  In this case the break in the light curve is due to the relativistic fireball slowing to $\\Gamma \\approx 1/\\theta_0$.  The light curve for \\object{GRB~990510} is best fit by a continuous function with an early-time slope of $\\alpha_1 = -0.54 \\pm 0.14$, a late-time slope of $\\alpha_2 = -1.98 \\pm 0.19$, and a slow transition between the two regimes approximately one day after the burst.  This is consistent with a collimated outflow with $\\theta_0 \\approx 5\\degr$ that is initially radiative, but undergoes a sideways expansion that begins approximately one day after the burst.  This sideways expansion is responsible for the slow break in the light curve. ",
        "introduction": "}  In the first few months of 1999 bright optical afterglows (OAs) were observed for two gamma-ray bursts (GRBs): \\object{GRB~990123} and \\object{GRB~990510}.  Extensive follow-up observations were made at optical wavelengths for both GRBs, so it was possible to follow their light curves down to $V \\approx 28$, with the deepest photometry being done with the \\emph{Hubble Space Telescope} (\\emph{HST}).  These data, when put onto a common photometric system, can be used to constrain the shape of each burst's light curve.  The resulting shape parameters (the time of the break, and the slopes before and after the break) can be used to constrain physical models of the OA, and the interaction of the relativistic ejecta from the initial blast with the surrounding medium.  At 9:46:56.12 UT on 1999 January 23 the Burst and Transient Source Experiment (BATSE) detectors on the \\emph{Compton Gamma Ray Observatory} satellite detected \\object{GRB~990123} in the constellation Bo{\\\"o}tes.  $X$-ray observations from the Dutch--Italian satellite \\emph{BeppoSAX} localized the position of the burst to within $5\\arcmin$ (Piro~\\cite{P99}) and the Robotic Optical Transient Search Experiment (Akerlof \\& McKay~\\cite{AM99}; Akerlof et~al.~\\cite{ABB1999}) detected an OA that reached a peak magnitude of $V = 8.86 \\pm 0.02$ just 47 seconds after the BATSE trigger.  Further optical observations constrained the redshift to between $1.60 \\le z < 2.04$ (Andersen et~al.~\\cite{AC99}; Kulkarni et~al.~\\cite{KD99}), and images taken with the \\emph{HST} Space Telescope Imaging Spectrograph (STIS) revealed a host galaxy with $V_0 = 24.25$ (Holland \\& Hjorth~\\cite{HH99}).  The burst had a peak fluence of $(5.09 \\pm 0.02) \\times 10^{-4}$ erg cm$^{-2}$ (Kippen~\\cite{K99a}), which places it in the top 0.3\\% of the BATSE fluence distribution.  This corresponds to a total energy release, assuming isotropy and depending on the details of the adopted cosmology, of between $\\approx 3 \\times 10^{54}$ and $\\approx 4.5 \\times 10^{54}$ erg, which is approximately the rest-mass energy of two neutron stars.  This amount of energy is difficult to explain using merging compact objects since it would require that the entire rest mass of both objects be converted to gamma radiation with an efficiency of nearly 100\\%.  \\object{GRB~990510} (the Anja burst) was detected by both the \\emph{Compton Gamma Ray Observatory} and \\emph{BeppoSAX} in the constellation Octans at 8:49:05.95 UT on 1999 May 10.  Optical follow-up observations 3.5 hours after the burst revealed an OA with $R \\approx 17.5$ (Axelrod et~al.~\\cite{AMS99}).  Vreeswijk et~al.~(\\cite{VG1999}) found a lower limit for the redshift of $z = 1.619$.  A faint galaxy ($V = 28.5 \\pm 0.5$) has been detected $0\\farcs066 \\pm 0\\farcs009$ from the GRB (Bloom~\\cite{B2000}; Fruchter et~al~\\cite{FHP2000}).  The burst had a peak fluence of $(2.56 \\pm 0.09) \\times 10^{-5}$ erg cm$^{-2}$ (Kippen~\\cite{K99b}), which places it in the top 9\\% of the BATSE fluence distribution.  This corresponds to a total isotropic energy release, depending on the details of the cosmology, of at least $10^{53}$ erg, approximately one order of magnitude less than the energy released in \\object{GRB~990123}.  The light curve for \\object{GRB~990123} became steeper approximately two days after the burst (Castro-Tirado et~al.~\\cite{CT99}, Kulkarni et~al.~\\cite{KD99}), while Harrison et~al.~(\\cite{HB99}), Israel et~al.~(\\cite{IMC1999}), and Stanek et~al.~(\\cite{SGK1999}) found that the light curve of \\object{GRB~990510} became significantly steeper approximately 1.5 days after the burst.  These breaks have been interpreted as evidence that the outflows from the bursts are collimated with opening angles of approximately $5\\degr$--$10\\degr$ (Sari et~al.~\\cite{SP99}, Castro-Tirado et~al.~\\cite{CT99}).  Such models provide a solution to the so-called ``energy crisis'' of GRBs since, if GRBs are collimated outflows, the total energy requirement drops by factor of between approximately 100 and 1000.  This makes the energetics of GRBs consistent with the energetics of supernovae and merging compact objects.  Before \\object{GRB~990123} GRB light curves were usually fit by power laws of the form $f_{\\nu}(t) = k t^{\\alpha}$ where $k$ is a normalization constant and $\\alpha$ is between approximately $-1$ and $-2$.  The steepest temporal decay was $\\alpha = -2.5$ for the OA associated with \\object{GRB~971227} (Djorgovski et~al.~\\cite{D98}). The first evidence that the decay of the optical light curve from an OA varies with time came from \\object{GRB~990123}, where $\\alpha$ rapidly steepened from $-1.10 \\pm 0.03$ to $-1.65 \\pm 0.06$ with the break occurring $2.04 \\pm 0.46$ days after the burst (Kulkarni et~al.~\\cite{KD99}).  Harrison et~al.~(\\cite{HB99}), Israel et~al.~(\\cite{IMC1999}), and Stanek et~al.~(\\cite{SGK1999}) found that the optical light curve for \\object{GRB~9900510} exhibited a similar behavior, but with a slow transition between the two regimes. Harrison et~al.~(\\cite{HB99}) used a four-parameter fitting function to find slopes of $-0.82 \\pm 0.02$ before the break and $-2.18 \\pm 0.05$ after the break with the break occurring $1.20 \\pm 0.08$ days after the burst.  However, Stanek et~al.~(\\cite{SGK1999}) found slopes of $-0.76 \\pm 0.01$ before and $-2.40 \\pm 0.02$ the break with the break occurring $1.57 \\pm 0.03$ days after the burst; and Israel et~al.~(\\cite{IMC1999}) found $-0.88 \\pm 0.03$ before and $-2.68 \\pm 0.13$ after with a break at $1.8 \\pm 0.2$.  The latter two groups used a different four-parameter fitting function from Harrison et~al.~(\\cite{HB99}), and the fits formally disagree with each other.  In this paper we collect the published photometry for the OAs associated with \\object{GRB~990123} and \\object{GRB~990510}, examine three fitting functions for broken light curves, and use these fits to constrain physical models for the burst.  We investigate the effects of using different fitting functions on determining the time of the break and the slope of the light curve before and after the break. Our goal is to determine reliable light curve parameters for each GRB and to use these parameters to distinguish between different physical models for the nature of each OA\\@.  Sect.~\\ref{SECTION:fitting} discusses the three fitting functions that we used to determine the break time, and the slopes of each light curve. Sect.~\\ref{SECTION:grb990123_data} presents our fits to the light curve of \\object{GRB~990123}'s OA and an estimate of the slope of the optical spectrum of the OA\\@.  Sect.~\\ref{SECTION:grb990510_data} presents our fits to the light curve of \\object{GRB~990510}'s OA and an estimate of the slope of its optical spectrum.  We discuss the effects of the choice of the fitting function on the determination of the break time and slopes of the light curves for each burst.  In Sect.~\\ref{SECTION:models} we discuss various models for the decay of the light curves, and in Sect.~\\ref{SECTION:interpertation} we derive some physical properties for the bursts based on our fits to the light curves.  We have written these sections in such a way that they can be used with data from other GRBs.  We have adopted a cosmology with a Hubble parameter of $H_0=65$ km s$^{-1}$ Mpc$^{-1}$, and assumed a density parameter of $\\Omega_0=0.2$ and no cosmological constant ($\\Omega_\\Lambda=0$) throughout this paper. ",
        "conclusions": "}   It is possible to learn a great deal about the nature of a GRB, and the environment where the burst occurred, from the optical light curve of the burst's OA\\@.  However, it is important that the decaying light curve be well-sampled in order to accurately determine the time and shape of the break, as well as the slope of the light curve before and after the break.  \\object{GRB~990123} and \\object{GRB~990510} are the first two GRBs for which breaks have been observed in their optical light curves.  We have collected all of the published photometry for these two bursts and determined the time of the break and the slopes of these light curves before and after the break using three different fitting functions.  We also estimated the spectral indices of each OA from near-simultaneous broad-band photometry.  These parameters were used to constrain models for the nature of each burst.  Our results suggest the following.  \\begin{enumerate}  \\item Both \\object{GRB~990123} and \\object{GRB~990510} exhibit breaks in their optical light curves at times between approximately one and two days after the bursts.  There is no evidence that the time of the break depends on frequency.  The break in \\object{GRB~990123}'s optical light curve appears to have occurred very fast, perhaps over a period of less than five hours.  This is in sharp contrast to the break in the optical light curve for \\object{GRB~990510}, which appears to have occurred slowly over approximately one day.  \\item Unlike \\object{GRB~000301C} (Jensen et~al.~\\cite{JFG2000}) the parameters of the light curve ($\\alpha_1$, $\\alpha_2$, and $t_b$) for \\object{GRB~990123} and \\object{GRB~990510} are not strongly dependent on the fitting function.  This means that our interpretations are not strongly dependent on the type of fitting function used to derive the slopes and break time.  However, the choice of fitting function can influence the interpretation of the light curve for some GRBs.  \\item Our favoured interpretation for the optical light curve of \\object{GRB~990123} is that the afterglow was collimated with a fixed opening angle of $\\theta_0 \\approx 5\\degr$.  Therefore, the observed break in the light curve is a geometric effect that occurs when the relativistic expansion of the fireball slows to $\\Gamma \\approx 1 / \\theta_0$.  This model reduces the energy released by the GRB by a factor of $\\approx 300$ to $E_{52} = 1.1$, which corresponds to the conversion of $\\approx 0.01$ Solar masses of material to gamma radiation.  The magnetic field strength is $\\epsilon_B \\approx 2.6 \\times 10^{-7}$ if the local number density is $\\approx 1$ cm$^{-3}$ and $\\epsilon_B \\approx 5.6 \\times 10^{-5}$ if the local number density is $\\approx 1000$ cm$^{-3}$.  This is well below the equipartition energy.  \\item The OA's light curve for \\object{GRB~990510} is consistent with a collimated outflow that is initially radiative. The transition to an adiabatic phase, and the onset of a sideways expansion of the jet, occurred near the time of the break. The opening angle was $\\theta_0 \\approx 5\\degr$ at that time, the same as \\object{GRB~990123}.  We are unable to estimate reliably the magnetic field strength because no measurements are available of the flux at frequencies above $\\nu_c$, the electron cooling frequency.  \\end{enumerate}  We have investigated several models for \\object{GRB~990123} and \\object{GRB~990510} and find that the OAs for these two bursts can not be explained with a single model.  This suggests that the local environment where a GRB occurs plays an important role in determining the evolution of the fireball, and the observed light curve of the OA\\@.  We note, however, that the uncertainties in determining the break time and slopes of the light curves make it difficult to unambiguously determine the physics of the optical afterglow and its environment."
    },
    "0010/astro-ph0010475_arXiv.txt": {
        "abstract": "We review the topic of rotation curves of spiral galaxies emphasizing the standard interpretation as evidence for the existence of dark matter halos. Galaxies other than spirals and late-type dwarfs may also possess great amounts of dark matter, and therefore ellipticals, dwarf spirals, lenticulars and polar ring galaxies are also considered. Furthermore, other methods for determining galactic dark matter, such as those provided by binaries, satellites or globular clusters, have to be included. Cold dark matter hierarchical models constitute the standard way to explain rotation curves, and thus the problem becomes just one aspect of a more general theory explaining structure and galaxy formation. Alternative theories also are included. In the magnetic model, rotation curves could also be a particular aspect of the whole history of cosmic magnetism during different epochs of the Universe. Modifications of Newtonian Dynamics provide another interesting possibility which is discussed here. ",
        "introduction": "There is a wide consensus that the rotation curves of spiral galaxies constitute an observational proof -perhaps the best proof- for the existence of dark matter in the Universe. Dark matter is of evident interest in Cosmology, hence the interest of a review on the topic, in this case from a post-graduate didactic point of view.  Dark matter is not an exotic or sophisticated hypothesis. Neutrinos, brown dwarfs and black holes are all candidates to be identified with dark matter and are, nonetheless, classical concepts in Physics, introduced by fully established theories. High Energy Physics actually predicts a number of particles that do not interact with photons. We cannot claim that all the existing matter emits or absorbs photons.  However, although the necessity of dark matter was proposed more than 60 years ago (Zwicky, 1937), this hypothesis is still not whole-heartedly accepted by some workers. This scepticism is never explicitly expressed, but is subjacent, implicitly revealed in sentences such as \"a galaxy {\\it{has}} a halo\". We should rather say  \"a galaxy {\\it{is}} a halo\" as a galaxy's mass may be at least 10 times its visible mass, the visible element then being a mere minor component, only important for us because it is what we see.  (We must, however, accept that when we state that if a galaxy {\\it{is}} a halo, the discovery of just one exception, i.e. of a visible galaxy with no halo, would represent a serious problem of interpretation, a problem as fundamental as  finding the light of a firefly not associated with a firefly.)  Once the  existence of dark matter is recognized as a conservative possibility, let us establish a difference between the problem of dark matter in the Universe and the problem of dark matter in galaxies. We implicitly assume throughout the paper that $\\Omega \\sim 1$, and that visible matter only contributes with $\\Omega_{V} \\sim 0.003$. The total matter contributes either with $\\Omega_M \\sim 1$, as classically assumed, or with $\\Omega_M \\sim 0.3$, coherent with the more recently assumed scenario deduced from the observations of early supernovae, leading to the re-accelerating Universe, the non-vanishing cosmological constant or other identifications of dark energy (e.g. Turner, 1999). The assumption of a plane Universe, $\\Omega \\sim 1$, is based on theoretical ideas about inflation, observations of large-scale dynamics, the interpretation of the CMB and even on some philosophical arguments. Our position on dark matter is based on the difference between $\\Omega_M \\sim 0.3$ and $\\Omega_{V} \\sim 0.003$, as observed. We only see, in the best of cases, 1\\% of the matter. This is the basic fact, rather than the existence and possible observation of dark matter in galaxies, that demonstrates that dark matter exists.  Rotation curves and other dynamical effects in galaxies suggest that total galactic mass is about 10 times larger than that observed. Even in this case, we should still find about 90\\% of dark matter elsewhere in the Universe, which means, given the relative uncertainty in all these figures, that practically all the required dark matter lies outside galaxies. The possible galactic contribution to dark matter is negligible to close the Universe. Therefore, the existence of galactic dark matter is clearly very important for our knowledge of galaxies and their dynamics, but not so decisive for the cosmological problem of identifying its nature and amount.  This notion allows us to make a more objective analysis of the topic of dark matter in galaxies, once it has been partially disconnected from the cosmological one. Rotation curves of spirals, and many other observations, are currently interpreted as evidence of the existence of massive large dark halos, but a critical analysis of the observations, and the theoretical interpretations involved, is permanently necessary. Even if dark matter is a major ingredient in the most widely accepted and orthodox picture of a galaxy, most authors only differing about its mass and size, we will see that the hypothesis of a total absence of galactic dark matter cannot be completely ruled out.  The problems involved in  determining galactic dark matter may be summarized as follows: the internal regions of galaxies require little or no dark matter and we must examine the external ones, where there are few stars and we must observe gas to determine the gravitational potential. However gas, in these regions where the gravitational force is low, may be influenced by magnetic fields. Thus, it is convenient to look for stellar systems lying far from the galaxy, and particularly satellite or companion galaxies. Then, however, it is very difficult to distinguish between galaxies with a halo and halo with galaxies, i.e. the hypothesis of a large common halo is difficult to reject.  Though the rotation curves of spirals is the main topic to be discussed, we must pay attention to other methods of determining DM, in particular those based on globular clusters, satellite galaxies, binary systems, polar-ring galaxies and so on. Lensing by spiral galaxies could become an important tool in the near future (Maller et al. 1999). We should also review the methods of DM estimation in other types of galaxies. With respect to the DM problem, spirals do not seem to be exceptional.  Other models, which do not require DM to explain rotation curves, have been reported in the literature and also require our attention in this study. One such model is MOND (Modified Newtonian Dynamics) and another is the magnetic model. The authors have contributed to the development of this latter model, which is commented in some detail at the end, together with its cosmological implications. Despite the inclusion of MOND and the magnetic approach as interesting possibilities, through out this paper we adhere to the most conservative point of view based on DM. Nevertheless, we emphasize the difficulties inherent to most methods and models.  The purpose of this review is mainly didactic, as a bibliographic source for postgraduate courses, but also critical. This critical approach might be considered unnecessary, in view of the wide acceptance of the dark matter hypothesis by the scientific community; but apart from its didactic interest, it is always pertinent to reconsider apparently solid beliefs. In this respect, it is convenient to remind the reader of four historical aspects related to the early adoption of the DM halo hypothesis, which were based on arguments that eventually became open to discussion or were made obsolete:   a) Kahn and Woltjer (1959) considered the dynamics of the double system formed by M31 and the Milky Way and concluded that their motion would require a binary system mass of $\\ge 1.8 \\times 10^{12} M_\\odot$, much greater than the visible matter in the two galaxies. Kahn and Woltjer formed the opinion that there existed an as yet unobserved mass in some invisible form. Although they identified this invisible mass with hot gas rather than dark matter in its present sense (unobserved rather than unobservable) this work gave a first proof of the missing mass. It should be remembered, however, that this mass was considered to lie either in M31 and the Milky Way or in the intergalactic space (either two halos or a common halo). It is still not clear whether this second possibility can be completely ruled out.  Therefore, this paper established the existence of dark matter, but not necessarily within galaxies. The doubt remained: either in or in between.  b) Oort (1960, 1965) found evidence for the presence of dark matter in the disk of our galaxy, although van dem Bergh (1999) wrote: ``However, late in his life, Jan Oort told me that the existence of missing mass in the galactic plane was never one of his most firmly held scientific beliefs''. After a long debate since then (see Binney and Tremaine, 1987; Ashman 1992) the discussion finally seems to be closed. Using HIPPARCOS data, Crez\\'e et al. (1998) have found that there is no evidence for dark matter in the disk: gas and stars perfectly account for the gravitational potential.  c) Ostriker and Peebles (1973) suggested that spherical dark matter halos around the visible component of the spiral galaxies were necessary to suppress bar instabilities. However, their arguments did not convince Kalnajs (1983) and Sellwood (1985), who showed that a central bulge was equally efficient to stabilize disks.  d) Babcock (1939) observed that the stars in M31 were rotating at an unexpectedly high velocity, indicating a high outer mass-to-light ratio, although he also considered  other possibilities: either strong dust absorption or, as he stated, ``new dynamical considerations are required, which will permit of a smaller relative mass in the outer parts''. This sentence, quoted by van dem Bergh (1999), is interesting, bearing in mind that gravitation is the only force considered at present to explain rotation curves, while other ``dynamical considerations'' are ignored. Optical rotation curves of other galaxies were obtained, until those published by Rubin et al. (1980), that were considered to be clear evidence of dark matter in galaxies. Today, however, many authors consider that optical rotation curves can be explained without dark matter (even without rejecting its contribution). For instance, Broeils and Courteau (1997) by means of r-band photometry and $H_\\alpha$ rotation curves for a sample of 290 spirals concluded that ``no dark halo is needed''.  Therefore, the basic initial arguments leading to the belief of dark matter halos around spiral galaxies failed or were not conclusive. Only the interpretation of the HI rotation curves by Bosma (1978) subsists unmodified. For early thoughts on dark matter, the paper by van dem Bergh (1999) and the Ph.D.  Thesis of Broeils (1992) are essential reading.  In brief, there is a standard interpretation of the rotation curve of spiral galaxies that is implicitly adopted throughout this paper, but we should not ignore other possibilities. ",
        "conclusions": ""
    },
    "0010/astro-ph0010643_arXiv.txt": {
        "abstract": "The similarity between observed velocity structures of Al III and singly ionized species in damped Ly$\\alpha$ systems (DLAs) suggests the presence of ionized gas in the regions where most metal absorption lines are formed. A simplified model consisting of {\\em Region 1)} a plane-parallel ionization bounded region illuminated by an internal radiation field, and {\\em Region 2)} a neutral region with a negligible metal content is considered. We calculate photoionization equilibrium models for region 1, and constrain the ionization parameter by the observed $N$(Al III)/$N$(Si II) column density ratio. Under these conditions we find that ionization effects are important.  If these effects are taken into account, the element abundance ratios in DLAs are quite consistent with those observed in Milky Way stars and in H II regions of local low-metallicity blue compact dwarf galaxies. In particular we cannot exclude the same primary N origin in both DLAs and metal-poor galaxies. No depletion of heavy elements on dust grains needs to be invoked, although our models do not exclude the presence of little depletion.  Although highly simplified and relying on the strong assumption of a significantly lower metal content in region 2, our model appears to be supported by recent data on a local DLA and it is not in contradiction with the current knowledge on high redshift DLAs. If correct, it offers a clear simplification in the understanding of heavy element abundance ratios in DLAs and their comparison with the local Universe. ",
        "introduction": "The above discussion shows that the properties of the damped Ly$\\alpha$ systems are yet poorly understood. While a significant amount of neutral gas is observed in these systems, it is not clear how the regions of ionized gas are related to those of neutral gas. What is the main source of ionization in DLAs - internal stellar radiation or harder background intergalactic radiation? Where are absorption lines of heavy elements formed - in ionized or neutral gas? Is the metallicity of ionized and neutral gas similar or not? Is heavy element depletion by dust grains important?  In the present paper we explore the following ``multi-component'' picture for DLA systems: {\\em 1)} an ionization bounded region illuminated by an internal radiation field complemented by {\\em 2)} a neutral region with a lower metal content. These working hypotheses are adopted to study the main effects intrinsic stellar ionizing radiation may have on observed heavy element patterns in such conditions.   To explore the consequences of such a models we use the CLOUDY ionization equilibrium code (version c90.05; Ferland 1996; Ferland et al.\\ 1998) to calculate the ionization state of region 1. For simplicity, we assume internal stellar radiation and consider a plane-parallel geometry. The outer boundary of the slab is determined by the condition that a low electron temperature of 2000 K is reached. For the cases discussed here (low metallicity and low dust content) this boundary condition is a good approximation for the outer boundary of ionization-bounded H II regions. For the ionizing spectrum, we adopt the spectrum of a typical O star. We use an ATLAS line-blanketed model atmosphere (Kurucz 1991) with effective temperatures $T_{\\rm eff}$ = 33000 K, 40000 K and 45000K and log $g$ = 4.0. The lower temperature is likely more appropriate for the conditions in the interstellar medium of metal-rich spiral galaxies and the local ISM in our Galaxy (cf.\\ Sembach et al.\\ 2000). In low-metallicity environments the temperature of the ionizing radiation tend to be higher as evidenced from studies of nearby extragalactic giant H II regions (e.g.\\ Stasi\\'nska \\& Leitherer 1996). The ionization parameter which is proportional to the ratio of the number density of the ionizing photons to that of the electrons is varied in the range $U$ = 10$^{-5}$ -- 10$^{-2}$. We consider the metal abundance of 0.1 solar, with relative element abundances equivalent to those observed in the low-metallicity BCDs (e.g., Izotov \\& Thuan 1999). In particular, C/O and N/O abundance ratios are reduced by factors of 2 and 5 respectively as compared to solar abundance ratios. We also adopt for $^4$He the mass fraction $Y$ = 0.252 (or log He/H = --1.07), which is higher than the primordial mass fraction $Y_{\\rm p}$ = 0.245 (Izotov et al.\\ 1999), but is typical for the BCDs with the heavy element abundance 1/10 solar (Izotov \\& Thuan\\ 1998). Dust grains with a dust-to-gas ratio of 1/10 that in Milky Way are taken into account in heating and cooling processes, although the differences in ionic fractions derived in the models with and without dust grains are small.  A similar model has recently been considered by Howk \\& Sembach (1999). The major difference with their approach is that they consider ionized density-bounded regions surrounding the neutral, H I--bearing clouds. The boundary of the ionized region is defined where the local fraction of neutral hydrogen climbs above 10\\%. With these assumptions the fraction of neutral species like O$^0$, N$^0$ or Ar$^0$ is significantly reduced because they always reside near the edges of H II region. Additionally, the variation of the local ionization fraction of neutral hydrogen at the edge of ionized density-bounded region results in large variations of some low-ionization species, in particular, of Cr$^+$ and Fe$^+$. Further resulting differences between the model by Howk \\& Sembach (1999) and our ionization-bounded model will be discussed below.  An important feature of both models is that the heavy element abundances in the ionized and neutral regions are a priori unrelated. This implies in particular that relative abundance determinations with respect to hydrogen, as derived from the observed column densities, may not be meaningful or are at least uncertain. ",
        "conclusions": ""
    },
    "0010/astro-ph0010369_arXiv.txt": {
        "abstract": "We present {\\sl XMM-Newton} observations of the eclipsing, disc accreting, cataclysmic variable OY Car which were obtained as part of the performance verification phase of the mission. The star was observed 4 days after an outburst and then again 5 weeks later when it was in a quiescent state. There is a quasi-stable modulation of the X-rays at $\\sim$2240 sec, which is most prominent at the lowest energies. We speculate that this may be related to the spin period of the white dwarf. The duration of the eclipse ingress and egress in X-rays is 20--30 sec. This indicates that the bulk of the X-ray emission originates from the boundary layer which has a negligible height above the surface of the white dwarf. The eclipse profile implies a white dwarf of mass $M_{1}$=0.9--1.1\\Msun\\hspace{1mm} and a secondary star of $M_{2}$=0.08--0.11\\Msun\\hspace{1mm}. ",
        "introduction": "Cataclysmic variables (CVs) are close binary systems in which the secondary (usually a dwarf main sequence star) fills its Roche lobe and transfers material onto the white dwarf primary. In non-magnetic systems, this material forms an accretion disc around the primary. Some of these show `dwarf nova' outbursts when the system brightens by several magnitudes for a few days or weeks. These outbursts recur on a timescale of weeks to months.  High inclination systems provide an opportunity to locate the X-ray emitting region(s). Eclipses are not seen in the X-ray emission of OY Car (Naylor et al 1988) nor in the EUV (Mauche \\& Raymond 2000) during outbursts, suggesting that the prime X-ray source (the boundary layer, where the disc is decelerated and merges with the white dwarf) is obscured at all phases. The residual X-ray flux is thought to be due to scattering from an accretion disc wind. In contrast, during quiescence X-ray eclipses have been seen in HT Cas (Wood et al 1995), Z Cha (van Teeseling 1997) and OY Car (Pratt et al 1999) when the boundary layer is not hidden by disc material. The duration of the eclipse ingress/egress shows that the X-ray source is small and close to the photosphere of the white dwarf, $<1.15R_{\\rm wd}$ in HT Cas (Mukai et al 1997).  X-ray observations of dwarf novae have been hampered by their relatively low flux levels. Using instruments with large effective areas, such as {\\sl XMM-Newton} (Jansen et al 2001), we can examine the eclipse profile of these systems with unprecedented precision and accurately locate the source of the X-ray flux. Having an orbital period of 1.51 hrs, OY Car was an ideal target for the performance verification phase of {\\sl XMM-Newton}. Specifically, OY Car was chosen to test the timing capability of the {\\sl XMM-Newton} Optical Monitor (OM, Mason et al 2001).  This paper presents an initial analysis of the optical and X-ray light curves. In a companion paper (Ramsay et al 2001), we present the results of an analysis of the X-ray spectra. ",
        "conclusions": "\\label{discuss}  \\subsection{The nature of the 2240 sec period}  The X-ray light curve of OY Car exhibits quasi-sinusoidal variability that may have an underlying periodicity of $\\sim$2240 sec. The variation is most prominent at softer energies but has a variable amplitude. Such a modulation could be due to energy dependent absorption, perhaps in disk material circulating above the orbital plane, or a quasi-stable bright spot on the photosphere of the white dwarf.  The latter scenario would imply that the spin period of the white dwarf ($P_{\\rm spin}$) is $\\sim$2240 sec. Marsh \\& Horne (1998) observed OY Car in the UV towards the end of a super-outburst and found evidence for dwarf nova oscillations (17.94 and 18.15 sec). One suggestion was that one of these periods represented the Keplerian frequency of the inner edge of the accretion disc and the other a beat frequency between the Keplerian frequency and $P_{\\rm spin}$. This scenario implies $P_{\\rm spin}\\sim$1500 sec.  Interestingly, a period of 1500 sec is close to the 1590-second orbital sideband of the 2240 sec period that we find in the amplitude spectrum (Figure \\ref{power}).  At this stage, we must regard this interpretation with a considerable degree of caution. However, it is interesting to note that a spin period of $P_{\\rm spin}$=2240 sec, implies $P_{\\rm spin}/P_{\\rm orb}$=0.41. The two Intermediate Polars (weakly magnetic CVs) which have orbital periods below the period gap -- like OY Car -- (RX J1238--38 and EX Hya) have $P_{\\rm spin}/P_{\\rm orb}$=0.42 (Buckley et al 1998) and 0.68 (C\\'{o}rdova, Mason \\& Kahn 1985) respectively. If the spin period in OY Car really is close to 2240 sec then it begs the question, is OY Car an Intermediate Polar?   \\subsection{The mass of the binary components} \\label{mass}  The most likely source of X-rays is from a boundary layer of negligible height which is extended in azimuth around the white dwarf. In \\S \\ref{eprof} we showed that in X-rays the ingress and egress of the eclipse was very rapid taking 20--30 sec. In addition, the duration of the eclipse in X-rays was 290$\\pm$10 sec. Using the observed times, we can place constraints on the system parameters.  There is good agreement regarding the system inclination: 83$^{\\circ}$ (cf Wood et al 1989). There is, however, a wide variation in the quoted mass of the white dwarf, $M_{1}$, varying from 0.33--1.26\\Msun. The mass of the secondary star, $M_{2}$, is not much better constrained, with estimates ranging from 0.07--0.15\\Msun. There is better agreement on the mass ratio, $q=(M_{2}/M_{1})$, where the more recent estimates quoted in Wood et al (1989) gave $q=0.10-0.11$.  Using an inclination $i=83.3^{\\circ}$ (Wood et al 1989) we determined the ingress time and total eclipse duration for a range of $M_{1},M_{2}$. We assume the Nauenberg (1972) mass-radius relationship for a white dwarf and the secondary star is a main sequence star (ie it is not over-massive for its spectral type as has been suggested for some secondary stars).  We determine the size of the eclipsed source by tracing the Roche potential out of the binary system along the line of sight from any point in the vicinity of the white dwarf. We then determine the ingress duration and eclipse duration for various combinations of $M_{1},M_{2}$. Comparing these times with the observed times (\\S \\ref{eprof}) we find that only for $M_{2}$=0.08--0.11\\Msun\\hspace{1mm} and $M_{1}$=0.9--1.1\\Msun\\hspace{1mm} can we match the observed times. Compared to previous estimates of the component masses these are tightly constrained. If the boundary layer has a non-negligible height, then $M_{1}$ will be more massive."
    },
    "0010/astro-ph0010019_arXiv.txt": {
        "abstract": "The ESA HIPPARCOS satellite has provided astrometry of unprecedented accuracy, allowing to reassess, improve and refine the pre-HIPPARCOS luminosity calibrations. We review the \"classical\" absolute magnitude calibrations with the Str\\\"omgren-Crawford in\\-ter\\-me\\-dia\\-te-band photometric system. A small zero point correction of about 2-4\\% seems necessary, as well as to refine the dependences on metallicity and projected rotational velocity. The need of a rigorous statistical treatment of the extremely precise HIPPARCOS data to derive definite dependences of the luminosity on physical stellar parameters is emphasized. ",
        "introduction": " ",
        "conclusions": ""
    },
    "0010/astro-ph0010549_arXiv.txt": {
        "abstract": "We construct a model for the formation of black holes within galactic bulges. The initial state is a slowly rotating isothermal sphere, characterized by effective transport speed $\\aeff$ and rotation rate $\\Omega$. The black hole mass is determined when the centrifugal radius of the collapse flow exceeds the capture radius of the central black hole. This model reproduces the observed correlation between black hole masses and galactic velocity dispersions, $\\mbh \\approx 10^8 M_\\odot (\\sigma/200 \\kms)^4$, where $\\sigma = \\sqrt{2} \\aeff$. This model also predicts the ratio $\\mrat$ of black hole mass to host mass: $\\mrat$ $\\approx$ 0.004 $(\\sigma/200 \\kms)$. ",
        "introduction": "Two groups have recently reported an observed relationship between a galaxy's velocity dispersion $\\sigma$ and the mass $\\mbh$ of its central (supermassive) black hole. This correlation can be written in the form \\be \\mbh = M_0 \\, (\\sigma/200 \\kms)^\\gamma \\, , \\label{eq:observed} \\ee where the two groups find $\\gamma$ = 3.75 (Gebhardt et al. 2000) and $\\gamma$ = 4.7 (Farrarese \\& Merritt 2000) and where the leading mass coefficient is $M_0 \\approx 1.2 - 1.3 \\times 10^8$ $M_\\odot$.  Both groups report little scatter about the relationship (the scatter in $\\mbh$ at fixed $\\sigma$ is bounded by 0.30 dex) and find no evidence for a subsidiary dependence on the Hubble type, profile type, or environment. This relationship supersedes a previous one claimed between the black hole mass and bulge luminosity (Richstone et al. 1998; Magorrian et al. 1998; van der Marel 2000; Kormendy \\& Richstone 1998). Furthermore, this measured scaling law poses a clear challenge for galaxy formation theories, which must ultimately account for this relationship.  Several theories imply a relationship between the central black hole mass and the galactic velocity dispersion. For example, a semi-analytic model of merger-driven starbursts with black hole accretion (Haehnelt \\& Kauffman 2000; Kauffman \\& Haehnelt 2000) provides a correlation of the form [\\ref{eq:observed}].  Several models are based on black hole accretion influencing star formation and gas dynamics in the host galaxy; this feedback can occur through ionization, mechanical work, and heating (e.g., Ciotti \\& Ostriker 1997, 2000; Blandford 1999; Silk \\& Rees 1998).  The model of Blandford (1999) gives $\\mbh < \\eta \\, \\sigma^{5/3}$ whereas Silk \\& Rees (1998) predict $\\mbh \\propto \\sigma^5$. Finally, the accretion of collisional dark matter indicates the scaling relation $\\mbh \\propto \\sigma^{4 - 4.5}$ (Ostriker 2000).  In this letter, we present a new theory of the $\\mbh - \\sigma$ relation, based on an idealized model for the collapse of the inner part of protogalaxies. The simplest variant of this theory is described in terms of particle dynamics in \\S 2. The role of gas dynamics is explored in \\S 3. We discuss the implications of this model and identify future issues in \\S 4. ",
        "conclusions": "In this paper, we have presented a simple model to describe the collapse flow that produces galactic bulges and the supermassive black holes living at their centers. The initial (pre-collapse) state is a slowly rotating isothermal sphere, characterized by an effective sound speed $\\aeff$ and an angular velocity $\\Omega$. These parameters $(\\aeff, \\Omega)$ thus represent the specification of the initial conditions. The velocity dispersion of the final stellar system is comparable to the initial sound speed and we make the identification $\\sigma \\approx \\sqrt{2} \\aeff$. In developing this basic picture, we find the following results:  [1] The black hole mass $\\mbh$ is determined by the condition that the centrifugal radius exceeds the capture radius of a Schwarzschild black hole. This requirement leads to the scaling law $\\mbh = M_0 (\\sigma/200 \\kms)^4$, which is consistent with observations both in its dependence on velocity dispersion $\\sigma$ and for the mass scale ($M_0 \\approx 10^8 \\Msun$) of the leading coeffecient (see equation [\\ref{eq:msig}] and Figure 1).  [2] The bulge mass is determined by the outer boundary of the collapsing region -- material at initial radii $r > R$ is rotationally supported and cannot collapse. This condition leads to the scaling law $M_B \\propto \\sigma^3$ (equation [\\ref{eq:mbulge}]) and predicts the ratio $\\mrat$ of black hole mass to bulge mass (equation [\\ref{eq:fraction}]). This mass fraction scales weakly with velocity dispersion and has a typical value $\\mrat \\approx 0.004$, similar to observed mass ratios (Figure 2).  Elliptical galaxies (and bulges) are described by four parameters: effective radius $R_e$, luminosity $L$, velocity dispersion $\\sigma$, and central black hole mass $\\mbh$. For this discussion, we substitute the total mass $M_B$ for the luminosity (for a given dark matter fraction, we thus assume that the transformation $M_B \\to L$ is determined by known, but perhaps complicated, stellar physics). We need to make the connection between these four basic quantities $(\\sigma, R_e, M_B, \\mbh)$ and our model, which is described by only two variables: the transport speed $\\aeff$ and the rotation rate $\\Omega$. Our model implies the following transformation between initial conditions and final system properties: [1] $\\sigma = \\sqrt{2} \\aeff$, [2] $R_e = \\aeff/ (2 \\Omega)$, [3] $M_B = 2 \\aeff^3 / (G \\Omega)$, [4] $\\mbh = 16 \\fa \\aeff^4 / (c G \\Omega)$, and [5] $\\mrat = 8 \\fa \\aeff/c$, where $\\fa \\approx 1.35$. The fifth relation is not independent, but because the rotation rate cancels out, it represents a robust prediction. The implications of these five scaling laws must be tested further against observations.  In this letter, we have presented a working collapse model for black hole and bulge formation. In future work, a number of issues should be addressed, including relativistic corrections to the infall solutions, additional black hole mass contributions from disk accretion, and feedback from the central black hole on the collapse. In order to successfully reproduce the observed correlations, our model uses initial conditions with a particular profile of specific angular momentum (determined by the initial $\\rho(r)$ and $\\Omega$); we thus need to explore more general initial conditions and make the connection between the required initial conditions and earlier stages of galaxy formation. In addition, galactic bulges are likely to experience merger events (White \\& Rees 1978) which lead to modest increases in the overall velocity dispersion (White 1979); we must determine how the black holes produced through this model evolve in the context of such galactic mergers. In any case, the properties of these supermassive black holes -- and their connection to their host galaxies -- will ultimately provide a vital diagnostic for the galaxy formation process."
    },
    "0010/astro-ph0010580_arXiv.txt": {
        "abstract": "In this paper, we discuss the current status of core collapse supernova models and the future developments needed to achieve significant advances in understanding the supernova mechanism and supernova phenomenology, i.e., in developing a supernova standard model. ",
        "introduction": "Beginning with the first numerical simulations conducted by Colgate and White\\cite{cw66}, three decades of supernova modeling have established a basic supernova paradigm. The supernova shock wave---formed when the iron core of a massive star collapses gravitationally and rebounds as the core matter exceeds nuclear densities and becomes incompressible---stalls in the iron core as a result of enervating losses to nuclear dissociation and neutrinos. The failure of this ``prompt'' supernova mechanism sets the stage for a ``delayed'' mechanism, whereby the shock is reenergized by the intense neutrino flux emerging from the neutrinospheres carrying off the binding energy of the proto-neutron star\\cite{w85,bw85}. This heating is mediated primarily by the absorption of electron neutrinos and antineutrinos on the dissociation-liberated nucleons behind the shock. This past decade has also seen the emergence of multidimensional supernova models, which have investigated the role convection, rotation, and magnetic fields may play in the explosion \\cite{hbhfc94,bhf95,jm96,mcbbgsu98a,mcbbgsu98b,fh99,khowc99}, in some cases invoking new explosion paradigms.  Although a plausible framework is now in place, fundamental questions about the explosion mechanism remain: Is there more than one mechanism? For example, are some core collapse supernovae neutrino driven while others driven by magnetohydrodynamic jets? Is there one mechanism for one class of massive stars---e.g., stars with progenitor masses between 10 and 20 M$_{\\odot}$---and another for another class of massive stars---e.g., more massive progenitors? Or is the mechanism tailored to each progenitor? For neutrino-driven supernovae, is the neutrino heating sufficient, or are multidimensional effects such as convection and rotation necessary? In addition, any ``standard model'' for core collapse supernovae must reproduce observed supernova phenomenology, such as neutron star kicks\\cite{fbb98} and the polarization of supernova emitted light\\cite{w99}. To answer these questions and to develop a ``standard model,'' simulations in one, two, and three dimensions must be performed and coordinated. ",
        "conclusions": ""
    },
    "0010/astro-ph0010255_arXiv.txt": {
        "abstract": "It is shown that photoabsorption via autoionizing resonances may be appreciable and used for abundance analysis. Analogous to spectral lines, the `resonance oscillator strength' $\\bar{f}_r$ may be defined and evaluated in terms of the differential oscillator strength df/d$\\epsilon$ that relates bound and continuum absorption. X-ray photoabsorption in KLL (1s2s2p) resonances of O~VI is investigated using highly resolved relativistic photoionization cross sections with fine structure. It is found that $\\bar{f}_r$ is comparable to that for UV dipole transition in O~VI (2s - 2p) and the X-ray ($1s^2 \\ ^1S_0 - 1s2p \\ ^1P^o_1)$ transition in O~VII. The dominant O~VI(KLL) components lie at $\\lambda\\lambda$ 22.05 and 21.87 $\\AA$. These predicted absorption features should be detectable by the {\\it Chandra X-Ray Observatory} (CXO) and the {\\it X-Ray Multi-Mirror Mission} (XMM). The combined UV/X-ray spectra of O~VI/O~VII should yield valuable information on the ionization structure and abundances in sources such as the `warm absorber' region of active galactic nuclei and the hot intergalactic medium. Some general implications of resonant photoabsorption are addressed. ",
        "introduction": "Whereas line absorption has been well studied and used for diagnostics and abundance analysis (Spitzer 1978), resonant absorption does not appear to have been similarly considered. This is probably due to the general complexity of resonances that require rather elaborate atomic physics calculations. On the other hand, resonances are ubiquitous, and may considerably  affect the effective cross sections. In this {\\it Letter} the theoretical treatment of resonant absorption is generalized using the quantity the differential oscillator strength that describes both bound-bound and bound-free absorption on either side of the ionization threshold. The method is applied to K-shell X-ray absorption in Lithium-like Oxygen.  The O~VI UV absorption in the $2s ^2S_{1/2}- 2p ^2P^o_{3,2,1/2}$ transition at $\\lambda\\lambda$ 1031.91 and 1037.61  $\\AA$ is widely observed in sources such as quasars and AGN (Mathur \\etal 1994, Tripp \\etal 2000), and {\\it Far Ultraviolet Spectroscopic Observer} (FUSE) sources (e.g. Savage \\etal 2000). Hellsten \\etal (1998) have predicted an `X-ray forest' of O~VII and O~VIII absorption lines from the low-z hot intergalactic medium as a probe of baryonic matter. Recently, X-ray absorption and emission line spectra have been reported from the CXO from H- and He-like ions such as O~VIII and O~VII (Kaspi \\etal 2000, Kaastra \\etal 2000). In their work Mathur \\etal (1994) reported on UV/X-ray absorption from the same element but in different ionization states, O~VI and O~VII, from the `warm absorber' region of AGN. The possibility of the same ionic species (O~VI) as both the UV and X-ray absorber is therefore of further interest for ionization structure and abundance studies. It is shown in this {\\it Letter} that resonant K-shell X-ray absorption by O~VI should lie among, but distinct from, the prominent emission lines of O~VII due to $2(^3S_1, ^3P^o_{1,2},\\ ^1P^o_{1}) \\longrightarrow 1(^1S_0)$ transitions.  KLL resonances are normally see as {\\it emission} lines (Gabriel 1972). Di-electronic recombination of highly ionized ions, for example $e~+~Fe~XXV \\longrightarrow Fe~XXIV$, leads to di-electronic satellite (DES) lines that are useful diagnostics of high-temperature sources such as tokamaks and solar flares (e.g. Bely-Dubau \\etal 1982, Beirsdorfer \\etal 1992). The radiative decay rates of many DES of Fe~XXV approach or exceed autoionization rates (e.g. Pradhan and Zhang 1997). For lighter elements, such as Oxygen, radiative decays are much smaller and these resonances should manifest themselves primarily in absorption, as demonstrated  in this work. ",
        "conclusions": "A few conclusions may be drawn from this study.  1. The strength of resonant photoabsorption may be computed and used for abundance analysis. Highly accurate and detailed atomic photoionization cross sections are required to obtain the corresponding PAR strengths. 2. The X-ray spectra from CXO and XMM should display the PAR absorption features at $22.05 \\AA$ and $21.87 \\AA$, lying in between the well known triplet emission features of He-like O~VII (Fig. 3). O~VI and O~VII exist in very different plasma conditions, with peak temperatures for maximum abundance that may differ up to an order of magnitude depending on photoionization and/or coronal equilibrium (Kallman 1995, Arnaud and Rothenflug 1985). The O~VI absorption in both UV and X-ray provides an additional tool for ionization and abundance studies. An examination of X-ray spectra is suggested for both the O~VII emission and the O~VI absorption features. 3. Radiative decay rates for autoionizing resonances may be obtained from the integrated $\\frac{df}{d\\epsilon}$ since the Einstein A-values are related to f-values (Wiese \\etal 1996). As confirmation of accuracy of the method presented, the computed $\\bar{f}_r$ in Table 1 are nearly equal (to two decimal figures) to those obtained from the $\\Gamma_r$. However, the quantity $\\frac{df}{d\\epsilon}$ is more general and represents photoabsorption in lines, resonances, and the non-resonant background at all energies. As such, it may be useful in complex cases with many overlapping resonances or lines. 4. The PAR features in absorption should manifest themselves as di-electronic satellites (DES) along an iso-electronic sequence as $\\Gamma_r \\sim Z^4$. Contrariwise, unlike heavier elements like iron where the DES are strong, for lighter elements like oxygen DES emission is very weak  (possibly undetectible), and the PAR satellites could be important absorption line diagnostics. 5. Radiative transfer in resonances may be significant and should be considered in NLTE and photoionization models. These and other points will be discussed in subsequent works."
    },
    "0010/astro-ph0010125_arXiv.txt": {
        "abstract": "We have obtained a moderately long ($10^5$ second) {\\it ASCA} observation of the Seyfert 1 galaxy Markarian 231, the most luminous of the local ultraluminous infrared galaxy (ULIRG) population. In the best-fitting model we do not see the X-ray source directly; the spectrum consists of a scattered power-law component and a reflection component, both of which have been absorbed by a column $N_H\\approx 3\\times 10^{22}\\pcmsq$. About $3/4$ of the observed hard X-rays arise from the scattered component, reducing the equivalent width of the iron K$\\alpha$ line. The implied ratio of $1-10$ keV X-ray luminosity to bolometric luminosity, $\\Lx/\\Lbol\\sim 2\\%$, is typical of Sy 1 galaxies and radio-quiet QSOs of comparable bolometric luminosities, and indicates that the bolometric luminosity is dominated by the AGN. Our estimate of the X-ray luminosity also moves Mrk~231 in line with the correlations found for AGN with extremely strong Fe II emission. A second source separated by about 2 arcminutes is also clearly detected, and contributes about $25\\%$ of the total flux. ",
        "introduction": "\\label{introduction} Sixteen years ago, the {\\sl IRAS} all-sky survey revealed the existence of a class of objects which emit nearly all of their energy at far-infrared wavelengths (Soifer \\etal\\ 1984). At the high-luminosity end of the distribution, $\\Lir\\ge 10^{12}\\Lsol$, the so-called ``ultraluminous infrared galaxies'' (ULIRGs), the space density of such objects is approximately twice that of optically-selected QSOs, the only objects with comparable bolometric luminosities (Sanders \\& Mirabel 1996). The far-infrared energy distributions of the prototype objects, Arp~220 and NGC 6240, are very similar to that of the classic starburst galaxy M82, peaking at $\\lambda\\simeq 60\\mu$m, and do not exhibit the ``warm'' {\\sl IRAS} colors typical of Seyfert galaxies (DeGrijp \\etal\\ 1992). A large fraction of the ultraluminous galaxies show evidence for interaction or merging with a companion galaxy (Sanders \\etal\\ 1988; Melnick \\& Mirabel 1990), and most show luminous molecular line emission, indicating substantial quantities of molecular gas, which could potentially fuel either a burst of star formation or an AGN (\\eg Sanders \\& Mirabel 1996, and references therein).  Ever since the discovery of this class of galaxies debate has raged over the nature of the dominant source of energy. The failure to detect any of the ULIRGs at X-ray wavelengths with HEAO-1 (Rieke 1988) strengthened a starburst interpretation, as did the recent failure of the {\\sl OSSE} instrument on the {\\sl GRO} to detect Arp~220, Mrk~231, or Mrk~273 in the $50-200$ keV energy range (Dermer \\etal\\ 1997); this rules out very luminous {\\sl Compton-thin} sources in these galaxies. Most recently, observations of a number of ULIRGs with the short-wavelength ({\\sl SWS}) spectrometer on the {\\sl Infrared Space Observatory} failed to detect any high-excitation lines in the mid-infrared, as might be expected from even a dust-shrouded AGN (Sturm \\etal\\ 1996; Genzel \\etal\\ 1998), and this has been taken to be evidence for starbursts rather than AGN. However, it is now clear that in many ULIRGs the column densities toward the nucleus are so high that the dust optical depth is greater than unity even in the mid-IR. Furthermore, the discovery of Compton-thick X-ray sources in a number of ULIRGs, notably including NGC 6240 (Iwasawa \\& Comastri 1998; Vignati \\etal\\ 1999), substantially weakens the argument against AGN-dominated sources based on hard X-ray non-detections. Such sources, in which even hard X-rays are seen only through reflection, are characterized by a relatively flat continuum above $E\\sim 3$ keV and very large equivalent widths of the iron $K\\alpha$ line, exceeding 1 keV (\\eg Krolik \\& Kallman 1987; Krolik, Madau \\& \\.Zycki 1994; Matt, Brandt, \\& Fabian 1996). Such large equivalent widths can be produced only if the X-ray continuum incident on the line-emitting region is much larger than that which we observe. If the reflected fraction in NGC 6240 is similar to that in NGC 1068, the bolometric luminosity of the AGN is of order $10^{45}$ \\ergps, and dominates the far-infrared output of the galaxy.  Foremost among the local ULIRGs is Markarian 231 ($z=0.042$) , which is the most luminous object in the local universe ($z < 0.1$) (Soifer \\etal\\ 1984). Its bolometric luminosity is $L_{bol}\\approx 3\\times 10^{12}\\Lsol$ ($H_o=75$ km s$^{-1}$ Mpc$^{-1}$), nearly all of which emerges in the far-infrared. Although classified as a Seyfert 1, Mrk~231 shows evidence for substantial amounts of reddening ($A_V\\sim 2$: Boksenberg \\etal\\ 1977) and when corrected for extinction its optical luminosity approaches quasar levels. Furthermore, Mrk~231 exhibits many features characteristic of (low ionization) Broad Absorption Line (BAL) QSOs: high optical polarization ($20\\%$ at 2800\\AA, Smith \\etal\\ 1995), strong Fe II emission (extremely strong in Mrk~231) (Boksenberg \\etal\\ 1977), red optical colors, and a large IR/optical flux ratio. {\\sl ASCA} and {\\sl ROSAT} observations of Mrk~231, analyzed by Iwasawa (1999) and Turner (1999), revealed an extended, soft X-ray component, with a luminosity $L\\sim 10^{42}\\ergps$, and a hard component exhibiting considerable absorption, with a 2-10 keV luminosity of only $L\\approx 2\\times 10^{42}\\ergps$. If this is the intrinsic X-ray luminosity, then an AGN is unlikely to dominate the energetics unless Mrk~231 is X-ray-quiet. However, the signal-to-noise of the {\\sl ASCA} spectrum was too low to determine whether the observed X-ray emission is largely reflected, in which case the true X-ray luminosity could be much higher.  In this letter, we present an analysis of a much deeper ASCA observation of Mrk 231 than that presented by Iwasawa. We find strong evidence for a powerful, buried AGN in this ULIRG which we observe only through reflection and scattering. Indeed, we show that the AGN in Mrk231 may well be the dominant energy source. ",
        "conclusions": "\\label{discussion} In his analysis of previous {\\it ASCA} data, Iwasawa (1999) was unable to distinguish between a very flat, unabsorbed power-law and a heavily absorbed, typical ($\\Gamma=1.8$) AGN spectrum. With the considerably better S/N provided by our {\\it ASCA} observation, we can show unequivocally that no amount of neutral absorption can steepen the intrinsic photon index into the normal AGN range. We are therefore led to conclude that there is a Compton-thick screen blocking the X-ray source along our line of sight, and we see the AGN only by reflection and scattering of the X-rays into our line of sight (Figure 3). Furthermore, unless the spectrum of the hard X-ray emission is unusually steep, the reflected/scattered emission is fairly heavily absorbed, by a neutral column $N_{\\rm H}\\approx 3\\times 10^{22}\\pcmsq$.  We can estimate the intrinsic luminosity of the X-ray source in two ways. When we subtract the contribution from the thermal emission, the observed $0.5-10$ keV luminosity is $2.0\\times 10^{42}\\,\\ergps$. Corrected for neutral absorption, this becomes $3.2\\times 10^{42}\\,\\ergps$, and with the reflection fraction ${\\cal R}$ set equal to zero, the scattered luminosity is $L_{\\rm sc}=2.4\\times 10^{42}\\,\\ergps$. (Note that this implies that about three-quarters of the {\\it observed} X-ray luminosity is due to scattering of X-rays into our line of sight, rather than reflection, thus reducing the equivalent width of the Fe K$\\alpha$ line from the value of $\\sim 1$ keV expected for a reflection-dominated spectrum.) Hence the intrinsic luminosity of the X-ray source is given by \\be L_{\\rm intr}=2.4\\times 10^{43}\\left({0.1\\over f_{\\rm sc}}\\right)\\ \\ergps \\ee where the scattering fraction $f_{\\rm sc}$ is expected to be on the order of $1\\%$ for electron scattering, assuming that the scattering column density is of the same order as the inferred absorption column density, \\ie a Thompson optical depth of order $0.01$.  Alternatively, we can estimate $L_{\\rm intr}$ from the reflected fraction and modeling of the reflection process. From the above numbers, the observed luminosity in the reflection component is $L_{\\rm refl}=8.0\\times 10^{41}\\ergps$. Using {\\sc xspec}, we conclude that the product of $L_{\\rm intr}$ and the reflection fraction $f_{\\rm refl}$ is \\be L_{\\rm intr}f_{\\rm refl}=2.0\\times 10^{43}\\ \\ergps \\ee and so \\be L_{\\rm intr}=2.0\\times 10^{44}\\left({0.1\\over f_{\\rm refl}}\\right)\\ \\ergps \\ee where we have normalized to $0.1$ under the assumption that only a small fraction of the reflecting surface is visible to us; for the Compton-thick Sy 2 NGC 1068, Iwasawa, Fabian \\& Matt (1997) infer $f_{\\rm refl}\\sim 0.05$. These two estimates of the luminosity agree for a ratio $f_{\\rm refl}/f_{\\rm sc}\\approx 10$, \\ie for a reflected fraction $f_{\\rm refl}\\approx 0.1$ for a scattered fraction $f_{\\rm sc}\\approx 1\\%$ which, as noted above, is the value expected if the scattering column is of the same order as the inferred absorbing column. (This ratio of $f_{\\rm sc}/f_{\\rm refl}$ is {\\it not} independent of the value ${\\cal R}\\approx 10$ derived from the spectral fitting.) Thus we infer that $f_{\\rm refl}\\approx 0.1$, $f_{\\rm sc}\\approx 0.01$, and $\\Lx\\approx 2\\times 10^{44}\\,\\ergps$. This is $200$ times larger than the value of Iwasawa (1999), and a factor $\\sim 20$ larger than the luminosity estimated by Turner (1999); hence our analysis implies that the AGN is much more X-ray-luminous than previously thought.  From our estimate of the intrinsic hard X-ray luminosity \\Lx, we can estimate the contribution of the AGN to the bolometric luminosity of Mrk 231. The implied ratio of $\\Lx/\\Lbol\\sim 2\\%$. For comparison, the ratio of $\\Lx/\\Lbol$ for the sample of Sy 1 and radio-quiet QSOs of Elvis \\etal\\ (1994) for which X-ray data are available ranges from $1.95\\%$ to $14.5\\%$, with the mean value decreasing somewhat with \\Lbol, from $6.3\\%$ for $\\Lbol=10^{45}-10^{46}$ (11 objects) to $4.3\\%$ for $\\Lbol=10^{46}-2.2\\times 10^{46}$ (7 objects); all the latter are classified as radio-quiet QSOs. Hence, within the uncertainties, the energetics of Mrk 231 are consistent with complete domination by the AGN; in any case, the fraction of the bolometric luminosity contributed by the AGN is almost certainly large ($\\gtrapprox 0.5$). Extrapolating to high energies, the predicted $50-200$ keV luminosity is $L_{50-200}\\approx 2\\times 10^{44}\\,\\ergps$, a factor of five below the Dermer \\etal\\ (1997) upper limit from {\\it OSSE}.  We also note that, with our estimate of the intrinsic X-ray luminosity, Mrk~231 is no longer an outlier in the significant correlations found by Lawrence \\etal\\ (1997) in the properties of AGN with extreme Fe II emission, in particular, those between the 2 keV -- 1$\\mu$m spectral index $\\alpha_{\\rm ix}$ and the Fe II emission strength and the emission line FWHM. Since our estimate of the flux density at 2 keV is $\\sim 270$ times larger than the value used by Lawrence \\etal\\ (from Rigopoulou, Lawrence, \\& Rowan-Robinson 1996), we find a value of $\\alpha_{\\rm ix}\\approx 1.2$, smaller by $0.8$ than the value inferred by Lawrence \\etal, which moves Mrk~231 into line with the bulk of strong Fe II emitters in their sample. Furthermore, the expected hard X-ray luminosity from the correlation between broad H$\\alpha$ and $2-10$ keV emission (Ward \\etal\\ 1988), using the H$\\alpha$ flux measured by Smith \\etal\\ (1995, uncorrected for extinction), is $L_x\\gtrapprox 2\\times 10^{44}\\ergps$, in excellent agreement with the inferred X-ray luminosity."
    },
    "0010/astro-ph0010313_arXiv.txt": {
        "abstract": "We present some early results from a study of five hot DA white dwarf stars, based on spectra obtained using STIS. All show multiple components in one or more of the strong resonance absorption lines typically associated with the stellar photosphere (e.g. \\civ, \\siiv, \\nv\\ and \\ov). Possible relationships between the non-photospheric velocity components and the interstellar medium or local stellar environment, are investigated, including contributions from gravitational redshifting. ",
        "introduction": "Several studies of hot DA white dwarf stars have revealed highly ionized species at non-photospheric velocities. While shifted features in the spectrum of Feige 24 could be explained by the binary nature of this system (Dupree \\& Raymond, 1982), {\\em IUE} observations of the isolated DA white dwarf CD -38\\deg10980 revealed Si and C absorption features at velocities displaced by -12 km/s with respect to the photospheric velocity. These features have been explained in terms of a dense gaseous halo in close proximity to the star (Holberg, Bruhweiler \\& Andersen (1995)).  We have examined high resolution EUV spectra of five hot (T$_{eff} > 54,000$ K) DA white dwarfs, % listed with their basic physical parameters in table 1. All show multiple components in the \\civ\\ resonance doublet, and some exhibit features in other lines. We compare the velocity of each component to that of the line-of-sight ISM lines observed in the spectra, and to the velocity of the local interstellar cloud (LIC). Using stellar models, gravitational redshifts are estimated to determine whether these shifted components could be produced by material within the gravitational well of the star, and column density estimates are derived to provide further information on the nature of the material. In the case of WD 2218+706, a possible link to a large molecular complex is suggested.  \\begin{table} \\caption{Physical parameters of the five stars included in this study} \\begin{tabular}{l c c c r r} \\tableline Star & T$_{eff}$ (K) & Mass (M\\sun) & Radius (R\\sun) & Log g & Dist (pc) \\\\ \\tableline G191-B2B & 54,000 & 0.51 & 0.020 & 7.5 & 69 \\\\ Feige 24 & 56,000 & 0.50 & 0.018 & $>7.5$ & 70 \\\\ WD 2218+706 & 56,900 & 0.40 & 0.033 & 7.0 & 436 \\\\ REJ 0558+165 & 62,860 & 0.58 & 0.019 & 7.7 & 295 \\\\ REJ 1738+665 & 69,340 & 0.53 & 0.024 & 7.4 & 239 \\\\ \\tableline \\end{tabular} \\end{table} ",
        "conclusions": "\\subsection{RE 1738+665} RE 1738+665 is the hottest DA white dwarf to be detected by {\\em ROSAT}, as described by Barstow et al. (1994). We determine a photospheric velocity $V_{phot} \\approx 30 \\pm 1$ km/s based on absorption features arising from Fe, Ni and O which show no multiple components; interstellar lines indicate a line-of-sight ISM velocity of $V_{ism} \\approx -18 \\pm 1$ km/s. The line-of-sight velocity of the LIC is estimated to be $V_{LIC} \\approx -3.4$ km/s. Figure 1 (left) shows the \\civ\\ resonance doublet in RE 1738, with shifted components at $-18.5\\pm 0.5$ km/s dominating the 30 km/s photospheric contribution.  Shifted features with similar velocities are determined for several other species, although in each case the photospheric component is dominant. The \\siiv\\ doublet shows clear evidence of multiplicity, with non-photospheric velocity components observed at -17.7 $\\pm 0.7$ km/s. Viewed individually, the lines of the \\nv\\ doublet (1238.821 and 1242.804 \\AA) show no evidence of companions, but co-addition of these features suggests an extra, weak line at $-15.2 \\pm 2$ km/s. The \\oiv\\ doublet (1338.612 and 1343.512 \\AA) shows no additional features, but the \\ov\\ line at 1371.292 \\AA\\ is accompanied by a weak shifted component at $-18.7 \\pm 0.5$ km/s.  \\begin{figure}[t!] \\plottwo{civ_1738.eps}{civ_g191.eps} \\caption{Velocity shifted, high ionization components in two of the stars under discussion. Left: clear multiplicity in the C IV doublet of RE 1738+665, with the photospheric components dominated by the blueshifted lines. Right: more subtle blending in the $\\lambda$1550 line of G191-B2B, shown with the compound Gaussian fit and its two components.} \\end{figure}  Our estimation of the gravitational redshift of this star is $V_{gr} \\approx 14$ km/s. Given uncertainties in this estimate and in the velocity of the shifted components, $V_{gr}$ is comparable to the velocity of these components, suggesting that the material does not reside within the gravitational well of the star.  RE 1738 exhibits shifted features in several species, at velocities consistent with the line-of-sight ISM velocity. It is therefore possible that these lines may be produced as a consequence of photoionization of material passing through the Str\\\"omgren sphere as the star moves through the ISM.  \\subsection{G191-B2B} We determine $V_{phot} \\approx 27 \\pm 2$ km/s and $V_{ism} \\approx 16 \\pm 1$ km/s which, considering the proximity of this star, is unsurprisingly close to V$_{lic}$ ($\\approx$ 20.6 km/s). Although not as clearly separated as the features in RE 1738, both lines of the \\civ\\ doublet are accompanied by a shifted component, the parameters of which have been estimated by fitting multiple gaussians to the data as illustrated in figure 1 (right). The component at photospheric velocity is of greater equivalent width, and the shifted elements are observed at velocities of $7.5 \\pm 0.5$ km/s. However, there is no evidence for corresponding features in any of the other lines examined.  We determine $V_{gr} \\approx 15.4$ km/s from stellar models, in agreement with the value of $19 \\pm 4$ km/s estimated by Reid \\& Wegner (1988). As in the case of RE 1738, this figure is comparable with the velocity difference between photospheric and shifted high ionization features, suggesting that the non-photospheric material probably resides outside the limit of the potential well. However, the velocity is substantially different to $V_{ism}$, and thus photoionization of the cloud responsible for the primary ISM lines is not a viable explanation for these features. No evidence exists for the presence of a planetary nebula surrounding this object, but we note observations by Green, Bowyer and Jelinsky (1990) who describe the possible detection of a \\hei\\ $\\lambda$584.3 emission feature in the stellar spectrum, postulating the existence of a hidden companion star or compact ($\\sim 10$ AU) circumstellar nebula as potential explanations for the line.  \\subsection{WD 2218+706} We estimate $V_{phot} \\approx -39 \\pm 1$ km/s, $V_{ism} \\approx -10$ km/s and $V_{lic} \\approx 5$ km/s. Lines of the \\civ\\ and \\siiv\\ doublets are clearly multiple, dominated by a photospheric contribution, but with companion features of comparable equivalent width at a velocity of $-16.5 \\pm 1$ km/s. WD 2218 is surrounded by an old planetary nebulae, and is discussed under the alternative designation DeHt 5, by Napiwotzki and Sch\\\"onberner (1995). The shifted components observed in this star are found at a velocity redward of the photospheric lines, suggesting that this material may be infalling; gravitational redshift % therefore provides no viable explanation. However, Dgani and Soker (1998) show that in regions where the ISM is reasonably dense (such as the galactic plane), Rayleigh-Taylor instabilities can develop in the outer regions of planetary nebulae, leading to fragmentation of the halo, and allowing the surrounding ISM to pass into the inner regions of the nebula where photoionization can occur.  Although WD 2218 is out of the galactic plane ($b_{II} = 11.6$\\deg), Kun (1998) describes the morphology of a nearby giant molecular cloud complex consisting of a large number of distinct regions previously identified in independent surveys; several are found close to WD 2218, and two are of particular interest (Lynds-1217 and Lynds 1219). The central portions of these clouds have galactic coordinates within less than 0.5\\deg of WD 2218, and their distance limits (from 380 to 450 pc) encompass the distance to WD 2218 (440 pc, from Napiwotzki \\& Sch\\\"onberner). This raises the possibility that the star may lie in an area where the ISM is particularly dense, allowing instability and inflow to take place (curve of growth analysis of the non-photospheric \\civ\\ features suggests a column density of $4.17 \\times 10^{13}$ cm$^{-2}$). However, in the absence of less equivocal evidence, this explanation must be regarded as conjecture. Alternative explanations, such as the presence of a hidden companion, are also deserving of investigation, and this work is currently in progress.   \\subsection{Feige 24} Feige 24 is a white dwarf+red dwarf binary system, and several detailed studies of this object have already been published (e.g. Dupree \\& Raymond (1982) and Vennes \\& Thorstensen (1994)). Vennes \\& Thorstensen estimate a systemic velocity of $62.0 \\pm 1.4$ km/s derived from absorption line measurements. We determine $V_{phot} \\approx 30 \\pm 1$ km/s, $V_{ism} \\approx 3 \\pm 1$ km/s and $V_{lic} \\approx 20.6$ km/s for the white dwarf. Multiple components are observed only in the \\civ\\ doublet, with dominant elements at the photospheric velocity, and secondary features at a velocity of $5.8 \\pm 1$ km/s. Our gravitational redshift estimate, $V_{gr} \\approx 17$ km/s, is somewhat higher than that derived by Vennes \\& Thorstensen ($V_{gr} \\approx 9 \\pm 2$ km/s), but is still too low to explain the secondary \\civ\\ features. Shifted components in the \\civ\\ doublet of Feige 24 have already been discussed by Dupree \\& Raymond, while Vennes \\& Thorstensen investigated the possibility that this material resided either in the photosphere, in a circumstellar shell, or in a wind from the red dwarf companion.  We note that the velocities of the shifted \\civ\\ features agree, within error, with the ISM velocity, and hence a link between the star and its immediate surroundings (beyond any circumstellar shell) cannot be discounted. However, Vennes \\& Thorstensen also estimate that a \\civ\\ column density ranging from $8 \\times 10^{11}$ to $8 \\times 10^{13}$ cm$^{-2}$, (corresponding to equivalent widths for the $\\lambda$1550 line of between 4-400 m\\AA), would be consistent with mass loss from the red dwarf companion. Although insufficient data is available to derive an unambiguous \\civ\\ column density in this study (figure 2 (right)), the range of possible values we obtain includes the estimates of Vennes \\& Thorstensen. Furthermore, the equivalent width of the $\\lambda$1550 line is measured as 24 m\\AA, consistent the earlier estimates. Clearly, much work remains to be done to improve our understanding of this system, and the recent release of a second STIS spectrum, taken at the opposite quadrature point, is a valuable addition to the available data.  \\begin{figure} \\plottwo{wd2218_cog.ps}{feige_cog.ps} \\caption{Curves of growth for shifted features associated with the \\civ\\ doublet in WD 2218 (left) and Feige 24 (right).} \\end{figure}  \\subsection{RE 0558+165} We determine $V_{phot} \\approx 28 \\pm 2$ km/s, $V_{ism} \\approx 10 \\pm 3$ km/s and $V_{lic} \\approx 25.2$ km/s. In common with Feige 24 and G191-B2B, this object shows evidence of non-photospheric components only in the \\civ\\ doublet, at a velocity of $10 \\pm 1$ km/s. Our gravitational redshift estimate suggests $V_{gr} \\approx 20$ km/s, so that RE 0558 is the only star in our sample with shifted high ionization features that could be formed by material {\\em within} the potential well, rather than the weakly shifted outer regions. However, the highly ionized, non-photospheric features lie at the velocity of the ISM, providing another possible explanation for the observations."
    },
    "0010/astro-ph0010307_arXiv.txt": {
        "abstract": "The 2dF Galaxy Redshift Survey has already yielded over 135,000 galaxy spectra (out of the planned 250,000). We present a method for the parameterisation of galaxy spectra in the 2dF and other large redshift surveys using a new continuous parameter $\\eta=0.44pc_1-pc_2$, based on a Principal Components Analysis (where $pc_1$ and $pc_2$ are the projections onto the first and second principal components respectively).  This parameter is designed to be robust to instrumental uncertainties and it represents the absorption/emission strength of a galaxy. ",
        "introduction": "The 2dF Galaxy Redshift Survey\\cite{ref1.1} is a combined UK-Australian effort to map the distribution of galaxies brighter than $b_J=19.45$ (with median redshift $z\\simeq0.1$). In so doing it will obtain a total of 250,000 galaxy spectra from which redshifts will be determined, a factor of 10 more than any previous survey.  Crucial to extracting the full scientific content of the survey is an accurate characterisation of the galaxy population in terms of spectral features. To this end we address the problem of obtaining an accurate and meaningful spectral classification that can be used in future analyses e.g. of luminosity functions and correlation functions. ",
        "conclusions": "We have attempted to derive a meaningful and robust parameterisation of the observed spectral distribution in order to characterise the galaxy population of the 2dFGRS. In so doing we have defined a new variable $\\eta$ which represents the most significant component of the galaxy sequence that is unaffected by instrumental uncertainties, but still retains physical information.  The parameter $\\eta$ is easily interpreted as an averaged absorption/emission strength and can be shown to have significant correlations with more conventional classifiers such as morphology and the H$\\alpha$ equivalent width.  There are certain limitations to our analysis, the most notable of which is that we have only derived a classification in the broadest sense. We have not attempted to identify the distinct sub-populations that are known to exist buried within the galaxy sequence, however there is no reason that such an analysis cannot be performed on the data in the future.  Other significant limitations to our analysis are the assumption that each spectrum can be reconstructed using a linear combination of only a few of the derived eigenspectra and the restriction to variance as a measure of information (both of which are implicitly assumed in the PCA).  These assumptions greatly simplify the analysis but are known to not be true in general.  It is in fact possible to extend the analysis without the use of these assumptions to perform an Independent Component Analysis. Another example of a non-linear method that has recently been investigated is that of an information-based clustering algorithm~\\cite{slonim}. Preliminary results from such analyses look promising."
    },
    "0010/astro-ph0010131_arXiv.txt": {
        "abstract": "\\bigskip  We analyze a morphologically-selected complete sample of 52 late-type (spiral and irregular) galaxies in the Hubble Deep Field North with total K-magnitudes brighter than K=20.47 and typical redshifts $z\\sim 0.5$ to 1.4. This sample exploits in particular the ultimate imaging quality achieved by HST in this field, allowing us to clearly disentangle the early- from late-type galaxy morphologies, based on accurate profiles of the surface brightness distributions. Our purpose was to investigate systematic differences between the two classes, as for colours, redshift distributions and ages of the dominant stellar populations. Our analysis makes also use of an exhaustive set of modellistic spectra accounting for a variety of physical and geometrical situations for the stellar populations, the dusty Interstellar Medium (ISM), and relative assemblies. The high photometric quality and wide spectral coverage allow to estimate accurate photometric redshifts for 16 objects lacking a spectroscopic measurement, and allow a careful evaluation of all systematics  of the selection [e.g. that due to the surface-brightness limit]. This sample appears to miss significantly galaxies above $z=1.4$ (in a similar way as an early-type galaxy sample previously studied by us), a fact which may be explained as a global decline of the underlying mass function for galaxies at these high redshifts. Differences between early- and late-types are apparent -- particularly in the colour distributions and the evolutionary star-formation (SF) rates per unit volume --, although the complication in spectro-photometric modelling introduced by dust-extinction in the gas-rich systems prevents us to reach conclusive results on the single sources (only future long-wavelength IR observations will allow to break the age/extinction degeneracy). However, we find that an integrated quantity like the comoving star-formation rate density as a function of redshift $\\Psi(z)$ is much less affected by these uncertainties: by combining this with the previously studied early-type galaxy sample, we find a shallower dependence of $\\Psi(z)$ on $z$ between $z=0.2$ and $z=1.5$ than found by Lilly et al. (\\cite{lilly}). Our present results, based on a careful modelling of the UV-optical-NIR SED of a complete galaxy sample -- exploiting the observed time-dependent baryonic mass function in stars as a constraint and attempting a first-order correction for dust extinction -- support a revision of the Lilly-Madau plot at low-redshifts for both UV- and K-band selected samples, as suggested by independent authors (Cowie et al. \\cite{cowie}). ",
        "introduction": "Cosmogonic models, in particular those based on the hierarchical clustering of cold dark matter halos, now including detailed physical descriptions of gas cooling, star-formation and feed-back processes in the baryonic component, make specific predictions about the evolutionary history of galaxy populations as a function of their morphology.  Basically, in the hierarchical scheme, forming galaxies acquire angular momentum from tidal interactions with the surrounding structure and then dissipate and collapse preferentially along the rotation vector and tend to form flattened rotational-supported structures (disk galaxies). There are indeed indications of a substantial population of large structures of this kind up to the highest redshifts from absorption-line studies in the distant quasar spectra (e.g. Wolfe A.M. \\cite{wolfe}). The pressure-supported stellar bulges dominating E/S0 galaxies, in this scheme, would originate from the violent relaxation and dynamical evolution following strong interactions and mergers  of primordial disk galaxies. At the zero-th order, these models predict that spheroidal galaxies are assembled somewhat later than spirals, although their stellar populations might not differ much if the merger occurs among gas-poor systems. One such extreme case has been discussed by Kauffmann \\& Charlot (\\cite{kauffmannb}) based on the $\\Omega=1$ standard CDM cosmology, predicting a substantial dearth of spheroidal galaxies already by z=1, most of them being formed at lower z.  More recently it has been pointed out that, within this scheme, the morphological appearence of a galaxy may repeatedly change with cosmic time not only from a late- to an early-type following a merging event, but also from a spheroidal to a disk configuration following the acquisition of new infalling gas from the environment (e.g. Ellis R.S. \\cite{ellis}). This may reflect in long-wavelength (V to K) colour distributions un-distinguishable between early- and late-types, while shorter-wavelength (U to V) colours would be dominated by the on-going SF in disks.  An opposite pattern is contemplated by the \"traditional\" models of galaxy formation, assuming that massive galaxies, in particular elliptical and S0's, originated first at high redshifts as single entities from rapid homologous collapse of primordial gas. Gas-rich systems, in this view, form instead more quiescently from progressive inflow of gas into the dark matter halos during most of the Hubble time. This formation scenario then predicts a marked differentiation in colours and ages for the stellar populations of the two classes of galaxies, late-type galaxies containing much younger stellar populations on average than the early-types. Also a substantial population of massive spheroids would be expected to be visible at $z\\sim 2$ in this case.  The ultra-deep integrations at various wavelengths performed by the Hubble Space Telescope in the Hubble Deep Field (Williams et al. \\cite{williams}; we consider here only the survey in the North area) offer an extremely valuable dataset to study morphological properties of high-redshift galaxies. Furthermore, the very accurate photometry achievable in such deep images allows accurate estimates of the photometric redshifts for vast numbers of faint galaxies in the field.  We have recently exploited these data to study the colours, masses, age distributions, and the star-formation history of a sample of elliptical-S0 galaxies (Franceschini et al.  \\cite{franceschini}, FA98 hereafter). The basic result was to find colours indicative of wide ranges of ages for the stellar populations and a remarkable absence of objects at $z>1.3$, both facts telling against the predictions of the \"traditional\" monolithic formation scenario.  As a natural complement, we present in this paper an analysis of late-type and irregular galaxies in the HDF. Similarly to what we did there, our primary selection is in the K-band, obtained from a deep KPNO image, to minimize the biases in the sample due to the effects of K- and evolutionary corrections. The completion of our previous analysis of E/S0 to account for the complementary set of late-type systems is also needed for a global evaluation of the star-formation history as a function of redshift. The advantage of our approach over previous attempts (Lilly et al. \\cite{lillyb}, Cowie et al. \\cite{cowie}) is in our careful treatment of dust extinction from a detailed fitting of the UV-optical-NIR spectral energy distributions (SED). In addition, the detailed knlowledge of the near-IR (NIR) spectrum for sources at the relevant redshifts is informative on the baryonic mass function in stars, which provides an essential constraint on the cumulative star-formation rate as a function of time.  In Section 2 we discuss the selection scheme and photometric corrections  used to construct a complete K-band flux limited sample of late--type galaxies. In Section 3 we describe the population synthesis code that we used to model the optical-NIR SEDs of our sample objects, taking into full account the effects of a dusty interstellar medium in the galaxy spectra. Our main results are then reported in Section 4, where we perform detailed analyses of the space distributions, colours and ages of the stellar populations of field spiral and irregular galaxies, compared with ellipticals and S0. We discuss the difficulties inherent in the spectral modelling of gas-rich systems affected by dust extinction. We finally attempt to construct the global star formation histories of field galaxies (E/S0+spiral/irregulars). In Section 5 we summarize our main conclusions.  We anticipate that the results of the  present analysis, based on a survey over a very small sky field, are to be considered as only tentative, untill larger areas will be surveyed to similarly deep limits.  We adopt $H_0=50~ Km~ s^{-1}~ Mpc^{-1}$ throughout the paper. For consistency with FA98 the analysis is made assuming $q_0=0.5$, and zero cosmological constant $\\Lambda$. ",
        "conclusions": "With the main goal to investigate systematic differences between early-type and late-type galaxies -- as for colours, redshift distributions, and ages of the dominant stellar populations -- we have analyzed a morphologically-selected complete sample of 52 spiral and irregular galaxies in the Hubble Deep Field North with total K-magnitudes brighter than K=20.47 and typical redshifts from $z\\sim 0.5$ to 1.5. The sample makes use of total photometry in the UBVI bands from HST and the JHK bands from ground, all carefully tested with an extensive set of Monte Carlo simulations.  The present sample exploits in particular the ultimate imaging quality achieved by HST in this field, allowing us to disentangle among galaxy morphologies, based on accurate profiles of the surface brightness distributions.  Our analysis makes also use of an exhaustive set of modellistic spectra accounting for a variety of physical and geometrical situations for the stellar populations, the dusty ISM, and relative assemblies. The high photometric quality and wide spectral coverage allowed us to estimate accurate photometric redshifts for 16 objects lacking a spectroscopic measurement.  We have also carefully evaluated all plausible systematic effects of the selection, in particular the redshift cutoff implied by the limiting surface-brightness achievable in the reference K band image.  A warning is in order, in any case, about the general conclusions derived from our sample of K-selected galaxies: they should be treated with caution, due to the very small field of view and modest spatial sampling of the present survey. Ferguson et al. (2000) and Eisenhardt et al. (2000) estimate that the number of $L^*$ galaxies in the total HDF co-moving volume between z=1 and z=2 is only a few dozens. Considering also the strong clustering inferred for Lyman break galaxies (Adelberger et al. 1998), statistical fluctuations imply large uncertainties on any conclusions based on samples like the HDF, untill more substantial surveys to similar depths will be made available.  Three the main results of our study.  \\begin{itemize}  \\item The sample galaxies are distributed in redshift up to $z=1.4$, but appear to be significantly missing above, compared with evolutionary models assuming standard recipes for the luminosity evolution and a substantial redshift of formation. We reported a similar finding in our previous study of early-type galaxies in the same area (Franceschini et al. 1998). Our conclusion is that, either the area has some peculiarities, or the underlying mass function for galaxies of all morphological kinds has a global decline at these high redshifts. Confirmation of this result will require more substantial sky areas to be surveyed to similar depths by large telescopes.  \\item Differences between early- and late-types are apparent in the rest-frame colour distributions and the evolutionary star-formation rates per unit volume. In particular, the short-wavelength rest-frame colours B-J, once dust reddening is taken into account, appear quite significantly bluer for late- than for early- galaxy types. On the contrary, the longer-wavelength colors V-K appear to be very similar for the two morphological classes. We interprete this as an indication that, while the stellar mix in spirals/irregulars includes young newly formed populations, less apparent in E/S0, the underlying older age component traced by the V-K colors has quite a similar origin and age distribution for the two galaxy categories.  We warn, however, that the complication in spectro-photometric modelling introduced by dust-extinction in the gas-rich systems prevents to reach conclusive results on the source by source basis. Only future long-wavelength IR observations, from space (SIRTF, FIRST, NGST) and from ground (10 m class telescopes in the mid-IR and interferometers in the sub-mm), will allow to break down the age/extinction degeneracy.   \\item We found that an integrated quantity like the comoving-volume star-formation rate density as a function of redshift $\\Psi(z)$ is much less affected by the uncertainties related to the dust distribution. The reason for this is mostly in the fact that our analysis is strongly constrained by the evolutionary baryonic mass function in stars traced by the near-IR galaxy luminosities, the estimate of the baryon mass at any redshifts being much more robust than that of the instantaneous rate of star formation (see Sect. 4.5).  By combining this with the early-type galaxy sample previously studied by FA98, we find a shallower dependence of $\\Psi(z)$ on $z$ between $z=0.2$ and $z=1.5$ than found by Lilly et al. (1996), i.e. $\\Psi(z)\\propto (1+z)^{1.4}$ rathen than $\\Psi(z)\\propto (1+z)^{4}$ as in Lilly et al. (\\cite{lillyb}). In this redshift interval our observed  $\\Psi(z)$ turns out to roughly agree with results published by Cowie et al. (\\cite{cowie}) and Treyer et al. (\\cite{treyer}).  Our present results, based for the first time on a careful modelling of the whole UV-optical-NIR Spectral Energy Distributions of galaxies, then support a revision of the Lilly-Madau plot at low-redshifts for UV- and K-band selected galaxies. UV-selected and near infrared selected galaxy samples display a remarkably similar evolution of the comoving SFR density $\\Psi(z)$, at $z<1$.  \\end{itemize}  The three above findings seem to favour the general scheme of hierarchical assembly for the formation of bright galaxies, envisaging their progressive build up during a substantial fraction of the Hubble time. After all, this is the most physically motivated present description.  In this context, a warning is in order concerning some published specializations of the Cold Dark Matter cosmogonic scheme, predicting that spheroidal galaxies in the field form at low redshift ($z<1$) from merging of spirals (Kauffmann \\& Charlot \\cite{kauffmannb}). This prediction is not supported by our results in Fig.\\ \\ref{sfrtot}, where Ellipticals and S0s appear to have been mostly formed at $z=1$, whereas spirals/irregulars keep some sustained SF activity at $z<1$. This result suggests that merging of spirals to form ellipticals at low redshifts cannot be a dominant process, quite in agreement with what found by Brinchmann \\& Ellis (\\cite{brinchmannb}).  It is conceivable, however, that minor modifications of the CDM hierachical scheme (e.g. in terms of different assumptions about the cosmological parameters $\\Omega_m$, $\\Omega_{\\Lambda}$) can explain the observed differences between morphological types. In our view, these differences in the population histories are quite probably related to the presence of different environments at different densities (and consequently different cosmic timescales of formation) in what we call the \"field\". In particular, moderately high-density environments (typically galaxy groups, as found very numerous in the spectroscopic survey by Cohen et al. \\cite{cohenb}), with an accelerated cosmic timescale of evolution and fast gas consumption, mix with truly low-density environments, where the transformation of primordial gas into stars slowly progresses during the whole Hubble time. We believe that the two galaxy morphologies analyzed in the present paper and in FA98 trace such different environments in the universe."
    },
    "0010/astro-ph0010288_arXiv.txt": {
        "abstract": "The simplest model of the afterglows of the gamma-ray bursts (GRBs) envisions a spherical blast wave with a power-law distribution of electron energy above some cutoff running into a constant density medium. A refinement involves a narrow jet, often invoked to explain the steep decline and/or steepening of light curves observed in some afterglows. The constant (ambient) density jet model has been applied to GRBs 991208 and 000301C, based to a large extent on radio observations. We show that, for these two sources, a spherical wind model (with an $r^{-2}$ density ambient medium) can fit the radio data as well as the jet model.    % The relatively steep decline and the fairly abrupt steepening of the R-band light curves of, respectively, GRB 991208 and GRB 000301C can be accounted for with a non-standard, broken power-law distribution of electron energy. Our model predicts a slower late decline for the radio flux than does the jet model. ",
        "introduction": "The simplest model of the afterglows of gamma-ray bursts (GRBs) involves a relativistic spherical blast wave running into a constant density, presumably interstellar, medium (M\\'esz\\'aros \\& Rees 1997; Katz 1994; see Piran 1999 for a review). The afterglows are emitted by non-thermal electrons whose energy distribution is usually assumed to be a power-law above some cutoff. The predicted power-law decay of the afterglow emission with time has been subsequently observed at X-ray (Costa et al. 1997), optical (van Paradijs et al. 1997), and radio (Frail et al. 1997) wavelengths, giving basic confirmation to this now ``standard'' picture.  There are observed features of afterglows that are difficult to accommodate by the simplest model. One of the most noticeable is the steepening of the optical light curves. Good examples include GRBs 990123 (Kulkarni et al. 1999), 990510 (Harrison et al. 1999; Stanek et al. 1999), and 991216 (Halpern et al. 2000). The steepening is usually attributed to a jet-like, instead of spherical, blast wave (Rhoads 1997; Sari, Piran \\& Halpern 1999).  Radio data are especially useful for analyzing the afterglows of GRBs because the self-absorption frequency, $\\nu_a$, and the characteristic frequency of the lowest energy electrons, $\\nu_m$, both typically lie in this range. GRB 970508 was followed extensively at radio wavelengths (Frail, Waxman, \\& Kulkarni 2000) and Frail et al. (2000) have suggested that there was an initial spherical relativistic expansion in a constant density medium (up to day 25), followed by lateral jet expansion (days 25--100) and spherical non-relativistic expansion (days 100--400). The same data were modeled by Chevalier \\& Li (2000) as a spherical blast wave running into an $r^{-2}$ density medium characteristic of a constant mass loss rate and velocity circumstellar wind, possibly of Wolf-Rayet star origin. The model  missed an early low frequency radio data point and some infrared points, but it  captured most of the observed behavior over 400 days. The question of a wind vs. a constant density surrounding medium is crucial for the question of the progenitors of GRBs, because massive stars, one of the leading candidates for GRB progenitors (Paczy\\'nski 1998), should be surrounded by a wind.  Fairly extensive radio observations are now available for GRBs 991208 and 000301C (Galama et al. 2000; Berger et al. 2000). At  optical wavelengths, the light curves of both sources are somewhat unusual: those of GRB 991208 are steeper than $t^{-2}$ (Sagar et al. 2000a), and those of GRB 000301C show pronounced steepening (Rhoads \\& Fruchter 2000; Masetti et al. 2000; Sagar et al. 2000b; Jensen et al. 2000). These two sources have been modeled as a jet running into a constant density medium by Galama et al. (2000) and Berger et al. (2000), based on both radio and optical data. The authors ruled out the simplest, spherical wind model with a standard, power-law electron energy distribution. Here, we show that the spherical wind model fits the observed radio data as well as  % the constant density jet model. We also demonstrate that a non-standard distribution of electron energy can account for both the steep decline and the steepening of the optical light curves of, respectively, GRBs 991208 and 000301C. In \\S\\ref{model}, we review the spherical wind model and describe the non-standard electron energy distribution. The model is then applied to the GRBs 991208 and 000301C in \\S\\ref{source}. We discuss our results % in  \\S\\ref{discuss}. ",
        "conclusions": "\\label{discuss}  Our study of these two GRBs  can be separated into the radio evolution and the optical/IR evolution. ISM + jet models have been proposed for the radio emission from both sources, but we find that spherical wind interaction models can produce fits of comparable quality. The same is true for GRB 970508 (Frail et al. 2000; Chevalier \\& Li 2000). In the case of GRB 991208, the approximate power law behavior observed over a factor 10 in time is intermediate between that expected for a spherical explosion in a constant density medium and the asymptotic lateral expansion of a jet (Galama et al. 2000). Kumar \\& Panaitescu (2000a) found that the transition to the asymptotic jet evolution should take at least an order of magnitude, perhaps several orders of magnitude, in observer time. Yet the model of Berger et al. (2000) for GRB 000301C assumes a rapid transition to the asymptotic jet evolution. The uncertainties in the jet models give them additional parameters that are not present in the wind interaction models which assume spherical symmetry throughout the evolution. Kumar \\& Panaitescu (2000a) found that for jet expansion in a wind medium, any break in the observed light curves would be unlikely to be detected, even for a narrow jet with an opening angle of about a few degrees. The spherical assumption, which we make, should be adequate even if a jet is present.  The problem with the simplest wind model is that the optical light curves of both GRB 991208 and GRB 000301C drop more rapidly than predicted. We have solved the problem by invoking a  steepening of the electron spectrum. We attribute the steepening to the acceleration process, although the details of how this occurs are not clear. Bednarz \\& Ostrowski (1998) have studied acceleration in relativistic shock waves and found that at high shock Lorentz factors, the spectrum tends to  a power law with $p\\approx 2.2$. However, at low Lorentz factors ($\\gamma\\approx 3$), which are characteristic of afterglow shock waves, there are a number of possibilities, including $p < 2$, depending on the shock parameters. A particle spectrum that is this flat must steepen at high energy in order to have finite total energy. The spectral steepening with $p$ increasing to a relatively large value of $3.3$ or more that we propose for GRBs 991208 and 000301C would be clearer if X-ray observations of the sources were available. Unfortunately, these two sources are not constrained by X-ray observations. Spectral steepening with $p$ increasing to a smaller value may also be possible. In such a case, the X-ray flux would be less affected.  The difference between an intrinsic spectral break and a laterally expanding jet is that, in the latter case, the break in a light curve should be achromatic. It appears to be difficult to decide this point solely on the basis of optical/IR observations because of the faintness of the sources and the sparse sampling. The difference between the radio and optical evolution might be clearer. The late radio light curve of GRB 000301C is most complete at 8.46 GHz and the wind model appears to give a better representation of these data than the jet model (compare Fig. [2] to Fig. [2] of Berger et al. 2000), although the uncertainties are too large to draw any firm conclusions. A similar test may be possible for GRB 991208 (see \\S~3.1). Another test provided by radio observations is at low frequencies. In the wind interaction model, self-absorption is high at early times so that the low frequency flux should be low. Both GRB 970508 and GRB 000301C have one early 1.4 GHz radio observation that is higher than expected in the wind interaction model (Chevalier \\& Li 2000; Fig. [2]); the 1.4 GHz observations of GRB 991208 show a decrease that is not expected in any model. More extensive early, low frequency observations would provide a useful test of  models."
    },
    "0010/astro-ph0010241_arXiv.txt": {
        "abstract": " ",
        "introduction": "M2$-$9 is a spectacular, well studied bipolar nebula, usually classified as a planetary nebula (PN).  It consists of two oppositely directed, narrow, cylindrical-like bright lobes, each extending to an angular projected distance of $\\sim 20 ^{\\prime \\prime}$ from the center of the nebula (Fig.~1a). Farther out, there are two ``ansae,'' one on each side along the symmetry axis (Schwarz et~al.\\ 1997), at $\\sim 60^{\\prime \\prime}$ from the center, moving outward at $\\sim 190 \\km \\s^{-1}$ (Schwarz et~al.\\ 1997). The inner lobes on the other hand, expand at a much slower speed, of  $\\sim 40 \\km \\s^{-1}$ (Schwarz et~al.\\ 1997; Solf 2000), or maybe even $15 \\km \\s^{-1}$ (Trammell, Goodrich \\& Dinerstein 1995; hereafter TGD). The ansae have a point symmetrical structure (namely, are at symmetrical points when reflected about the central star).  The nebula and its central region show many interesting properties, which are summarized in, e.g.\\ Balick (1989), TGD, Hora \\& Latter (1994), Phillips \\& Cuesta (1999), Solf (2000), and Doyle et~al.\\ (2000; hereafter DBCS). Balick (1989), for example, finds an H$\\alpha$ linewidth of $11,000 \\km \\s^{-1}$, implying the presence of a wind having a velocity of $\\sim 5,500 \\km \\s^{-1}$ from the nucleus. Balick further notes that ``the spectrum of M2$-$9's nucleus is more similar to the slow nova RR~Tel, some symbiotic stars, and Seyfert (type~1.9) galactic nuclei...'' (see also Solf 2000).  In view of these, and other intriguing properties (see below), it it important to identify the nature of this nebula and to place it in a broader evolutionary context.   In the present paper we attempt to do precisely that. We also address specifically a perplexing positional (time dependent) shift of the bright knots in the inner lobes, $\\sim 8^{\\prime\\prime}$ from the nucleus (Allen \\& Swings 1972), which was recently studied in detail by DBCS. The shifts in the northern and southern lobes are in the {\\it same\\/} direction, i.e.\\ there is a mirror-symmetry rather than a point-symmetry of the bright knots. The picture that emerges from many studies (e.g.\\ Allen \\& Swings 1972; van den Bergh 1974; Goodrich 1991; Kohoutek \\& Surdej 1980; and for more information see \\S1 of TGD) is that of two UV illuminating beams in a mirror symmetry, which move around the central UV source at a period of $\\sim 100 \\yrs$.   TGD conducted a detailed study of the ionization state of the shifting knots. One of their conclusions was that the off-knots regions are ionized by attenuated UV radiation. They assumed that the lobes contain rings of higher density, and concluded that: ``The UV source illuminates a section of the lobe, causing the rings of material to glow,  producing the knots and explaining and naturally leading to their ``fixed'' north-south positions. Obscuring material near the star keeps one side of the lobe from being illuminated by the full force of UV source.'' TGD also estimated that the column density of the obscuring material is $\\sim 10^{18-18.4} \\cm^{-2}$.  Goodrich (1991; see also Scarrott, Scarrott \\& Wolstencroft 1993) suggested that the radiation that ionizes the knots comes from a companion, and not from the central star itself. DBCS, on the other hand, attribute the ionization to a fast jet hitting the bright spots. As we shall explain in the Discussion (\\S4), we identify some difficulties in the scenario proposed by DBCS.  Instead, we propose in the present paper a different model, which characterizes M2$-$9 as a stage in the evolution of interacting, detached binary systems.  The important properties of M2$-$9, as they can be deduced from the different studies are as follows. \\begin{enumerate} \\item The bolometric luminosity of the central star is not known. The star is heavily obscured (e.g.\\ no point source is observed in the snapshot HST survey; H.~Bond 2000, private communication). A fit to the IR spectrum gives a total flux of $1.1\\times10^4$~L$_{\\odot}$(D/kpc)$^2$ (S.~Kwok 2000, private communication), but the distance is rather uncertain (Schwarz et~al.\\ 1997 suggest $\\sim650$~pc). \\item Lim \\& Kwok (2000) argue that the mass loss rate from the central source is $\\ll 10^{-5}$~M$_\\odot \\yr^{-1}$, and maybe even below $10^{-6}$~M$_\\odot \\yr^{-1}$. \\item The two ansae at very large distances from the lobes expand at $\\sim 200 \\km \\s^{-1}$ (Schwarz et~al.\\ 1997), a high velocity compared with that of the lobes which expand at 15--$40 \\km \\s^{-1}$ (Solf 2000). \\item The ionizing source is very probably a WD companion (e.g.\\ Goodrich  1991; Schwarz et~al.\\ 1997; Lim \\& Kwok 2000). \\item Material moving at $\\gtrsim 5,000 \\km \\s^{-1}$ is found near the center of the nebula (Balick 1989).  This wind was suggested to result from an accretion disk around the WD companion (e.g.\\ Torres-Peimbert \\& Arrieta 1998). This suggestion is consistent with the general finding that outflow speeds are always of the order of the Keplerian speed around the central object (Livio 1999, 2000). \\item The ionizing radiation escapes mainly in two directions, north and south, with an inclination of $\\sim 30^\\circ$ to the symmetry axis. The inclination to the symmetry axis is in the {\\it same\\/} direction for both ``beams'' (west for both, or east for both; Fig.~1a).  The beams rotate around the central star with a period of $ 120 \\yrs$ (DBCS). \\item The ionizing radiation does reach other regions around the lobes, but it is attenuated in those directions (TGD). \\item The Galactic latitude of M2$-$9 is the largest in the list of 43 bipolar PNe of Corradi \\& Schwarz (1995). For an assumed distance of $650 \\pc$ it implies a Galactic height of  $200 \\pc$. This may imply that the progenitor of M2$-$9 was not very massive compared to other bipolar PNe progenitors (perhaps a late~A or~F rather than a B~star). \\end{enumerate}  We argue that {\\it all\\/} of these properties can be adequately explained in the context of a model in which a central binary system that is a supersoft X-ray source (SXS), is in the transition phase from a symbiotic nebula (possibly generated by a symbiotic nova; Solf 2000) to a PN. In particular, we demonstrate that a system consisting of an asymptotic giant branch (AGB) or a post-AGB star (with its associated slow wind) and a hot WD (which has a fast wind), in an orbital period of $\\sim120$~yrs, can ionize the ``walls'' of the lobes in one direction, while the ionization is substantially attenuated in other directions. The revolving ionizing source is explained by a combination of the attenuating effects of the post-AGB star's wind and by the effects of a collimated fast wind (CFW) blown by the WD companion, which is bent by the slower wind from the post-AGB star. This type of flow, where the CFW is being strongly pushed upon and bent by the AGB (or post-AGB) wind is termed a `weak-CFW' flow by Soker \\& Rappaport (2000, hereafter SR00). The bent ``jets'' can both compress the ring in each lobe, and clear a way for the ionizing radiation in the two specific directions around the symmetry axis. The circumbinary material has special properties (in terms of its optical depth, see below) for a relatively short period of time (of the order of a few hundred years), and this enables the formation of the circulating ionizing beams.  Our model provides an important link between planetary nebulae and symbiotic nebulae.  The fact that these two types of nebulae may be related in their shaping mechanisms (and that the nebulae are sometimes misclassified) has been known for some time (e.g.\\ Morris 1987, 1990; Corradi 1995; Corradi \\& Schwarz 1995; Schwarz \\& Corradi 1992; Corradi et~al.\\ 2000). Furthermore, our proposed model for M2$-$9 emphasizes the important role played by binaries in the shaping of nebulae in general, and in the formation of jets in particular. The formation of collimated jets in some of these systems, as evidenced, for example, by the recent observations of He2$-$90 (Sahai \\& Nyman 2000), provides important clues for the processes by which astrophysical jets are accelerated and collimated in general (Livio 2000).  Our dynamical model is presented in \\S2, the properties of the ionizing beams are discussed in \\S3, and a summary and discussion follow. ",
        "conclusions": "M2$-$9 is a unique nebula in that it shows a variability on a time scale which is consistent with the orbital period expected from the binary model for the formation of bipolar nebulae. In this model most (although not all) bipolar nebulae are formed by binary systems with orbital periods in the range of $\\sim 1$--$10^3 \\yrs$ (e.g.\\ SR00). The time varying features exhibit a mirror symmetry, rather than a point-symmetry (TGD; DBCS). As we have shown, a mirror symmetry is expected from the interaction of two winds, a slow wind emitted by the AGB or post AGB star, and a jet (or collimated fast wind [CFW]) emanating from the accretion disk around the WD companion. The mirror symmetry is produced as the CFW is pushed and bent by the slow wind. Our model identifies the evolutionary state of M2$-$9 as one which relates three different classes of binary systems: supersoft X-ray sources (SXSs), symbiotic nebulae, and PNe (or proto-PNe). The ionizing source is the white dwarf companion, and not the mass-losing star (as already concluded by e.g.\\ Goodrich  1991; Schwarz et~al.\\ 1997; Solf 2000; Lim \\& Kwok 2000). At the same time, the fast outward moving ``ansae'' (Schwarz et~al.\\ 1997) and the large and bright nebula strongly suggest that in the recent past the system has both undergone a more intense mass loss episode, and an outburst associated with the accreting WD companion.  Hence, our evolutionary scenario for the nebula can be characterized as a symbiotic nebula evolving to a PN.  While M2$-$9 is unique in its time variation, it is not in its morphology. The bipolar proto PN~Henize 401, for example, has a similar structure (Sahai, Bujarrabal, \\& Zijlstra 1999), but with no evidence for circulating ionizing beams. We note that Sahai et~al.\\ (1999) argue that the bipolar structure in He~401 was formed by collimated bipolar jets, and not by the more commonly assumed interacting winds model, in which a spherical fast wind catches up with a previously ejected bipolar slow wind (e.g.\\ Balick 1987).  The model of Sahai et~al.\\ has in fact some features that are similar to those of our model for M2$-$9.  Our model offers an interpretation to the bright knots that is different from the one proposed by DBCS. We showed that the bright knots are exposed to a higher flux of ionizing radiation (from the hot WD) than the rest of the lobes, by virtue of the path cleared by the CFW. DBCS, on the other hand, attribute the brightness/ionization of the knots to shock heating, caused by jets impacting on lobe walls. We see a few problems with this suggestion: (i)~it offers no clear explanation for the mirror symmetry (the jets will collide with slow wind material ejected half a period earlier and will be slowed down substantially). (ii)~The fact that the off-knots regions are ionized as well (TGD) argues for an ionizing source that is not confirmed to the small cross section of a jet, which is present at all times. (iii)~We find the `corkscrew' model of DBCS (see e.g.\\ their Fig.~2) to provide for a rather poor description of the nebula (e.g.\\ the lines that are supposed to represent the emission in 1989 and 1999 do not follow very well the actual emission). We rather interpret the 1989 and 1999 images as follows.  The bright emission regions (from the center of the nebula to the bright blobs along the symmetry axis) are moving in the same sense, with the 1989 emission being toward the outskirts (east side) relative to the 1999 emission. The change in the east-west location occurs only near the bright blobs (N$_3$ and S$_3$), and even there it represents a complicated flow structure, and radiative transfer from these regions outward. Nevertheless, our model does share some common features with that of DBCS (e.g.\\ a fast outflow from the companion).  One point in our model deserves further discussion. A cleared path in a certain direction for the ionizing beam (and therefore enhanced ionizing radiation), could result, in principle, in a more extended, rather than brighter, nebula (which is not observed). We should therefore check that for the parameters of M2$-$9 there is no significant penetration of the ionization front in the direction {\\it perpendicular\\/} to the lobe walls. For an electron density in the lobes of $n_e\\sim10^4$~cm$^{-3}$ (DBCS), an ionizing source of $\\dot S\\sim4\\times10^{46}$~s$^{-1}$ (eq.~8), and a path clearing for about 10\\% of the period, one can easily show that the penetration of the ionization perpendicular to the wall would be by less than $\\sim0.1D$ (where $D$ is the radial distance to the S$_2$, N$_2$ knots). Thus, while we do not expect any significant perpendicular penetration, we do expect the bright regions to extend {\\it along\\/} the lobes (due to the lower density), as observed (Fig.~1a). \\newpage  To conclude, the spectacular nebula M2$-$9 provides an opportunity to identify an evolutionary link among three types of (empirically defined) binaries: PNe with binary central stars, symbiotic binaries, and supersoft x-ray sources.   {\\bf ACKNOWLEDGMENTS:} ML acknowledges support by NASA Grant NAG5-6857. NS acknowledges partial support by a grant from the Israel Science Foundation, and a grant from the US-Israel Binational Science Foundation.  We acknowledge extremely helpful comments and suggestions from Saul Rappaport."
    },
    "0010/gr-qc0010030_arXiv.txt": {
        "abstract": "We have investigated the possibility of having a late time accelerated expansion phase for the universe. We have used a dissipative fluid in Brans-Dicke(BD) theory for this purpose. The model does not involve any potential for the BD scalar field. We have obtained the best fit values for the different parameters in our model by comparing our model predictions with SNIa data and the also with the data from the ultra-compact radio sources. ",
        "introduction": " ",
        "conclusions": ""
    },
    "0010/astro-ph0010077_arXiv.txt": {
        "abstract": "We present near-infrared spectroscopy for a complete sample of 33 ultraluminous infrared galaxies at a resolution of \\( R\\approx 1000 \\). Most of the wavelength range from 1.80--2.20 \\( \\mu  \\)m in the rest frame is covered, including the Pa\\( \\alpha  \\) and Br\\( \\gamma  \\) hydrogen recombination lines, and the molecular hydrogen vibration-rotation 1--0 S(1) and S(3) lines. Other species, such as He~I, {[}Fe~II{]}, and {[}Si~VI{]} appear in the spectra as well, in addition to a number of weaker molecular hydrogen lines. Nuclear extractions for each of the individual galaxies are presented here, along with spectra of secondary nuclei, where available. The Pa\\( \\alpha  \\) emission is seen to be highly concentrated on the nuclei, typically with very little emision extending beyond a radius of 1~kpc.  This survey was carried out primarily to search for signatures of active nuclei via velocity-broadened hydrogen recombination or the presence of the {[}Si~VI{]} coronal line. These signatures are rare in the present sample, occurring in only two of the 33 galaxies. The extinction to the hydrogen recombination lines is investigated via the Pa\\( \\alpha / \\)Br\\( \\gamma  \\) line ratio. It is found that visual extinctions to the nuclei in excess of 10 magnitudes are relatively common among ULIRGs, and that visual extinctions greater than 25 mag are necessary to conceal a QSO emitting half the total bolometric luminosity. The ionized hydrogen regions in many ULIRGs are sufficiently obscured that dust-enshrouded active galactic nuclei would remain hidden at 2 \\( \\mu  \\)m at the current level of sensitivity. The vibration-rotation lines of molecular hydrogen appear to be predominantly thermal in origin, with effective temperatures generally around 2200 K. The relative nuclear velocities between double nucleus ULIRGs are investigated, through which it is inferred that the maximum deprojected velocity difference is \\( \\sim 200 \\) km~s\\( ^{-1} \\). This figure is lower than the velocities predicted by physical models of strong interactions/mergers of large, gas-rich galaxies. ",
        "introduction": "Understanding the nature of the ultimate power source in ultraluminous infrared galaxies (ULIRGs) has been a primary driver for numerous scientific investigations ever since the discovery in the 1980's that ULIRGs represent the most luminous class of galaxies in the local universe. With total power outputs comparable to those of quasars, characterizing the source of power is of fundamental importance in understanding the composition and evolution of normal galaxies. ULIRGs are found to share the common trait that virtually all have recently or are currently experiencing major encounters with other galaxies \\citep{twm96,clements,sanders,carico}. Encounters like these may play a significant role in the process of building normal galaxies. With the discovery of faint sources at sub-millimeter wavelengths having characteristic spectral energy distributions like that typically found in ULIRGs (e.g., Arp~220) and redshifts \\( z\\geq 2 \\), the importance of understanding these systems and their role in the evolution of normal galaxies has been heightened.  The present survey was designed to look for near-infrared spectral signatures of obscured active galactic nuclei (AGN) in ULIRGs, either by the presence of velocity-broadened Pa\\( \\alpha  \\) emission or by the appearance of the high excitation {[}\\ion{Si}{6}{]} coronal line. A summary of the conclusions of this survey, along with a median ULIRG near-infrared spectrum, are presented in \\citet{twm99}. Only two of the 33 sample galaxies show clear indications for AGN activity, with the rest appearing to be dominated by star formation processes. Either the AGN phenomenon is relatively rare among ULIRGs, or the extinction at 2 \\( \\mu  \\)m is still too high to permit a proper assessment. Mid infrared studies \\citep[e.g.,][]{genzel,rig} favor the former conclusion, though it is found in this study that the extinction at 2 \\( \\mu  \\)m may often be greater than four magnitudes, resulting in over one magnitude of extinction at mid-infrared wavelengths.  This paper presents the individual nuclear spectra for the sample galaxies, for both the primary and secondary nuclei where possible. Two-dimensional spectra are also presented to give a qualitative view of the line emission distributions typically found in ULIRGs. Some discussion of extinction measures, molecular hydrogen excitation mechanisms, and a summary of typical line ratios found in ULIRGs is included, as is a discussion of the velocity differences found among double nucleus systems. ",
        "conclusions": "Near-infrared spectra in the 2 \\( \\mu  \\)m window have been obtained for a nearly complete, volume-limited sample of 33 ULIRGs. Lines of atomic hydrogen and helium recombination, molecular hydrogen vibration-rotation, {[}\\ion{Fe}{2}{]}, and {[}\\ion{Si}{6}{]} are seen in the spectra. The Pa\\( \\alpha  \\) is predominantly seen as a nuclear event, with the majority of the emission arising in the central kpc of most ULIRGs. Initiated as a search for optically hidden active galactic nuclei, this survey reveals only two such galaxies, both of which are characterized as Seyfert galaxies based on visible-light spectra. It is found via comparison of the Pa\\( \\alpha  \\) and Br\\( \\gamma  \\) lines that significant extinction of several magnitudes persists into the near-infrared bands, such that buried active nuclei can not be ruled out in many objects. It is estimated that with the sensitivity of the current survey, at least 25 mag of visual extinction would be needed to hide a QSO with half the bolometric luminosity of the ULIRG.  Strong molecular hydrogen emission is seen in all of the primary nuclear spectra. The molecular hydrogen spectrum indicates that most H\\( _{2} \\) emission in ULIRGs stems from thermal excitation rather than fluorescent processes. Some ULIRGs appear to have a small fraction of their molecular hydrogen emission owing to fluorescence, though none approach pure fluorescence.  The relative radial velocities found in this sample of ULIRGs does not exceed \\( \\sim 120 \\) km~s\\( ^{-1} \\), with an inferred maximum deprojected velocity differential of around 200 km~s\\( ^{-1} \\). This relatively low velocity differential is difficult to produce from interactions between galaxies with masses comparable to that of the Milky Way using simple physical assumptions. It is suggested that ULIRGs in the late stages of merging may be selectively found near the time of apocenter, leading to a lower average relative velocity between nuclei."
    },
    "0010/astro-ph0010657_arXiv.txt": {
        "abstract": "We use mid-IR (primarily 10 $\\mu$m) photometry as a diagnostic for the presence of disks with inner cavities among 32 pre-main sequence stars in Orion and Taurus-Auriga for which rotation periods are known and which do not show evidence for inner disks at near-IR wavelengths. Disks with inner cavities are predicted by magnetic disk-locking models that seek to explain the regulation of angular momentum in T Tauri stars. Only three stars in our sample show evidence for excess mid-IR emission. While these three stars may possess truncated disks consistent with magnetic disk-locking models, the remaining 29 stars in our sample do not. Apparently, stars lacking near-IR excesses in general do not possess truncated disks to which they are magnetically coupled.  We discuss the implications of this result for the hypothesis of disk-regulated angular momentum. Evidently, young stars can exist as slow rotators without the aid of present disk-locking, and there exist very young stars already rotating near breakup velocity whose subsequent angular momentum evolution will not be regulated by disks. Moreover, we question whether disks, when present, truncate in the manner required by disk-locking scenarios. Finally, we discuss the need for rotational evolution models to take full account of the large dispersion of rotation rates present at 1 Myr, which may allow the models to explain the rotational evolution of low-mass pre-main sequence stars in a way that does not depend upon braking by disks. ",
        "introduction": "A keystone of the current paradigm of rotational evolution among low-mass pre-main-sequence (PMS) stars is the regulation of stellar angular momentum by circumstellar disks. Indeed, most rotational evolution models today rely chiefly upon some form of disk-regulated stellar angular momentum to connect the rotational properties of T Tauri stars (TTS) to those of zero-age main sequence (ZAMS) stars \\citep{bouvier97,krishnamurthi,cameron95}. The theoretical edifice supporting the disk-regulation paradigm has been developed by several researchers \\citep{konigl91,ostrikershu,yi95,cameron93,ghosh}, wherein a steady-state, magnetically mediated, star-disk coupling mechanism keeps the star at roughly constant angular velocity for the lifetime of the disk, preventing the star from spinning up as it contracts. This magnetic disk-locking model provides an attractive solution to the long-standing puzzle of how stars deplete angular momentum as they evolve to the main sequence \\citep{stauffhart}.  Observational investigations of disk-regulated angular momentum have yielded mixed results. In support of disk-locking, several authors have, either directly or indirectly, linked disks to slow stellar rotation. For example, Herbst and collaborators \\citep[and references therein]{ch} have reported the distribution of TTS rotation periods in the Orion Nebula Cluster (ONC) to be distinctly bimodal. The slow rotators ($P_{\\rm rot} > 4$ days), they suggest, represent those stars that are presently, or were until recently, disk-locked. More direct support has been presented by \\citet{edwards93}, who reported a correlation between the presence of excess near-IR emission and slow stellar rotation among 34 TTS in Taurus and Orion. Considering stars in their sample later than K5, those authors found excess near-IR emission among none of the rapid rotators but among half of the slow rotators. \\citet{bouvier93} reported similar results among 26 TTS in Taurus-Auriga. They found that the classical TTS (CTTS; those showing evidence of active accretion and circumstellar disks) in their sample rotate more slowly on average than their weak-lined (WTTS) counterparts. In contrast to these studies, we did not find strong evidence for a bimodal distribution of rotation periods, nor did we find evidence for a linkage between disks and slow stellar rotation, in our own study of disks and stellar rotation in Orion \\citep[hereafter Paper~I; see also Stassun et al.\\ 2000 for a concise summary]{stass}. Among 254 stars with $0.1 < P_{\\rm rot} < 8$ days, we found the distribution of rotation periods to be statistically consistent with a uniform distribution. Moreover, we found near-IR excesses to be present with roughly equal frequency among both slow and rapid rotators when we combined our rotation periods with near-IR photometry from the literature \\citep{hill98} for more than 100 stars in our sample.  \\citet{herbst99} have shown that subdividing the ONC sample by mass can reconcile some of these contradictory findings. They find that the bimodal distribution of rotation periods in Orion holds true when they consider only stars with $M_\\star > 0.25 {\\rm M}_\\odot$, i.e.\\ earlier than about M3. For these stars, they also find a weak but statistically significant correlation between near-IR excess and stellar rotation period; the slower rotators tend to exhibit larger near-IR excesses.  Whether these IR observations of TTS in fact agree with the predictions of disk-locking models is an important question that has yet to be fully explored. While the various prescriptions for magnetic star-disk coupling differ in detail, a key prediction of the models is that the stellar magnetosphere will truncate the disk at an inner radius, $R_{\\rm trunc}$, that is very nearly equal to the co-rotation radius, $R_{\\rm c}$ (the distance from the star at which disk material rotates with Keplerian angular velocity equal to the stellar angular velocity). Consequently, stars with $R_{\\rm c} \\gg R_\\star$ (i.e. slow rotators) should have $R_{\\rm trunc} \\gg R_\\star$, and so are {\\it not} expected to show their disks at near-IR wavelengths (e.g.\\ \\citet{meyer97}). Ironically, while previous authors have upheld the disk-locking hypothesis by linking slow stellar rotation to the presence of excess near-IR emission, it is the slow rotators that {\\it lack} near-IR excesses that may be most consistent with the disk-locking scenario. An important question, therefore, is whether slow rotators lacking near-IR excesses indeed possess truncated disks, or if they are simply diskless.  A similar question applies with respect to {\\it rapid} stellar rotation as well.  We reported in Paper~I the existence of very rapidly rotating ($P_{\\rm rot} < 2$ days) TTS in the ONC, some with rotation periods corresponding to breakup velocity. Roughly one-third of the rapid rotators in the ONC do not evince excess emission in the near-IR, indicating that they do not possess inner disks. What is to be the rotational fate of these apparently diskless PMS ultra-fast rotators, and what implications do they carry for rotational evolution models? At $\\lesssim 1$ Myr \\citep{hill97}, these stars still face contraction in size by factors of 5--10, yet {\\it already rotate at or near breakup velocity}. If these stars are truly diskless as the near-IR data suggest, then they must shed most of their angular momentum prior to the main sequence without the aid of disk-locking.  In this paper we present new mid-IR (primarily $N$-band) photometry for a sample of low-mass TTS in the ONC and in Taurus-Auriga for which rotation periods have been determined and which do not show excess emission in the near-IR. We supplement these new data with existing data from the literature to form a sample of 32 stars, roughly evenly divided between slow ($P_{\\rm rot} > 4$ days) and rapid rotators. Our aim is two-fold: First, we wish to determine whether slow rotators lacking near-IR excesses possess truncated disks, consistent with disk-locking models, or if these stars are simply diskless. Second, we wish to determine whether {\\it rapid} rotators lacking near-IR excesses are in fact diskless, or if they possess truncated disks to which they may be coupled in a non-steady-state and which may influence their subsequent rotational evolution.  The data, and the methods we employ in their analysis, are described in \\S 2. The results of our search for truncated disks are presented in \\S 3. We find evidence for excess mid-IR emission among only three of the stars in our sample. These stars may possess disks with $R_{\\rm trunc} \\approx R_{\\rm c}$, consistent with disk-locking models. However, no disks are evident at mid-IR wavelengths among the remaining 29 stars in our sample; these stars are evidently diskless. It thus appears that many stars can exist as slow rotators at 1 Myr without the aid of present disk-locking. In addition, there exist very young stars already rotating near breakup velocity whose subsequent angular momentum evolution will not be regulated by disks. In \\S 4 we discuss the implications of these results both for the hypothesis of disk-regulated angular momentum and for rotational evolution models of low-mass PMS stars. We summarize our conclusions in \\S 5. ",
        "conclusions": "We have used new and existing mid-IR photometry to search for circumstellar disks among 32 low-mass PMS stars in the ONC and in Taurus-Auriga for which rotation periods have been measured but which do not show evidence for disks at near-IR wavelengths. Our principal aim in this study has been to determine whether these stars harbor disks with large inner truncation radii, or if they are simply diskless. The stars in our sample are roughly evenly divided between slow ($P_{\\rm rot} > 4$ days) and rapid ($P_{\\rm rot} < 4$ days) rotators. The slow rotators, if shown to possess disks, would be consistent with large truncation radii predicted by magnetic disk-locking models. We similarly wish to determine if the rapid rotators in our sample are coupled to truncated disks, perhaps undergoing rapid spin-down at present, which could help explain the rotational fate of stars already rotating near breakup velocity at 1 Myr \\citep{stass}. The most rapid rotators require a braking mechanism before reaching the ZAMS; evidence for disk material might suggest non-steady-state disk coupling.  The primary observational result of this study is that, with a few possible exceptions, the stars in our sample---both the rapid and slow rotators---are not presently disk-locked, because they do not have disks to which the stars may magnetically couple. Any disk material around these stars must be situated at $\\gtrsim 1$ AU, well beyond the typical co-rotation radius of $\\sim 0.05$ AU for the stars in our sample. By extension, this implies that roughly one-third of slowly rotating PMS stars exist at 1 Myr without the aid of present disk-locking, and that a similar fraction of very rapidly rotating TTS (including some rotating near breakup velocity) will not have their subsequent angular momentum evolution to the main sequence regulated by circumstellar disks.  Combining this new result with those of \\citet{stass}, we draw two key interpretations with respect to magnetic disk-locking and rotational evolution models of low-mass PMS stars: \\begin{enumerate} \\item The dichotomy of disked slow rotators and diskless rapid rotators is a false one. Thus, we do not find compelling evidence on the basis of disk frequency to suggest that slow rotators are the evolutionary precedents of rapid rotators at 1 Myr. Low-mass PMS stars at 1 Myr possess a large dispersion of rotation rates that is similar to---and that indeed may exceed---the dispersion observed among low-mass ZAMS stars. This includes a PMS population of ``ultra-fast rotators\" rotating near breakup velocity. The inclusion of rapid rotators at the beginning of rotational evolution calculations may lessen the degree to which the models require the braking action of disks \\citep{sills}.  \\item Disks, when present, do not appear to truncate at co-rotation as required by disk-locking models for the regulation of stellar angular momentum. Models of TTS disks suggest that disks must extend well inside of co-rotation, i.e. $R_{\\rm trunc} \\ll R_{\\rm c}$, in order to explain the near-IR excesses observed among disked TTS. Furthermore, the previously reported positive correlations between slow stellar rotation period and excess near-IR emission are in the opposite sense of that predicted from disk-locking models. \\end{enumerate}  These results, in concert with those previously discussed by us \\citep{stass} and by others \\citep{armitage99,safier,chakrabarty}, present some fundamental challenges to current models of angular momentum evolution among young, low-mass PMS stars, which have come to rely upon disk-locking. The full impact of these conclusions awaits rotational evolution modeling that incorporates the rotational initial conditions advocated here, including rotational differences among PMS stars of different masses \\citep{herbst99}, as well as a more thorough investigation of angular momentum evolution prior to the stellar birthline."
    },
    "0010/astro-ph0010461_arXiv.txt": {
        "abstract": "We introduce the properties of the Two Micron All-Sky Survey (2MASS) survey for IAU Symposium 204.  2MASS is a near-infrared survey of the entire sky characterized by high reliability and completeness.  Catalogs and images for 47\\% of the sky are now available online.  This data release has been used by Wright (2000) and Cambr\\'esy et al. (2000) to subtract the stellar foreground at 1.25 and 2.2 microns from COBE DIRBE data, revealing the cosmological near-infrared background. ",
        "introduction": "The Two Micron All-Sky Survey (2MASS) is an ongoing survey to map the entire sky at near-infrared wavelengths.  Motivated by the particular application of 2MASS to the determination of the extragalactic near-infrared background, our goal in this paper is to review the status and quality of 2MASS. The 2MASS dataset has many applications, and it is our hope that this introduction to the data will motivate other investigators to utilize the 47\\% of the sky presently publically available. The 2MASS catalogs and images are available online at http://www.ipac.caltech.edu/2mass/ and are fully described by Cutri et al. (2000). ",
        "conclusions": "Nearly half the sky is now publically available.  At the time of this writing (11 October 2000), 99.8\\% of the sky has been observed with catalog-quality data.   The present data release to the astronomical community has enabled studies from the solar system to cosmology even before the final processing of the full 2MASS dataset."
    },
    "0010/astro-ph0010182_arXiv.txt": {
        "abstract": "We present ISO observations of several \\Hd\\ pure-rotational lines (from  S(0) to S(5)) towards a sample of 16 molecular clouds distributed along the central $\\sim 500$ pc of the Galaxy. We also present \\CuoO\\ and \\utCO\\  \\juce\\ and \\jdu\\ observations of these sources made  with the IRAM-30m telescope. With the CO data we derive \\Hd\\ densities of 10$^{3.5-4.0}$~\\cmmt\\ and  \\Hd\\ column densities of a few 10$^{22}$~\\cmmd. We have  corrected the \\Hd~ data for $\\sim 30$ magnitudes of visual extinction using a self-consistent method. In every source, we find that the \\Hd\\ emission   exhibits a large temperature gradient. The S(0) and S(1) lines trace temperatures ($T$)  of $\\sim 150$~K while the S(4) and S(5) lines indicate temperatures of $\\sim 600$~K. The warm \\Hd\\ column density is  typically $\\sim 1-2\\times 10^{22}$~\\cmmd, and is predominantly gas with $T$=150 K. This is the first direct  estimate of the total column density of the warm molecular gas in the Galactic center region. These warm \\Hd\\ column densities represent a fraction of $\\sim 30 \\%$ of the gas traced by  the CO isotopes emission. The cooling by \\Hd~ in the warm component is comparable to that by CO. Comparing our \\Hd\\ and CO data with available ammonia (\\NHt) observations from literature one obtains relatively high \\NHt\\ abundances of a few $10^{-7}$ in both the warm and the cold gas. A single shock or Photo-Dissociation Region (PDR)  cannot explain  all the observed \\Hd\\ lines. Alternatives for the heating mechanisms are discussed. ",
        "introduction": "The interstellar matter in the inner few hundreds parsecs of the Galaxy  (hereafter GC)  is mainly molecular. In this region there are  molecular clouds  and huge cloud complexes like Sgr~B2 which can be as large as 70 pc, with masses of  10$^6$ solar masses. The physical conditions in the GC clouds differ appreciably to those of the clouds of the disk of the Galaxy. The GC clouds have average densities of $\\sim 10^4$ \\cmmt\\ instead of 10$^2$ \\cmmt\\ typical of the clouds of the disk. In addition, with widespread high temperatures (up to 200 K), GC clouds are hotter than disk clouds.  The temperatures of the warm  gas are known mainly by observations of ammonia (\\NHt ) metastable lines. G\\\"usten et al. (\\cite{Gusten81}, \\cite{Gusten85}) derived rotational temperatures ($T_\\mathrm{rot}$) of 60-120 K in several GC clouds, most of them in the Sgr A complex. Morris et al. (\\cite{Morris83}) showed that $T_\\mathrm{rot} \\sim 30-60$~K are common in the region $|l|<2^\\circ$. The most complete study of the temperature structure of the molecular gas in the GC, was carried out by H\\\"uttemeister et al. (\\cite{Huttemeister93}). They presented a multilevel study of \\NHt\\ metastable lines  of 36 molecular clouds distributed all along the ``Central Molecular Zone\" (CMZ, in notation of Morris \\& Serabyn \\cite{MS96}) and the ``Clump 2\" complex, which, although not belonging to the actual CMZ, exhibits similar properties. They detected warm gas at all galactic longitudes and showed that the \\NHt\\ emission can be characterized by two temperature components since the $T_\\mathrm{rot}$ derived from the (1,1) and (2,2) levels is $\\sim 20-30$~K and that derived from the (4,4) and (5,5) levels is  $\\sim$ 70--200~K. Unfortunately, the {\\it a priori} unknown abundance of the \\NHt\\ molecule has made it difficult to estimate the total column density of warm gas in the GC clouds.  The heating of the molecular gas over large regions ($\\sim 10$~pc) where the dust temperature is lower than 30 K (Odenwald \\& Fazio \\cite{OF84}, Cox \\& Laureijs \\cite{CL89}, Mart\\'{\\i}n-Pintado et al. \\cite{MP99}, Rodr\\'{\\i}guez-Fern\\'andez et al. \\cite{RF00}) is a puzzle. Indirect arguments such as  the large widths of molecular lines or large abundances in gas phase of molecules such as SiO (Mart\\'{\\i}n-Pintado et al. \\cite{MP97}, H\\\"uttemeister et al. \\cite{Huttemeister98}) or \\NHt\\ points towards a mechanical heating. Wilson et al. (\\cite{Wilson82}) proposed the dissipation of turbulence induced by differential Galactic rotation as a possible heating source.  For the first time, we have  measured the total  column densities of warm gas in the GC clouds by observing the lowest \\Hd\\ pure-rotational transitions with the {\\em Infrared Space Observatory} (ISO; Kessler et al. \\cite{Kessler96}). The \\Hd\\ pure-rotational lines trace gas with temperatures of a few hundred K (see Shull $\\&$ Beckwith \\cite{SB82} for a review on the properties and the notation of the \\Hd\\ molecule). ISO has  detected  \\Hd\\ pure-rotational lines in a variety of sources such as: Young Stellar Objects (Van den Ancker \\cite{vdA99}); galactic nuclei (see e.g. Kunze et al. \\cite{Kunze99}); Photo Dissociation Regions (PDRs) like NGC 7023 (Fuente et al. \\cite{Fuente99}, \\cite{Fuente00}) or S140 (Timmermann et al. \\cite{Timmermann96}); shock-excited sources such as Orion Peak 1 (Rosenthal et al. \\cite{Rosenthal00}); and proposed x-ray excited regions (XDRs) like RCW 103 (see Wright \\cite{Wright00}).  Our sample consists    of 18 molecular clouds from the samples of H\\\"uttemeister et al. (1993) and Mart\\'{\\i}n-Pintado et al. (1997). Two of these show a non-equilibrium \\Hd\\ ortho-to-para ratio and have been studied in detail  by Rodr\\'{\\i}guez-Fern\\'andez et al. (2000). In this paper we present the other 16 clouds of the sample. The clouds are distributed along the CMZ, from the Sgr E region to the vicinity of Sgr D and the ``Clump 2\" complex. Four  clouds are located in the Sgr C complex, three in the vicinity of Sgr A (two are in  the radio Arc). Two clouds are situated in the cold dust ridge reported by Lis $\\&$ Carlstrom (\\cite{LC94}) that seems to connect the radio Arc and Sgr B. Other three clouds belong to the Sgr B complex.  This  paper is organized as follows. In Sect. 2 we present \\CuoO\\ and \\utCO\\   IRAM-30m observations and        \\Hd\\ ISO observations. The analysis of the CO isotopes and \\Hd , is presented in Sect. 3 and 4, respectively. The results and the possible heating mechanism of the warm gas are discussed in Sect. 5. ",
        "conclusions": "We have observed the S(0) to S(5) \\Hd\\ pure-rotational lines  with the SWS spectrometer on-board ISO toward a sample of 18 molecular clouds of the Galactic center region. The S(3) line is strongly affected by dust extinction due to the 9.7 $\\mu$m band of the silicates. After correcting the \\Hd\\ data for extinction using a self-consistent method, and assuming that the ortho- and para-\\Hd\\ populations are in equilibrium one finds that the S(0) and S(1) lines indicate temperatures of $\\sim 150$ K. Extrapolating to the lowest levels at that temperature, a total \\Hd\\ column density of $\\sim 1-2\\,10^{22}$~\\cmmd\\ is derived. This is the first direct estimate of the column density of warm gas in the GC clouds. In addition, it shows a complex temperature structure of the warm gas.  The temperature derived from the S(5) and S(4) levels is $\\sim 600$ K in the sources in which it can be derived,  however the column density of gas at this temperature is less than $1\\%$ of the column density at $T$=150~K. Assuming an OTP ratio of  $\\sim 2$ the temperatures would be $10\\%$ larger than those derived assuming a LTE OTP ratio, while the total \\Hd\\ column densities at those temperatures  would be a factor of  $\\sim 1.8$ lower than the column densities derived assuming the ortho- and para-\\Hd\\ populations in equilibrium. Comparing the \\Hd\\ warm column densities with the column densities derived from our CO data by LVG calculations one finds that the average fraction of warm  \\Hd\\ to the gas observed in CO is $\\sim 30 \\% $. With our data and the \\NHt\\ observations of H\\\"uttemeister et al. (1993) we derive relatively high \\NHt\\ abundances of a few 10$^{-7}$ in both the warm and the cold components.  Several indirect arguments point to shocks as the heating mechanism of the warm gas but PDRs may also play a role. Direct comparison of the \\Hd\\ data with  PDRs and shocks models indicate  that the S(4) and S(5) trace the densest gas in the GC clouds (\\lsim 10$^6$ \\cmmt) heated in  PDRs or shocks. Nevertheless, such dense PDRs or shocks  fail to explain the S(0) and  S(1) lines: several less dense PDRs, low velocity shocks ($< 10$~\\kms) or both along the line of sight would be needed to explain the observed emission.  The cooling by \\Hd~ in the warm component of GC clouds is comparable to the cooling by CO. Equating the \\Hd~ cooling rate with the heating rate by dissipation of supersonic turbulence, one finds that this mechanism could also  contribute to the emission in the two lowest \\Hd~ lines. In one source (\\ns ), we have also found some  evidence of large scale shocks that should be checked with higher spectral resolution \\Hd\\ observations."
    },
    "0010/astro-ph0010092_arXiv.txt": {
        "abstract": "We present a photometric method for identifying stars, galaxies and quasars in multi-color surveys, which uses a library of $\\ga 65000$ color templates for comparison with observed objects. The method aims for extracting the information content of object colors in a statistically correct way, and performs a classification as well as a redshift estimation for galaxies and quasars in a unified approach based on the same probability density functions. For the redshift estimation, we employ an advanced version of the Minimum Error Variance estimator which determines the redshift error from the redshift dependent probability density function itself.  The method was originally developed for the Calar Alto Deep Imaging Survey (CADIS), but is now used in a wide variety of survey projects. We checked its performance by spectroscopy of CADIS objects, where the method provides high reliability (6 errors among 151 objects with $R<24$), especially for the quasar selection, and redshifts accurate within $\\sigma_z \\approx 0.03$ for galaxies and $\\sigma_z \\approx 0.1$ for quasars.  For an optimization of future survey efforts, a few model surveys are compared, which are designed to use the same total amount of telescope time but different sets of broad-band and medium-band filters. Their performance is investigated by Monte-Carlo simulations as well as by analytic evaluation in terms of classification and redshift estimation. If photon noise were the only error source, broad-band surveys and medium-band surveys should perform equal, as long as they provide the same spectral coverage. In practice, medium-band surveys show superior performance due to their higher tolerance for calibration errors and cosmic variance.  Finally, we discuss the relevance of color calibration and derive important conclusions for the issues of library design and choice of filters. The calibration accuracy poses strong constraints on an accurate classification, which are most critical for surveys with few, broad and deeply exposed filters, but less severe for surveys with many, narrow and less deep filters. % ",
        "introduction": "Sky surveys are designed to provide statistical samples of astronomical objects, aiming for spatial overview, completeness and homogeneous datasets. Mostly they serve as a database for rather general conclusions about abundant objects, but another attractive role is allowing to search for rare and unusual objects. For both purposes, it is very useful to predict rather precisely the appearance of the different known types of objects. The object types can then be discriminated successfully, and allow to extract the information content from the survey. Also, unusual objects can be found as inconsistent with all known sorts of objects, but they might as well hide among the bulk of normal objects mimicking their appearance.  In this picture, we of course want a survey to perform as reliable and as accurate as possible in measuring object characteristics like class, redshift or physical parameters. Since surveys aim typically for large samples upon which future detailed work is based, their results are often not extremely reliable and accurate for a given single object. But for a statistical analysis of large samples, we can usually do without perfect accuracy in the measurement of features and we can also accept occasional misclassifications.  In astronomical surveys pointing off the galactic plane, obvious classes to start out with could basically be {\\it stars}, {\\it galaxies}, {\\it quasars} and {\\it strange} objects. These can be further differentiated into subclasses, based on physical characteristics derived from their morphology or spectral energy distribution (SED). Therefore, morphology and color or prominent spectral features are the typical observational criteria applied to survey data for classifying the objects contained.  Presently, surveys concentrate mostly on either imaging or spectroscopy. While spectroscopic surveys deliver a potentially high spectral resolution, they have expensive requirements for telescope time. Imaging multi-color surveys can expose a number of filters consecutively, and deliver {\\it morphological} information and crude {\\it spectral} information for all objects contained in the field of view.  Since the subject of this paper is the {\\it spectral} information in multi-color surveys, we want to mention {\\it morphological} information only briefly: The morphology is only of limited use for classifying objects into stars, galaxies and quasars: Objects observed as clearly extended are certainly not single stars, but the smaller ones could either be galaxies, low-luminosity quasars, or chance projections of more than one object. Objects consistent with point-sources can be stars, compact galaxies or quasars. Also, the morphological differentiation depends on the seeing conditions and typically reaches not to the survey limits set by the photometry.  The power of {\\it spectral} classification in a multi-color survey depends both on the filter set used and the depth of the imaging, where the optimum choices are determined by the goal of the survey. If a survey aims at identifying only one type of object with characteristic colors, a tailored filter set can be designed. E. g., when looking exclusively for U-band dropouts \\cite{Ste95}, the UGR filter set is certainly a very good choice. The performance of such a dropout survey depends mostly on the depth reached in the U-band, so the photon flux detection limit in U is the key figure. Also, number count studies are limited by the completeness limit in the filter of concern. Quasar search is very often done with color excess rules \\cite{Haz90}, where the limit is given by the flux errors combined from two or three filters. E.g., the evolution of quasars between redshift 0 and 2.2 was established using the UV excess method \\cite{SG83,BSP88}. At higher redshift quasars display rather star-like broad-band colors, motivating more advanced approaches like the selection of outliers in an n-dimensional color space \\cite{WHO91}.  If we now intend to focus different survey programs on a common patch of sky to maximise synergy effects from the various efforts, then we might as well combine the individual surveys into one that identifies every object, and avoid double work. Then we have to ask for a filter set which enables identifying virtually every object above some magnitude limit unambigously. In this case, the key number for the performance is the magnitude limit for a successful classification as needed for various science applications. If the classification takes all available color data into account, like template fitting procedures do, then the flux limit of a single filter is not the only relevant number, since the performance will depend to a large extent on the filter choice. This applies also for the estimation of multi-color redshifts, an idea dating back to Baum (1962), who used nine-band photoelectric data to estimate the redshifts of galaxy clusters.  Most multi-color surveys conducted to date obtained spectral information via broad-band photometry. They have been used e.g. to search for quasars or for high-redshift galaxies. However, they always needed follow-up spectroscopy to clarify the true nature of the candidates and to measure their redshift. The SLOAN Digital Sky Survey \\cite{York00} is now the most ambitious project to provide a broad-band color database, on which the astronomical community might perform a large number of ``virtual surveys''.  So far, only very few survey projects make extensive use of medium-band and narrow-band photometry, e.g. the Calar Alto Deep Imaging Survey \\cite{Mei98}. Surveys like CADIS with typically 10 to 20 filters are sampling the visual spectrum with a resolution comparable to that of low resolution imaging spectroscopy. CADIS fostered the development of a scheme for spectral classification, that distinguishes {\\it stars}, {\\it galaxies}, {\\it quasars} and {\\it strange objects}. Simultaneously, it assigns multi-color redshifts to extragalactic objects.  Using 162 spectroscopic identifications Wolf et al. (2000a, henceforth paper II) have shown, that it is reliable for virtually all objects above the 10-$\\sigma$ limits of the CADIS survey. Also, the photometric redshifts are accurate enough ($\\sigma_z \\approx 0.03$ for galaxies and $\\sigma_z \\approx 0.1$ for quasars around the 10-$\\sigma$ limit), so that follow-up spectroscopy is not needed for a number of analyses, e.g. the derivation of galaxy luminosity functions \\cite{Fri00}.  After this algorithm was developed for CADIS, it is now used for classification in additional projects. It provides multi-color redshifts in lensing studies of the cluster Abell 1689 \\cite{Dye00}, aiming at determining the cluster mass after identifying cluster members and weakly lensed background objects. It is also employed for an ongoing widefield survey to search for high-redshift quasars, to provide multi-color redshifts for galaxy-galaxy lensing studies, to search for high-redshift galaxy clusters and to perform a census of L* galaxies at $z \\approx 1$ \\cite{Wolf00b}.  The purpose of this paper is to present our classification scheme and discuss the optimization of its use for optimum survey strategies. The statistical algorithm for the scheme is presented in Sect. 2 and our choice for the template libraries is detailed in Sect. 3. In Sect. 4 we report on simulations of a few competitive filter sets and their expected classification performance. We include an analytic discussion on the comparison of filter sets and conclude that medium-band surveys are altogether more powerful, even when being limited by available telescope time. Sect. 5 outlines a few real datasets using this classification and draws conclusions about the expected performance. Paper II demonstrates real CADIS data based on which we gained experience during the development of the scheme, and show, that the conclusions from the simulations compare well to the real dataset. ",
        "conclusions": "\\subsection{Calibration of colors}\\label{calib_data}  Obviously, the measurements also need a careful calibration among the wavebands. A large erroneus offset can be disastrous for the photometric classification of narrow class structures in color space. If, e.g., true stars were measured with shifted colors, the classification would potentially find it rather in the location of library galaxies or quasars, and vice versa. Also, the redshift estimates would be thrown off by color offsets.  Calibration problems are of greatest concern, when rare objects are searched and their class gets contaminated. Especially, when class volumes are almost touching in color space, already small calibration errors can push objects into the wrong class. E.g., in many filter sets the quasar class is not well separated from stars and galaxies. In the presence of a calibration error, abundant galaxies can be pushed into the quasar class potentially making up the largest population among the precious candidates. The shape of class volumes is likely to cause quite some redshift dependence for the contamination. Then objects in some redshift range can become virtually unidentifiable, if they are overwhelmed by contaminants.  If calibration errors were known and quantified, they could as well be removed. If they were present but not realized, the measurements would look too accurate and a seemingly faithful classification would be derived, which is potentially wrong. Thus, as long as the calibration errors are unknown, it is still important to take their potential size into account for the error estimates on which the classification is based. As a result, the performance of the classification for bright objects is indeed limited by the calibration error.  We assume calibration errors on the order of 3\\% for the colors in our surveys, which implies that the quality of the classification saturates for objects that are more than $1\\fm5$ brighter than the 10-$\\sigma$-limits of the survey. On the other hand, if we assume for the moment poor data reduction or uncorrected galactic reddening changing the colors by, e.g., 10\\%, this would turn an entire survey catalog into a collection of ``less-than-10-$\\sigma$-objects'' --- a devastating effect for the survey quality.  An accurate relative calibration among many wavebands is best ensured by establishing a few spectrophotometric standard stars in each of the survey fields, a successful approach that we have made into a standard procedure in CADIS. This task can be carried out in a photometric night by taking spectra of the new standards and connecting these to published standard stars \\cite{Oke90}. This way, spectrophotometric standards are available in every one of the survey exposures, which will not require any further calibartion efforts regardless of the conditions under which the regular imaging is carried out. Obviously, standard star spectra are supposed to cover the entire filter set, but if a mixture of (e.g. optical and infrared) instruments is used, the calibration will involve different procedures to be matched.   \\subsection{The optimum survey strategy}  The most basic result of our study on the performance of different multi-color surveys is, that even for small systematic errors in the color indices of $s_{g-h} = 0\\fm03$, a survey with 17 bands performs better in classification and redshift estimation than one with only few bands. For the 17 band case we found that the limiting magnitude for {\\it reasonable} performance is reached when the typical statistical (i.e. photon noise) errors are on the order of 10\\%. It is obvious, that larger systematic errors will worsen the performance and will allow even higher statistical errors before the survey deteriorates significantly. For the survey strategy this implies that pushing the statistical errors in each band well below the systematic errors will add nothing to the survey performance. When $\\Delta t_{int}$ is the integration time required to reach $\\sigma \\approx 1/2 s_{g-h}$, the optimum number of bands $N$ for a given amount of total time $t_{tot}$ is roughly \\begin{eqnarray} N = t_{tot} / \\Delta t_{int} ~. \\end{eqnarray}  Although our present study has been confined to the wavelengths region attainable by optical CCDs and did not address the total wavelength coverage of the survey explicitely, it is predictable that further bands extending the wavenlength coverage (e.g. by adding NIR bands) will have a larger effect than splitting the optical bands. In particular, the maximum redshift for a reliable galaxy classification will be extended.  As the color indices are the prime observables entering the classification and redshift estimation process, it is clear that any multi-color survey has to be processed such, that these indices are measured in an optimum way. For ground-based observations it is of great importance to avoid that variable observing conditions introduce systematic offsets between bands when the observations are taken sequentially. First of all, this requires to assess the seeing point spread function on every dataset very carefully. Second, one has to correct for the effect of variable seeing which might influence the flux measurement of star-like and extended objects in a different way.  In CADIS, we essentially convolve each image to a common {\\it effective} point spread function and measure the central surface brightness of each object (see paper II for details). This has the disadvantage, that the spatial resolution (i.e. the minimum separation of objects neighboring each other) is limited by the data with the worst seeing. However, it is not clear whether the obvious alternative --- deconvolution techniques --- can be optimized such that the systematic errors can be kept below a few percent for a wide variety of objects.  The performance of the MEV estimator depends critically on the assumption that not only the color indices but also their errors are determined correctly. For the survey strategy this implies, that an optimization of the photon noise errors under the expense that an accurate estimation of these errors is no longer possible may lead to worse performance than slightly larger errors which are known accurately.   \\subsection{Ongoing applications and their scientific goals}  In this section, we want to mention examples for survey applications using this method and comment on the usefulness of our classification approach. A number of multi-color surveys have been conducted, where filters and exposure times were chosen to match some primary survey strategy. Although, none of these might have been optimal choices in terms of a general classification, we used or intend to use our approach to extract class and redshift data on the objects contained. These surveys are in chronological order of their beginning:  \\begin{enumerate} \\item The {\\it Calar Alto Deep Imaging Survey} (CADIS): Three broad-band and twelve medium-band filters have mostly been chosen to match the needs of the emission line survey in CADIS, while some of them fill in gaps in the spectral coverage. The multi-color part of CADIS has been used to study the evolution of the galaxy luminosity function at $z \\la 1$, to search for quasars at all visible redshifts and to use the observed faint stellar population to check models of the Galactic structure and the stellar luminosity function \\cite{Mei98,Wolf99,Fri00,Phl00,Wolf00a}.  \\item A lensing study of the galaxy cluster Abell 1689: Two broad-band and seven medium-band filters have been chosen to separate well between the cluster galaxies at $z \\approx 0.19$ and the background population. The galaxy luminosity function in the background of the cluster is compared to a control field taken from CADIS, and the cluster mass is estimated from weak lensing effects on the apparent luminosities \\cite{Dye00}.  \\item A widefield project for {\\it Measuring Agn redshifts by Medium-Band Observations} (MAMBO): The filters (setup A from Sect.\\,\\ref{simu_sets}) are chosen to provide a selection function and a redshift accuracy for quasars and galaxies, which is as independent of redshift as feasible. The data will be used to study the faint end of the quasar luminosity function at all accessible redshifts $z \\ga 1$ and galaxy-quasar correlation at $z \\la 1$, as well as weak lensing effects in the cluster group Abell 901/2 and in the open galaxy field \\cite{Wolf00b}.  \\item The {\\it Sloan Digital Sky Survey} (SDSS): Five broad filters have been chosen, which span the entire range of presently available CCD sensitivity. We intend to apply our classification to search for quasars and to separate stars from compact galaxies, where morphology data are not sufficient. \\end{enumerate}  From simulations of the classification scheme presented in this paper, we expect in all these projects, that we should be able to classify virtually all objects above some magnitude limit purely by color, and that especially the medium-band surveys should have selection functions which are not very dependent of redshift. This way, we can omit morphological criteria for defining catalogs of the stellar vs. galaxy vs. quasar population. This conclusion leads to a number of advantages for our method, we like to state explicitely here:  \\begin{itemize} \\item The star-galaxy separation reaches deeper and avoids confusion better than if based on morphology. Accordingly, studies of the stellar population and of the galaxy population can be extended to much fainter magnitude.  \\item Quasars can be found at all accessible redshifts, especially in the medium-band surveys, where the overlap with the stellar sequence at redshifts of $2.2 \\la z \\la 3.5$ is much reduced in the color space.  \\item Quasars do not need to be selected from a subcatalog of point sources only, allowing for objects with resolved host galaxies to be found, including high-redshift Seyfert I galaxies.  \\item We also believe that the multi-color redshifts from the medium-band surveys are accurate enough so that the applications listed above do not require follow-up spectroscopy. E.g., the evolution of the galaxy luminosity function can be analysed with the multi-color redshifts, as long as no trends in a redshift intervall $\\Delta z < 0.1$ are searched for. \\end{itemize}"
    },
    "0010/astro-ph0010571_arXiv.txt": {
        "abstract": "With the recent publication of the measurements of the radiation angular power spectrum from the BOOMERanG Antarctic flight (de Bernardis et al. 2000), it has become apparent that the currently favoured spatially-flat cold dark matter model (matter density parameter $\\Omega_{\\rm m}=0.3$, flatness being restored by a cosmological constant $\\Omega_{\\Lambda}=0.7$, Hubble parameter $h=0.65$, baryon density parameter $\\Omega_{\\rm b}h^2=0.02$) no longer provides a good fit to the data.   We describe a phenomenological approach to resurrecting this paradigm.  We consider a primordial power spectrum which incorporates a bump, arbitrarily placed at $k_{\\rm b}$, and characterized by a Gaussian in log $k$ of standard deviation $\\sigma_{\\rm b}$ and amplitude ${\\rm A}_{\\rm b}$, that is superimposed onto a scale-invariant power spectrum.  We generate a range of theoretical models that include a bump at scales consistent with cosmic microwave background and large-scale structure observations, and perform a simple $\\chi^2$ test to compare our models with the $COBE$ DMR data and the recently published BOOMERanG and MAXIMA data.  Unlike models that include a high baryon content, our models predict a low third acoustic peak.  We find that low $\\ell$ observations (20 $< \\ell <$ 200) are a critical discriminant of the bumps because the transfer function has a sharp cutoff on the high $\\ell$ side of the first acoustic peak.  Current galaxy redshift survey data suggests that excess power is required at a scale around 100 Mpc, corresponding to $k_{\\rm b}$ $\\sim$ 0.05 $h$ Mpc$^{-1}$.  For the concordance model, use of a bump-like feature to account for this excess is not consistent with the constraints made from recent CMB data.  We note that models with an appropriately chosen break in the power spectrum provide an alternative model that can give distortions similar to those reported in the APM survey as well as consistency with the CMB data (Atrio-Barandela et al. 2000; Barriga et al. 2000).  We prefer however to discount the APM data in favour of the less biased decorrelated linear power spectrum recently constructed from the PSCz redshift survey (Hamilton \\& Tegmark 2000).  We show that the concordance cosmology can be resurrected using our phenomenological approach and our best-fitting model is in agreement with the PSCz observations. ",
        "introduction": "The recent BOOMERanG and MAXIMA measurements of an acoustic peak in the angular power spectrum of the cosmic microwave background (CMB) temperature at $l \\approx 200$ (de Bernardis et al. 2000; Hanany et al. 2000) has provided remarkable confirmation that the growth via gravitational instability of primordial adiabatic density fluctuations seeds large-scale structure. One consequence of the location of this peak, due to the compression of an acoustic wave on first entering the horizon of last scattering, is that the spatial geometry of the universe is flat.  However, the weakness of the second acoustic peak at $l \\approx 400$, due to the subsequent first rarefaction of the acoustic wave on the last scattering horizon, has provoked considerable speculation as to the additional freedom that could be added to the concordance cold dark matter (CDM) model (matter density parameter $\\Omega_{\\rm m}=0.3$, flatness being restored by a cosmological constant $\\Omega_{\\Lambda}=0.7$, Hubble parameter $h=0.65$, baryon density parameter $\\Omega_{\\rm b}h^2=0.02$) to accommodate such an effect.  Ideas that have been proposed include enhancement of the baryon fraction (Lange et al. 2000; White et al. 2000), a large neutrino asymmetry (Lesgourgues \\& Peloso 2000), delay of recombination (Hu \\& Peebles 2000), an admixture of a component of cosmological defects (Bouchet et al. 2000) and models employing double inflation in supergravity (Kanazawa et al. 2000).  Here we suggest a more phenomenological solution, which is motivated by suggestive, although not overwhelming, evidence from galaxy surveys that there is excess power relative to the scale-invariant $(n \\approx 1)$ fluctuation spectrum of the conventional model near 100 $h^{-1}$ Mpc. The case for excess power has not hitherto been completely convincing because one is probing the limit of current surveys. Nevertheless, several independent data sets have provided such indications (See e.g. Broadhurst et al. 1990; Einasto et al. 1997; Landy et al. 1996).  In fact the multiple inflationary model of Adams, Ross \\& Sarkar (1997) predicts the suppression of the second acoustic peak through the generation of features in the primordial power spectrum from phase transitions that occur during inflation.  More generally, there are strong theoretical arguments which suggest that arbitrary features can be dialled onto the primordial power spectrum predicted by generic inflationary models (See e.g. Chung et al. 1999; Garc\\'{\\i}a-Bellido et al 1996; Lesgourges et al. 1998; Linde \\& Mukhanov 1997; Martin et al. 1999; Randall et al 1996; Starobinsky 1998).  We therefore consider a primordial power spectrum which incorporates a phenomenological bump, arbitrarily placed at $k_{\\rm b}$, and characterized by a Gaussian in log $k$ of standard deviation $\\sigma_{\\rm b}$ and amplitude ${\\rm A}_{\\rm b}$, that is superimposed onto a scale-invariant power spectrum as advocated by Silk \\& Gawiser (1998). We examine the constraint on the bump parameters for the $\\Lambda$CDM concordance model (Ostriker \\& Steinhardt 1995) ($\\Omega_{\\rm m}=0.3, \\, \\Omega_{\\Lambda}=0.7, \\, h=0.65, \\, \\Omega_{\\rm b}h^2=0.02)$ posed by the CMB data, and restrict the choice of bump parameters to the region of parameter space that is consistent with observations of large-scale power and CMB anisotropies. ",
        "conclusions": "We describe a toy model for a bump to be included in the primordial density power spectrum with the hope of reviving the standard model without resorting to revising fundamental cosmological theories.  We have confronted our theoretical models with the recent BOOMERanG and MAXIMA data and have shown that it is indeed possible to resurrect the standard model, although we are somewhat restricted with where we place our additional feature.  We find that our model, unlike models that include a high baryon content, predicts a low third acoustic peak.  There are two CMB measurements that will help to discriminate between such a bump and cosmological alternatives for suppressing the second peak.  One is of course the detection of the third peak.  In addition, although low $\\ell$ (20-100) observations have received relatively little attention and hence are currently a poor constraint on cosmological parameters, we have found that the low $\\ell$ power is potentially a critical discriminant for the possible bump feature. This  because the transfer function has a much sharper cutoff on the high $\\ell$ side of the first peak, relative to the cutoff on the low $\\ell$ side.  Current galaxy redshift survey data suggests that excess power is required at a scale around 100 Mpc, corresponding to $k_{\\rm b}$ $\\sim$ 0.05 $h$ Mpc$^{-1}$ (See e.g. Broadhurst et al. 1990; Einasto et al. 1997; Landy et al. 1996).  For the concordance paradigm, use of a bump-like feature to account for this excess is not consistent with the constraints from the CMB data.  We note that  models with an appropriately chosen break in the power spectrum provide an alternative model that can give distortions similar to those reported in the APM survey as well as consistency with the CMB data (Atrio-Barandela et al. 2000; Barriga et al. 2000).  We prefer however to discount the APM data in favour of the less biased decorrelated linear power spectrum recently constructed from the PSCz redshift survey (Hamilton \\& Tegmark 2000).  The incorporation of a bump in the primordial spectrum at a scale of $k_{\\rm b} = 0.004 \\, h$ Mpc$^{-1}$, as the CMB data prefers, is in good agreement with the PSCz power spectrum with a bias parameter of 1.07.  Future surveys such as 2DF and SDSS should be able to probe the large-scale structure power spectrum at the depths required to further test our conjecture. Large-scale velocity field data  are useful only at higher $k$ as a discriminant of bump-like features, and we will address this issue in a later paper."
    },
    "0010/astro-ph0010437_arXiv.txt": {
        "abstract": "We analyze the possibility that the spatial periodicity of 128$h^{-1}$ Mpc  in the galaxy count number observed recently in deep pencil-beam surveys could be explained  by the {\\it mere} existence of structure  with the appropriate scale in the galaxy distribution in the Universe. We simulate a universe where the distribution of galaxies has an intrinsic length scale and then investigate the probability of observing in a given direction a spatial periodicity similar to that observed in the pencil-beam surveys of the galactic polar regions. A statistical analysis shows that this probability is of the order of $10^{-8} $, a value which excludes the afore mentioned explanation. This result contrasts with the estimates based on Voronoi foam models which yield probabilities of the order $10^{-1} - 10^{-2}$. The conclusion that emerges from these studies put together with the present one, is that although they can be obtained in some models, the presence of an intrinsic scale in the structure is not sufficient to achieve reasonable probabilities for observation of the detected periodicities. ",
        "introduction": "The cosmic microwave background radiation shows a remarkable isotropic structure of the order $\\delta T/T \\approx 10^{-5}$, which is taken as an evidence for the smoothness of the mass density of the Universe on the largest scales comparable to the Hubble radius (Wilkinson 1987), $H_0^{-1} \\approx 3000h^{-1}$ Mpc ($H_0 = 100 h $ km s$^{-1}$ Mpc$^{-1}$). On the other hand, the existence of galaxies and clusters of galaxies as well as the (larger than unity) amplitude of the galaxy-galaxy correlation function (Peebles 1980) indicate that the Universe possesses a highly irregular structure on scales less than 10$h^{-1}$ Mpc.  On intermediate scales, ranging from about 30$h^{-1}$ Mpc to around 2000$h^{-1}$ Mpc, the existing data do not allow a conclusive interpretation about the mass density distribution of the Universe. Nevertheless, there are a number of different observations suggesting that the structure on these scales is quite varied. The most interesting of these observations include the voids with sizes greater than 30$h^{-1}$ Mpc observed in the distribution of bright galaxies (Kirschner et al. 1981; de Lapparent et al. 1986) the (larger than unity) amplitude of the cluster-cluster correlation function on scales up to 30$h^{-1}$ Mpc, the peculiar velocity field in our local vicinity on a scale of about 50$h^{-1}$ Mpc, and the ``Great Wall\" at a distance of approximately 100$h^{-1}$ Mpc (Geller \\& Huchra 1980). Also,  noteworthy is the analysis of the spatial distribution of superclusters on the intermediate scales (Einasto et al. 1997). For this analysis a sample of more than 200 superclusters (systems with more than 8 clusters where the distances between nearest-neighbors among member clusters do not exceed 24$h^{-1}$ Mpc) was used, where the estimated redshift of simple members reaches up to $z=0.12$ (around 1/4 of the redshift reached in the pencil-beam galactic surveys). These three-dimensional data of the distribution of superclusters has been used to calculate the correlation function as well as the power spectrum. In the analysis of the power spectrum it was found that it is characterized by a peak with a wavelength of ($120\\pm 15)h^{-1}$ Mpc.  But probably the most puzzling discovery concerns the recent observations which show an apparent periodicity in the distribution of redshifts in pencil-beam surveys of the north and south galactic polar regions (Broadhurst et al. 1990). These surveys extend to a redshift of around $z\\approx 0.5$ in both directions, and there is a strong evidence for a periodicity of 128$h^{-1}$ Mpc which extends for over 13 periods. Although this result was considered initially as a statistical anomaly (Kaiser \\& Peacock 1991), later and more detailed studies (Kopylov et al. 1988; Mo et al. 1992; Fetisova et al. 1993; Einasto \\& Gramann 1993; Szalay et al. 1993; Einasto et al. 1997) provided new evidence for the alluded periodicity on the same scales.  If this periodicity would be established in most directions, a naive interpretation would imply that galaxies and clusters are distributed on concentric shells centered around our galaxy, clearly a blow to our cosmological conceptions. Another more plausible explanation is based upon the idea that the observed periodicity is just a visual effect induced by a periodic oscillation of the gravitational constant (Morikawa 1990; Hill, Steinhardt \\& Turner 1990; Salgado, Quevedo \\& Sudarsky 1996, 1997; Quevedo, Salgado \\& Sudarsky 1997; Sudarsky, Quevedo \\& Salgado 1998), and that the mass density of the Universe on intermediate scales remains homogeneous and isotropic as on the largest scales. This explanation requires the introduction of a time-dependent scalar field into the model (the oscillating $G$ model) which is non-minimally coupled to gravity.  In view of the remarkable conclusions that would derive from any of these explanations it seems imperative to consider whether the observations are likely to occur in the absence of these scenarios, and can be the result of chance augmented with the assumption of the presence  of structure at the appropriate scale but with an otherwise random distribution. We will study the plausibility of this latter explanation inspired on the idea,  suggested by Szalay et al. (1993), that the structure of the Universe on these scales consists of a regular network of superclusters and voids with a step size of the order of the wavelength discovered in the power spectrum. The  specific structure that is studied by Van de Weygaert (1991) and Subba Rao and Szalay (1992) resembles a collection  of honeycombs or a three-dimensional chess-board. Such structures known as Voronoi foams result in a reasonable probability for the observation of a periodicity similar to that in the actual observations.   The aim of this work is to investigate whether it is sufficient to have a universe with an intrinsic length scale in its large scale structure in order to explain the observation of the periodicity mentioned above. In particular, the present work can be seen as a test of the suggestion that a hypothetical ``built-in scale in the initial spectrum'' arising from double inflation models and supported by cosmic microwave background radiation data (Einasto, 1997), could be the explanation of the above mentioned periodicity. Therefore, we will consider a model that does not introduce any other large scale pattern besides the existence of the above mentioned characteristic scale.  Needless is to say that neither our model nor the Voronoi foam models are anything more than ``toy'' models as they lack, for example, a proven dynamical scenario for their formation starting from reasonable initial data. We want to construct a simple realization of such a model. We will do this by assuming that the galaxies are evenly distributed over the surface of three-dimensional spheres with an average radius comparable with the appropriate scale, with the spheres themselves randomly distributed in the Universe. In a universe filled by these spheres we choose a point as the ``observational site'' and perform a statistical analysis to investigate the probability of observing a periodic distribution with the characteristics of those reported in (Broadhurst et al. 1990) in a particular direction. We obtain an upper bound for the value of such probability of the order $10^{-8}$, a value that obviously excludes the alluded structure as an acceptable explanation of the observations. The assumption that the galaxies are homogeneously distributed on the surface of the spheres is not expected to affect the main result of our numerical analysis in the  direction that would invalidate this upper bound. ",
        "conclusions": ""
    },
    "0010/hep-ph0010178_arXiv.txt": {
        "abstract": "We point out that neutrino events observed at Kamiokande and IMB from SN1987A disfavor the neutrino oscillation parameters preferred by the LSND experiment.  For $\\Delta m^2 > 0$ (the light side), the electron neutrinos from the neutronization burst would be lost, while the first event at Kamiokande is quite likely to be due to an electron neutrino.  For $\\Delta m^2 < 0$ (the dark side), the average energy of the dominantly $\\bar{\\nu}_e$ events is already lower than the theoretical expectations, which would get aggravated by a complete conversion from $\\bar{\\nu}_\\mu$ to $\\bar{\\nu}_e$.  If taken seriously, the LSND data are disfavored independent of the existence of a sterile neutrino.  A possible remedy is CPT violation, which allows different mass spectra for neutrinos and anti-neutrinos and hence can accommodate atmospheric, solar and LSND data without a sterile neutrino.  If this is the case, Mini-BooNE must run in $\\bar{\\nu}$ rather than the planned $\\nu$ mode to test the LSND signal.  We speculate on a possible origin of CPT violation. ",
        "introduction": " ",
        "conclusions": ""
    },
    "0010/astro-ph0010398_arXiv.txt": {
        "abstract": "In large stars that have exhausted their nuclear fuel, the stellar core collapses to a hot and dense proto-neutron star that cools by the radiation of neutrinos and antineutrinos of all flavors. Depending on its final mass, this may become either a neutron star or a black hole.  Black hole formation may be triggered by mass accretion or a change in the high-density equation of state.  We consider the possibility that black hole formation happens when the flux of neutrinos is still measurably high.  If this occurs, then the neutrino signal from the supernova will be terminated abruptly (the transition takes $\\lesssim 0.5$ ms).  The properties and duration of the signal before the cutoff are important measures of both the physics and astrophysics of the cooling proto-neutron star.  For the event rates expected in present and proposed detectors, the cutoff will generally appear sharp, thus allowing model-independent time-of-flight mass tests for the neutrinos after the cutoff.  If black hole formation occurs relatively early, within a few ($\\sim 1$) seconds after core collapse, then the expected luminosities are of order $L_{BH} = 10^{52}$ erg/s per flavor.  In this case, the neutrino mass sensitivity can be extraordinary.  For a supernova at a distance $D = 10$ kpc, SuperKamiokande can detect a $\\bar{\\nu}_e$ mass down to 1.8 eV by comparing the arrival times of the high-energy and low-energy neutrinos in $\\bar{\\nu}_e + p \\rightarrow e^+ + n$.  This test will also measure the cutoff time, and will thus allow a mass test of $\\nu_\\mu$ and $\\nu_\\tau$ relative to $\\bar{\\nu}_e$.  Assuming that $\\nu_\\mu$ and $\\nu_\\tau$ are nearly degenerate, as suggested by the atmospheric neutrino results, masses down to about 6 eV can be probed with a proposed lead detector of mass $M_D = 4$ kton (OMNIS). Remarkably, the neutrino mass sensitivity scales as $(D/L_{BH} M_D)^{1/2}$.  Therefore, {\\it direct} sensitivity to all three neutrino masses in the interesting few-eV range is realistically possible; {\\it there are no other known techniques that have this capability}. ",
        "introduction": "In the past several years, the growing evidence for neutrino oscillations has caused a great deal of excitement over the implied nonzero neutrino masses.  Oscillation phenomena, however, are sensitive only to differences of the squared neutrino masses, and thus provide only a lower bound on the heavier mass.  Without further input, the deduced masses can be increased, and the difference of masses decreased, providing exactly the same difference of squared masses and hence the same oscillation phenomena.  It is therefore of crucial importance to experimentally measure or constrain the absolute scale of the neutrino masses.  Two indirect techniques have been proposed.  First, the sum of the neutrino masses can be constrained by cosmological arguments. The requirement of not overclosing the universe gives an upper bound of about 100 eV~\\cite{cosmo1}.  This bound may be improved by considering the effects of neutrino masses on the cosmic microwave background and the clustering of galaxies; the claimed (future, in some cases) sensitivity is about 1 -- 10 eV~\\cite{cosmo2}.  These arguments require that the other cosmological parameters are independently known and may not apply if the neutrinos decay.\\footnote{Furthermore, it has recently been shown that in scenarios with a low (MeV-scale) reheating temperature, the neutrinos may decouple without reaching equilibrium, leading to a substantially lower density than in the usual scenario; this may weaken the cosmological neutrino mass bounds by a factor of 10 or more~\\cite{lowreheat}.} Second~\\cite{Barger}, if all of the neutrino masses are connected by small measured mass-squared differences, then each mass is constrained by the limit on the electron neutrino mass from tritium beta decay, now about 3 eV~\\cite{tritium} (the direct laboratory limits on the mu and tau neutrino masses are 170 keV~\\cite{Assamagan} and 18 MeV~\\cite{Barate}, respectively).  If neutrinoless double beta decay were discovered (i.e., neutrinos were confirmed to have a Majorana character), then this could anchor the masses at an even lower value~\\cite{2betamass}; the present limit on the combination of masses measured in double beta decay is about 0.2 eV~\\cite{Baudis}. Strictly speaking, to use the arguments of Ref.~\\cite{Barger}, each oscillation signal must first be decisively confirmed, including precise measurement of the mixing parameters and identification of the oscillated flavors.  Until then, we must allow for the possibility that there are more relevant flavors than there are measured mass-squared differences.  For example, if the solar neutrino problem is solved by $\\nu_e \\rightarrow \\nu_s$ oscillations, and if the LSND signal is ruled out, then the atmospheric neutrino problem can be solved by $\\nu_\\mu \\rightarrow \\nu_\\tau$ oscillations with a small mass difference and large masses, say 10 or 100 eV, as long as $\\delta m^2 \\simeq 10^{-3}$ eV$^2$~\\cite{SKatm}.  Thus, while the indirect constraints on the neutrino masses are valuable, it would be much more satisfying to have a direct experimental measurement.  Presently, the best possibility for direct measurement of the mu and tau neutrino masses is by time-of-flight differences using neutrinos from a Galactic core-collapse supernova. \\footnote{As noted by Shrock~\\cite{Shrock}, if neutrinos are mixed, then beta decay spectra consist of incoherent contributions from each mass eigenstate, where the endpoints depend on the masses, and the weighting on the mixing angles.  The presence of kinks in the spectrum would thus allow direct measurement of $m_2$ and $m_3$ and their mixing angles.  In order to experimentally separate such kinks from an endpoint turnover due to $m_1$, the mass differences and the mixing angles must be large enough.  For light neutrinos, the $\\bar{\\nu}_e$ disappearance experiments presently provide more restrictive limits on these parameters.} At lowest order, a neutrino with mass $m$ (in eV) and energy $E$ (in MeV) will experience an energy-dependent delay (in s) relative to a massless neutrino in traveling over a distance D (in 10 kpc): \\begin{equation} \\Delta t(E) = 0.515 \\left(\\frac{m}{E}\\right)^2 D\\,. \\label{eq:delay} \\end{equation} The distance is scaled by the approximate distance to the Galactic center, though a supernova may be detected from anywhere in the Galaxy and its immediate companions (e.g., the Magellanic Clouds). SuperKamiokande (SK) and the Sudbury Neutrino Observatory (SNO) would have good sensitivity to a Galactic supernova, collecting of order $10^4$ and $10^3$ events, respectively (see Refs.~\\cite{SKpaper,SNOpaper,JHUproc} and references therein).  Unless the decreasing neutrino luminosity is interrupted by black hole formation, it should be possible to measure it to very late times (some tens of seconds); either outcome would be an important probe of the nuclear equation of state~\\cite{Janka,Pons}.  The primary interest for mass tests is to measure the mu and tau neutrino masses relative to the nearly massless electron neutrino.  A neutrino mass test~\\cite{SKpaper,SNOpaper} based on the average event arrival times $\\langle t \\rangle$ can measure a mu or tau neutrino mass as small as 45 eV in SK and 30 eV in SNO.  If the mu and tau neutrinos are maximally mixed with nearly degenerate masses, then the sensitivity on either mass eigenstate is better by a factor of about $\\sqrt{2}$, i.e., about 30 eV in SK and 20 eV in SNO~\\cite{SKpaper,SNOpaper}.  This test is independent of supernova neutrino emission models, though it does assume that the luminosities of the different flavors have similar {\\it shapes} as a function of time, as expected on general grounds and also seen in the supernova models~\\cite{SNmodels}.  In the absence of a model, a Kolmogorov-Smirnov test can be used to compare the event rates for different flavors of neutrinos; it can be shown that this reduces to the $\\langle t \\rangle$ test~\\cite{triang}.  Other tests proposed in the literature are explicitly model-dependent, and the models have large uncertainties.  While the $\\langle t \\rangle$ test could improve the limit on the tau neutrino mass by almost six orders of magnitude, it seems very difficult to reach the eV range suggested by the cosmological and tritium arguments above.  It can be shown~\\cite{SNOpaper} that the mass sensitivity generically scales with the detector mass $M_D$ as $1/M_D^{1/4}$; therefore, another order of magnitude in sensitivity in neutrino mass would require detectors $10^4$ times larger, which seems impossible.  It can also be shown~\\cite{SNOpaper} that the sensitivity is {\\it independent} of the distance to the supernova in the case where the deduced neutrino mass is compatible with zero and only an upper limit is placed.  In this paper, a comprehensive study that follows our recent {\\it Letter}~\\cite{BHprl}, we consider the case that the proto-neutron star forms a black hole, instead of gradually cooling as a stable neutron star.  If that happens early enough, then the neutrino signals will be abruptly terminated as the neutrinospheres are enveloped by the event horizon of the black hole.  In Section II, we discuss the conditions required for this to happen and to be observable, as well as the expected details of the neutrino signal.  In Section III, we derive the mass effects on the detected neutrino event rate in the general case.  In Section IV, we show how to measure the black hole cutoff time in SuperKamiokande, with or without the complicating effects of a possible electron neutrino mass.  In Section V, we show how to make a time-of-flight mass measurement of the mu and tau neutrino masses relative to the cutoff time measured in SK.  Finally, in Section VI, we discuss some remaining issues and conclude. ",
        "conclusions": "\\subsection{Distance Dependence of the Neutrino Mass Sensitivity}  Throughout this paper, we have assumed that the next Galactic supernova will be at a distance of 10 kpc.  In the Bahcall-Soneira Galactic model~\\cite{Bahcall1,Bahcall2}, 25\\%, 50\\%, and 75\\% of supernovae are within about 7, 10, and 14 kpc of Earth, respectively. If the events during the short ($\\simeq 0.5$ ms) cutoff can be disregarded, then the results for other distances can be scaled with Eq.~(\\ref{eq:m2}), and are shown in Fig.~\\ref{fig:ddepE} and Fig.~\\ref{fig:ddepL}.  Other errors, for example the error in $N_{del}$ that comes from the small error on $t_{BH}$, are independent of $D$ in their relative importance.  A close supernova at 1 kpc would obviously have 100 times as many events as we have assumed, and would naively have mass sensitivity about 3 times better than at 10 kpc, i.e., about 2 eV in the Early case.  However, there could be a number of events during the short cutoff that would make defining $t_{BH}$ more difficult than for a more distant supernova (even assuming that the high event rate in SK does not saturate the detector).  Assuming a $\\simeq 0.5$ ms duration~\\cite{Baumgarte96b}, there could be 40 such events in SK in the Early case and about 4 in the Late case.  Note that these are estimated simply by the area of the triangle with height given by the event rate at $t_{BH}$ and width given by 0.5 ms.  In fact, the neutrino temperatures are falling rapidly during these 0.5 ms, due to increasing gravitational redshift; taking that and the detection threshold into account would reduce these numbers.  Even if $t_{BH}$ could be defined with negligible error, there could still be neutral-current events after $t_{BH}$ due to the $\\simeq 0.5$ ms duration of the cutoff: perhaps 5 events in OMNIS in the Early case and 0.5 events in the Late case. Again, these are conservatively large estimates. The presence of events during the cutoff would weaken the mass sensitivity, and it would no longer decrease with decreasing distance.  However, the real behavior of the luminosity and temperature during the cutoff is not well known, and further modeling along the lines of Baumgarte et al.~\\cite{Baumgarte96b} is needed.  For an extremely close (and hence rare) supernova, e.g., Betelgeuse at $\\sim 0.1$ kpc, the possibilities are even greater, particularly for exploring the process of black hole formation~\\cite{Baumgarte96b,Keil}, provided that the neutrino observatories can accommodate the enormous event rates.  \\begin{figure}[t] \\centerline{\\epsfxsize=3.25in \\epsfbox{ddepE.eps}} \\caption{The mass sensitivity as a function of the supernova distance (solid lines), for the Early case, for $m_{\\nu_\\mu} \\simeq m_{\\nu_\\tau}$ measured in the OMNIS detector, and for $m_{\\nu_e}$ measured in SK.  This figure is appropriate if $N_{del} = 0$ is measured and only a limit is being placed on the neutrino mass (if $N_{del} > 0$ is measured and hence a nonzero mass is discovered, see Table~I).  The dashed line is the present laboratory upper limit on $m_{\\nu_e}$~\\protect\\cite{tritium}.  In using Eq.~(\\ref{eq:m2}) to make this figure, we assumed that the events in the $\\simeq 0.5$ ms tail can be disregarded.  Depending on the unknown details of the tail, this assumption will break down at perhaps $\\sim 3$ kpc and the mass sensitivity will not improve further with decreasing distance.} \\label{fig:ddepE} \\end{figure}  \\begin{figure}[t] \\centerline{\\epsfxsize=3.25in \\epsfbox{ddepL.eps}} \\caption{As in Fig.~\\ref{fig:ddepE}, but for the Late case.  Because of the lower luminosity, the mass sensitivity may flatten out only below about $\\sim 1$ kpc.  Note the change in the vertical scale.} \\label{fig:ddepL} \\end{figure}   \\subsection{Neutrino Oscillations}  While a full discussion of neutrino oscillations is beyond the scope of this paper, we make a few brief comments.  Oscillations of $\\nu_\\mu \\leftrightarrow \\nu_\\tau$ are not important in the sense that these flavors cannot be distinguished experimentally.  The atmospheric neutrino results suggest that both are massive, with a small mass difference and a large mixing angle~\\cite{SKatm}; if so, the measured mass corresponds to the nearly degenerate mass eigenstates.  Oscillations of $\\nu_\\mu, \\nu_\\tau \\rightarrow \\nu_s$ will decrease the number of neutral-current events; this is irrelevant in the sense that the mass measurement depends on the measured, not predicted, event rate at $t_{BH}$. Oscillations of $\\nu_\\mu,\\nu_\\tau \\leftrightarrow \\nu_e$ (and their antiparticles) can in principle complicate the mass tests.  However, because of the higher temperature for $\\nu_\\mu$ and $\\nu_\\tau$, such oscillations would greatly increase the number of charged-current events and would harden the electron or positron spectrum; see, e.g., Refs.~\\cite{Fuller,SNosc}.  If evidence of such oscillations were seen, the formalism presented here could easily be enlarged to include oscillations.  The positron spectrum from $\\bar{\\nu}_e + p \\rightarrow e^+ + n$ from SN1987A appears to exclude large $\\bar{\\nu}_\\mu,\\bar{\\nu}_\\tau \\leftrightarrow \\bar{\\nu}_e$ mixing~\\cite{87Aosc}.   \\subsection{Conclusions}  If a black hole forms early in a core-collapse supernova, then the fluxes of the various flavors of neutrinos will be abruptly and simultaneously terminated when the neutrinospheres are enveloped by the event horizon.  For a massive neutrino, the cutoff in the arrival time will be delayed by $\\Delta t \\sim (m/E)^2$ relative to a massless neutrino.  The SK detector can measure both $t_{BH}$ and $m_{\\nu_e}$ by the arrival times of low- and high-energy $\\bar{\\nu}_e + p \\rightarrow e^+ + n$ events, for which the neutrino energies can be measured.  The mu and tau neutrinos are detectable only by their neutral-current interactions, in which their energies are not measured.  However, their masses can be measured by counting the number of these neutral-current events detected after $t_{BH}$.  The mass sensitivity depends on the supernova neutrino luminosity $L_{BH}$ at cutoff, the distance $D$, and the detector used.  For luminosities of $10^{52}$ erg/s per flavor at cutoff (the Early case), and a distance of 10 kpc, SK will be able to measure an electron neutrino mass as small as 1.8 eV and OMNIS would be able to measure $m_{\\nu_\\mu} \\simeq m_{\\nu_\\tau}$ as small as about 6 eV. These results are perhaps even slightly conservative, as the luminosities in Refs.~\\cite{Burrows88,Mezzacappa} were in fact a few times larger than assumed in the Early case. As discussed, the mu and tau neutrino masses were assumed to be degenerate because of the atmospheric neutrino results~\\cite{SKatm}; in this case the masses are really those of the relevant mass eigenstates.  Using the neutral-current channels in SNO and SK, the neutrino mass sensitivity is nominally $\\simeq 10$ eV for each.  However, it appears that various practical effects will degrade those results.  For other luminosities, distances, and detector masses, the mass sensitivity scales as in Eq.~(\\ref{eq:m2}), i.e., \\begin{equation} m_{lim} \\sim \\sqrt{\\frac{D}{L_{BH} M_D}}\\,. \\end{equation} This should be contrasted with the case in which the neutrino luminosities are not truncated by black hole formation, where $m_{lim} \\sim 1/M_D^{1/4}$ and is independent of $D$.  As we have discussed, there seems to be a good chance that the ongoing and proposed neutrino detectors can observe the truncation of the neutrino signals caused by black hole formation in a Galactic core-collapse supernova.  This would have profound consequences, even if no delayed events were observed and only limits were placed on the neutrino masses. Besides the obvious astrophysical importance of such an observation, this could improve the limit on the tau neutrino mass by a factor of almost $10^7$.  Moreover, the technique discussed in this paper is the {\\it only} known possibility for {\\it direct} measurement of the $\\nu_\\mu$ and $\\nu_\\tau$ masses (either Dirac or Majorana) in the crucial eV range suggested by the indirect neutrino mass tests~\\cite{cosmo1,cosmo2,Barger} discussed in the Introduction."
    },
    "0010/astro-ph0010351_arXiv.txt": {
        "abstract": "By using the comoving distance, we derive an analytic expression for the optical depth of gravitational lensing, which depends on the redshift to the source and the cosmological model characterized by the cosmic mass density parameter $\\Omega_m$, the dark energy density parameter $\\Omega_x$ and its equation of state $\\omega_x = p_x/\\rho_x$. It is shown that, the larger the dark energy density is and the more negative its pressure is, the higher the gravitational lensing probability is. This fact can provide an independent constraint for dark energy.\\\\ ",
        "introduction": "The standard cosmological model is based on three cornerstones: the Hubble expansion, the Cosmic Microwave Background Radiation(CMBR) and the primordial Big Bang Nucleosynthesis. Now these three kinds of observations, such as Hubble's relation for fifty-some Ia type supernovae(SNeIa) out to redshifts of nearly one\\cite{per98rie98}, the anisotropy of the CMBR\\cite{lin98} and the deuterium abundance measured in four high redshift hydrogen clouds seen in absorption against distant quasars\\cite{bur98}(combined with baryon fraction in galaxy clusters from X-ray data\\cite{whi93}), have made a strong case for the existence of a nearly uniform component of dark energy with negative presure. It seems that determining the amount and nature of the dark energy is emerging as one of the most important challenges in cosmology.  One apparent plausible candidate for the dark energy is the cosmological constant or vacuum energy density\\cite{wei89,kra95}. The possibility of a nonzero cosmological constant $\\Lambda$ has been advocated and then discarded several times in the past for theoretical and observational reasons\\cite{coh98sah99}. Due to this checked history and the difficulty in understanding the observed $\\Lambda$ in the framework of modern  quantum field theory\\cite{wei89}, Most physicists and astronomers believe that the cosmological constant should be zero because of some unknown physics. Other candidates for the dark energy include: a frustrated network of topological defects (such as strings or walls)\\cite{defect} and an evolving scalar field (referred to by some as quintessence)\\cite{quint}. As shown in literatures, it is difficult to discriminate well against these different possibilities either by the SNIa data alone\\cite{gar98} or only by the CMBR data\\cite{hue99}. This led some authors to consider the combination of the SNIa measurements with the anisotropy of CMBR\\cite{tur97} or the large scale structure\\cite{per99}.  In this paper, an independent means for probing the amount and nature of dark energy is proposed. It relies on the gravitational lensing statistics, which has been shown to be an efficient tool for determining the cosmological parameters\\cite{lensing}. Some authors even have given the general expressions for the optical depth and mean image seperation in general Friedman-Robertson-Walker (FRW) cosmological models\\cite{analytic}. But these expressions are complicated and thereby hard to apply in practice. By using the comoving distance we derive an analytic and simple expression for the optical depth of gravitational lensing that depends on the redshift to the source and the cosmological model characterized by the cosmic mass density parameter $\\Omega_m$, the dark energy density parameter $\\Omega_x$ and its equation of state $\\omega_x = p_x/\\rho_x$. It is shown that for a flat universe the lensing probability is very sensitive to $\\Omega_x$ and $\\omega_x$, and hence provides an independent probe for the dark energy. ",
        "conclusions": "Knowing the analytic expressions for the lensing optical depth in general FRW cosmologies, it is easy to demonstrate how the lensing probability depends sensitively on the amount and nature of dark energy. For sake of simplicity,  we now concentrate on flat universe models, which is strongly supported by various observational evidences\\cite{per98rie98,lin98,bur98} and prefered by the inflationary scenario. Fig.1 shows how the probability $p$ depends on a vacuum energy component (cosmological constant) for a flat universe. In our calculations, the source has been set at $z_s= 1, 2, 3, 4, 5$ respectively. Indeed, the probability $p$ depends sensitively on the cosmological constant, this is why astronomers take it as a promising means to determine $\\Lambda$ \\cite{lensing}.  Generally, the dark energy is parameterized by ($\\Omega_x, \\omega_x = p_x/\\rho_x$). Therefore the optical depth of gravitational lensing depends on the amount of dark energy $\\Omega_x$ as well as its equation of state $\\omega_x$. The larger the dark energy amount is, the higher the gravitational lensing probability is; the more negative the dark energy pressure is, the higher the optical depth is (see Fig.2). This fact can provide an independent constraint for dark energy, as a complement to other methods. Fig.2 is the contour plots for gravitational lensing probability (normalized to the parameter $F$) in the ($\\omega_x, \\Omega_x$) plane. Along each contour, the dark energy model is degenerated for a lensing optical depth. Fortunately, this degeneracy can be resolved when combining the lensing data with SNIa and CMBR measurements\\cite{wag99}. Thus the gravitational lensing statistics can serve as an efficient but independent tool for probing the dark energy.  In summary, by using the comoving distance we derive an analytic and simple expression for the optical depth of gravitational lensing that depends on the redshift to the source and the cosmological model characterized by the cosmic mass density $\\Omega_m$, the dark energy density $\\Omega_x$ and its equation of state $\\omega_x = p_x/\\rho_x$. It is shown that, for a flat universe, the lensing probability is very sensitive to $\\Omega_x$ and $\\omega_x$ and hence provides an independent probe for the dark energy.\\\\  This work was supported by the National Natural Science Foundation of China, under Grant No. 19903002."
    },
    "0010/astro-ph0010484_arXiv.txt": {
        "abstract": "We consider filamentary and planar large scale structures as possible refraction channels for electromagnetic radiation coming from cosmological structures. By this hypothesis, it is possible to explain the quasar luminosity distribution and, in particular, the presence of \"twin\" and \"brother\" objects. Method and details of simulation are given. ",
        "introduction": "Gravitational lensing (Straumann et al. 1998) is today considered a basic tool to investigate the large scale structure of the universe and to perform tests of cosmological models. Furthermore, it plays a relevant role in the dark matter search from small to large scales. On the other hand, the lensing effects can be explained by considering the action of a weak gravitational field deflecting the light rays which come from a fixed source. In other words, it is possible to define a refraction index, connected to the Newtonian gravitational field.  However, it is worthwhile to note that, in optics, there are other kinds of effects and instruments, like optical fibres and waveguides, which use the same deflection principles: The analogy with the gravitational field can be extended also to these processes taking into account a sort of ''nonlinearity'' in gravitational lensing.  Considering the concept of ''channel'' of sister sciences, we have refraction channels, acoustic channels and several other examples where propagating waves are trapped by some structures.  In cosmology, matter distributions, able to produce gravitational potential inducing such ''channelling effects'', could explain several anomalies which are hard to connect to standard gravitational lensing models, like, for example, the existence of objects with the same spectrum and redshift, but placed in the sky at large angular distances (e.g. twins). The huge luminosity of quasars (Fan et al. 2000; Zheng et al. 2000) could be explained by using planar cosmological structures too. In fact, taking into account expanding bubbles, produced during first order phase transition, a matter density gradient is generated when they come in touch each other. Such expanding bubbles create tridimensional spatial structures like hexahedrons (e.g. honeycombs) to minimize the interstitial space. However the growth is different with respect to crystal models since in these last cases the growth is stopped in such a direction as well as a cell of the lattice touches with an other expanding crystal in the same direction. In cosmology, we have to take into account the expansion of the universe.  Our hypothesis is that the surfaces of such hexahedrons, which are contact surfaces, could be approximated by planar structures acting as channels for electromagnetic radiation.  Furthermore, as we will discuss below, also filamentary structures could act like channels too.  In principle, all primordial topological defects as cosmic strings, textures and domain walls (Vilenkin 1984; Vilenkin 1985) could be evolved so that today they can trap electromagnetic or gravitational radiation (Capozziello \\& Iovane 1998; Turner et al. 1986).  This gravitational waveguiding effect has the same physical reason of the gravitational lensing effect, which account for the light deflection given by a gravitational field acting like a refraction medium . However, there are some substantial differences: the gravitational lenses are usually compact objects. On the contrary, the density can be low in these gravitational channels; moreover the radiation is deflected by the lens, but it does not pass through the object, while the light have to travel ''inside'' the matter distribution in the case of gravitational channels.  The existence of a waveguiding effect could explain the huge luminosity of very far objects like quasars. Another effect could be the generation of twins and brothers. In other words, an observer could detect two images of the same object: The real image and the one catched by gravitational channel.  From a theoretical point of view, the matter distribution should locally produce an effective gravitational potential of the form $\\Phi \\left( r\\right) \\sim r^{2}$.  It is worthwhile to note that the specific form of $r$ (and then of the effective gravitational potential) depends on the particular symmetry of the situation. As we shall discuss below, we have $r=x$ for ''planar'' mass distributions while $r=\\sqrt{x^{2}+y^{2}}$ in the case of ''filamentary'' distributions. In both cases, we are assuming an almost constant density $% \\rho $ which is substantially different with respect to the background.  The planar or filamentary distributions, essentially connected to a gradient of matter density, could be generated at the contact surfaces among several expanding bubbles.  The paper is organized as follows. In the Sec.2, the theoretical model for gravitational channels is developed. Sec.3 is devoted to the discussion of early cosmological phase transition, topological defects and possible connections to the today observed large scale structures. In Sec. 4, we present a simulation by which we try to explain the luminosity distribution of quasars assuming \\ that a part of them has been ''twined'' because the radiation emitted from them have undergone a channelling process by intervening filamentary or planar large scale structures. The results are discussed in Sec.5. Conclusions are drawn in Sec.6. ",
        "conclusions": "\\bigskip In this paper, we propose a mechanism by which it is possible to explain two of major puzzles of quasar theory: the huge luminosity and the twin (or brother) objects. Both of them, till now, have invoked exotic explanations as the presence of a supermassive black hole inside the core radius or lensing effects, where the position and the size of intervening deflector (usually a galaxy) has to be in same sense ''fine tuned''. By our cosmological refraction channeling effects both the issues are addressed in a natural way.  The huge luminosity at far red-shift is nothing else but the effect of the channel which ''draws'' the luminosity of the source closer to the observer. In this sense, quasars could be just primordial (in same sense ''ordinary'') galaxies whose luminosity in not dispersed as $1/R^{2}$ This fact means that the source brightness will turn out to be much stronger than the brightness of analogous objects located at the same distance (i.e. at the same redshift \\textit{Z}) and the apparent energy released by the source seems anomalously large. Besides, the large angular distance separation of twins and brothers is due to the geometry of (filamentary or planar) channels without posing ''ad hoc'' lens galaxies between the source and the observer. This hypothesis, as shown, naturally explain the double peak luminosity function of quasars.  However, what we have discussed here is nothing else but a toy model. In fact, the discrete nature of matter distribution within the channels affect the light propagation (we have to take into account scattering points, dissipation, absorbtion, and dispersion in a more realistic model).  Depending on the distribution of discrete matter, we could have several places where the light could lake out.  Furthermore, the distribution of our filamentary and planar structures has to be compared to the microwave background in order to preserve the overall homogeneity and isotropy of observed very large scale structure.  All these issues deserve further studies and the comparison to the observational data."
    },
    "0010/astro-ph0010167_arXiv.txt": {
        "abstract": "We report the detection of well resolved, extended X-ray emission from the young planetary nebula \\BD\\ using the Advanced CCD Imaging Spectrometer (ACIS) aboard the Chandra X-ray Observatory.  The X-ray emission from \\BD\\ appears to lie within, but is concentrated to one side of, the interior of the shell of ionized gas seen in high-resolution optical and IR images. The relatively low X-ray temperature ($T_x \\sim 3\\times10^6$ K) and asymmetric morphology of the X-ray emission suggests that conduction fronts are present and/or mixing of shock-heated and photoionized gas has occurred and, furthermore, hints at the presence of magnetic fields. The ACIS spectrum suggests that the X-ray emitting region is enriched in the products of helium burning. Our detection of extended X-ray emission from \\BD\\ demonstrates the power and utility of Chandra imaging as applied to the study of planetary nebulae. ",
        "introduction": "Planetary nebulae (PNs) represent the last stages of evolution for stars of initial mass 1--8 $M_\\odot$. The central star that generates a PN has terminated its asymptotic giant branch (AGB) evolution, but its ejected stellar envelope has yet to disperse entirely, such that the star's newly exposed hot core (the eventual white dwarf remnant) ionizes and illuminates the envelope via UV radiation. The photoionized gas in a typical PN is far too cool to emit X-rays and the central stars are, at best, sources of very soft X-rays. However, theorists have long predicted that high-temperature X-ray emission should arise from a so-called ``hot bubble'' formed interior to the optically emitting ionized gas (e.g., Kwok, Purton, \\& Fitzgerald, 1978; Mellema \\& Frank 1995). The hot bubble is formed when the new, fast ($\\sim1000$ km s$^{-1}$) wind emanating from the white dwarf rams into the ejected red giant envelope that is coasting outward at $\\sim10$ km s$^{-1}$, in principle shock-heating the material to $>10^7$ K.  {\\it Einstein} and {\\it ROSAT} detected X-rays from more than a dozen planetaries (de Korte et al.\\ 1985; Tarafdar \\& Apparao 1988; Apparao \\& Tarafdar 1989; Zhang, Leahy, \\& Kwok 1993; Leahy, Zhang, \\& Kwok 1994; Rauch, Koeppen, \\& Werner 1994; Chu \\& Ho 1995; Hoare et al.\\ 1995; Leahy et al.\\ 1996; Chu, Chang, \\& Conway, 1997; Chu, Gruendl, \\& Conway 1998; Chu, Guerrero, \\& Gruendl 2000) but, in most cases, the X-ray spectra are quite soft ($T_x \\le 2\\times10^5$ K) and therefore clearly emanate from hot central stars rather than hot bubbles (Guerrero, Chu, \\& Gruendl 2000 and references therein). But a handful of planetaries do display rather hard spectra, in data obtained with the {\\it ROSAT} Position-Sensitive Proportional Counter (PSPC). The brightest, hardest, and therefore most convincing example among the {\\it ROSAT} PSPC detections was the compact young planetary nebula BD $+30^\\circ$ 3639 (Kreysing et al. 1992), whose high X-ray temperature ($T_x \\sim 3\\times10^6$ K) was subsequently confirmed by {\\it ASCA} (Arnaud,  Borkowski, \\& Harrington 1996). Furthermore, Leahy, Kowk, \\& Yin (2000) found the X-ray emitting region of \\BD\\  to be possibly extended in {\\it ROSAT} High Resolution Imager (HRI)  data.  BD $+30^\\circ$ 3639 is a fascinating object in many other respects as well. Its youth is apparent from the compact morphology and low ionization state of its HII region, the low temperature of its central star ($T_\\star \\sim$ 30,000 K), and its large ratio of neutral to ionized gas mass (Sahai \\& Trauger 1998; Leahy et al.\\ 2000; and references therein). Taken together these observations suggest that the onset of ionization in this nebula is a recent event. Furthermore $\\sim10^{-3}$ $M_\\odot$ of its neutral gas is concentrated in a pair of diametrically opposed ``molecular bullets'' that have been expelled at $\\sim50$ km s$^{-1}$ (Bachiller et al.\\ 2000). The presence of such highly confined molecular outflows is strongly suggestive of the presence of a binary companion to the central star (see reviews in Kastner, Soker, \\& Rappaport 2000a), and detection of these ``bullets'' called into question whether the interacting winds model was applicable to BD $+30^\\circ$ 3639 (Bachiller et al.\\ 2000).  With the launch of the {\\it Chandra X-ray Observatory}, it is possible for the first time to obtain subarcsecond images of X-ray sources and simultaneously to perform moderate resolution X-ray spectroscopy, with the Advanced CCD Imaging Spectrometer (ACIS). In this {\\it Letter} we report the unambiguous detection of extended X-ray emission from \\BD. In a second paper (Kastner, Vrtilek, \\& Soker 2000b) we will present more detailed analysis of the \\BD\\ data. ",
        "conclusions": "The Chandra image of \\BD\\ shows an extended region of X-ray emission that appears to lie within, and is concentrated to one side of, the interior of the elliptical shell of ionized gas seen in high-resolution optical and IR images (Figs.\\ 1, 2). The approximate extent of the X-ray emitting region is $5''\\times4''$, or $\\sim5000\\times4000$ AU (for an assumed distance of $\\sim1$ kpc; Kastner et al.\\ 2000b). However, the X-ray emission is quite asymmetric in comparison to the optical and infrared morphologies of BD +30 3639, with a compact ``hot spot'' apparent toward the northeast rim of the optical/IR nebulosity and little or no emission toward the southwest rim. This result confirms the viability of the interacting winds model, but shows that another ingredient is necessary. In particular, the asymmetry suggests that the hot bubble is inhomogeneous in density, temperature, and/or abundance. Alternatively, a spatially varying absorbing column may be responsible for the asymmetry; note the close correspondence between the outline of X-ray emission and the inner ``hole'' seen in the IR nebulosity.  The central star of \\BD\\ blows a fast wind at a velocity of $v_f = 700$ km s$^{-1}$ (Leuenhagen, Hamann, \\& Jefferyet 1996).  The shocked fast wind is expected to have a temperature of $\\sim 10^7$ K, and be distributed isotropically around the central star. However, we find here a much lower temperature (see below), and the X-ray emission is localized and asymmetric, suggesting that the X-ray emission results from cooler, clumpy gas. Such cooler and denser gas can result from mixing of the hot gas with cooler, nebular gas (Chu \\& Ho 1995; Arnaud et al.\\ 1996) and/or from the presence of a heat conduction front (Soker 1994). The latter possibility suggests that magnetic fields play an important role in determining the asymmetric X-ray emission morphology of \\BD, since the presence of localized magnetic fields would suppress heat conduction, and therefore X-ray emission, in specific regions of the inner nebula (Soker 1994). More detailed analysis is postponed to our next paper (Kastner et al.\\ 2000b).  The Chandra/ACIS spectrum of BD $+30^\\circ$ 3639 (Fig.\\ 3) shows that nearly all the X-ray photons detected by ACIS have energies between 0.3 keV and 1.7 keV. The spectrum displays strong emission from a blend of He-like Ne lines centered at 915 eV, and an apparent blend of emission lines (and, possibly, continuum emission) between $\\sim300$ and $\\sim700$ eV.  Fits of a variable-abundance MEKAL (Mewe--Kaastra--Liedahl collisional equilibrium) model indicate that the abundance of C is enhanced relative to solar, Ne is roughly solar, and N and O are depleted; these fits also suggest negligible abundances for all elements heavier than Ne.  The apparent presence of significant continuum emission may indicate greatly enhanced He abundance, but this result is tentative given the modest spectral resolution of ACIS.  From the model fitting we deduce an emitting region temperature of $T_x = 2.7\\times10^6$ K ($\\pm$3\\%), intervening absorbing column of $9.9\\times10^{20}$ cm$^{-2}$ ($\\pm$4\\%), and absorbed flux of $5.7\\times10^{-13}$ ergs cm$^{-2}$ s$^{-1}$. Assuming $D=1$ kpc, the model normalization and intrinsic (unabsorbed) flux indicate, respectively, an emission measure of $5.0 \\times 10^{54}$ cm$^{-3}$ ($\\pm$12\\%) and X-ray luminosity of $L_x \\sim 1.6\\times10^{32}$ ergs s$^{-1}$.  This emission measure suggests an electron density $n_e \\sim200$ cm$^{-3}$, given the approximate diameter of the emitting region; as noted above, however, the asymmetry of the Chandra image suggests the emitting region is quite clumpy, and this result for $n_e$ should only be considered an average value. Collectively, these results are consistent with those obtained by Kreysing et al.\\ (1992) from ROSAT data and by Arnaud et al.\\ (1996) from ASCA data, although analysis of the latter dataset was complicated by contamination from a nearby supernova remnant.  The ACIS spectroscopy strongly suggests that the chemical abundances in the X-ray-emitting region are enriched in, and indeed may consist almost exclusively of, products of main sequence and post-main sequence nuclear burning. Furthermore, the large Ne abundance suggests that the nebula and white dwarf are the products of a relatively massive progenitor that obtained very high core fusion temperatures and, hence, efficient production of Ne via the reaction $^{16}$O $+$ $\\alpha$ $\\rightarrow$ $^{20}$Ne. This interpretation relies on the results for CNO abundances, however, which can only be tentatively deduced given the modest resolution of the CCD spectrum. The abundance anomalies implied by the X-ray spectrum may also hint at the central star's binary nature (Soker \\& Rappaport 2000)."
    },
    "0010/astro-ph0010578.txt": {
        "abstract": "Fourier transform power spectra of the distribution of neutral hydrogen emission in the Large Magellanic Cloud is approximately a power law over $\\sim2$ decades in length. Power spectra in the azimuthal direction look about the same as the rectilinear spectra. No difference is seen between power spectra of single channel maps and power spectra of either the peak emission map or the integrated emission map at the same location. There is a slight steeping of the average 1D and 2D LMC power spectra at high spatial frequencies. Delta-variance methods also show the same power-law structure.  These results suggest that most of the interstellar medium in the LMC is fractal, presumably the result of pervasive turbulence, self-gravity, and self-similar stirring. The similarity between the channel and integrated maps suggests they cover about the same line-of-sight depth. The slight steepening of the power spectra at high spatial frequency, corresponding to wavelengths smaller than $\\sim100$ pc, could mark the transition from large-scale emission that is relatively shallow on the line of sight to small-scale emission that is relatively thick on the line of sight. Such a transition, if real, would provide a method to obtain the thickness of a face-on galactic gas layer.  To check this possibility, three dimensional fractal models are made from the inverse Fourier transform of noise with a power-law cutoff. The models are viewed in projection with a Gaussian density distribution on the line of sight to represent a face-on galaxy disk with finite disk thickness. The density structure from turbulence is simulated in the models by using a log-normal density distribution function with a scale factor dependent on the Mach number.  Additional density structure from simulated HI phase transitions is included in some models. After tuning the Mach number, galaxy thickness, and mathematical form of the phase transition, the models can be made to reproduce the observed LMC power spectra, the amplitude of the HI brightness fluctuations, and the probability distribution function for brightness.  In all cases, the HI structure arises from a relatively thin layer in the LMC; the thick part of the HI disk has little spatial structure.  The large amplitude of the observed intensity variations cannot be achieved by turbulence alone; phase transitions are required.  The character of the fractal HI structure in the LMC is viewed in another way by comparing positive and negative images of the integrated emission. For the isotropic fractal models, these two images have the same general appearance, but for the LMC they differ markedly. The HI is much more filamentary in the LMC than in an isotropic fractal, making the geometric structure of the high-emission regions qualitatively different than the geometric structure of the low-emission (intercloud) regions.  The high-emission regions are also more sharply peaked than the low-emission regions, suggesting that compressive events formed the high-emission regions, and expansion events, whether from explosions or turbulence, formed the low-emission regions.  The character of the structure is also investigated as a function of scale using unsharp masks and enlargements with four different resolutions. The circular quality of the low emission regions and the filamentary quality of the high emission regions is preserved on scales ranging from several tens to several hundreds of parsecs.  The spatial scales for sources of turbulent energy input may be illustrated by rms variations in the power spectra with position in the galaxy.  This rms decreases from $\\sim0.6$ at kpc scales to $\\sim0.25$ on $\\sim20$ pc scales. The large scale variations are probably from known supershells. The smaller scale variations could be the result of a combination of turbulent cascades from these large-scale energy inputs and additional energy sources with smaller sizes. ",
        "introduction": "Power spectra of HI in the Small Magellanic Cloud show a power law decrease with increasing wavenumber over 2.3 decades in length (Stanimirovic et al. 1999). Box-counting techniques and perimeter-area measures of the HI in M81 group galaxies indicate a scale-free behavior there too, over 1.7 decades (Westpfahl et al. 1999).  These results for whole galaxies extend to much larger scales the fractal nature of interstellar gas that was previously found in Milky Way clouds using the same methods of power spectra (Crovisier \\& Dickey 1983; Green 1993; Lazarian \\& Pogosyan 2000; Stutzki et al. 1998) and perimeter-area correlations (Falgarone, Phillips, \\& Walker 1991) . The similarity of these gas structures to the distribution of smoke particles and other passive tracers in incompressible laboratory turbulence (Sreenivasan 1991) and to the distribution of density in simulations of compressible turbulence (Elmegreen 1999; Rosolowsky et al. 1999; MacLow \\& Ossenkopf 2000; Pichardo et al. 2000) imply that turbulence is an important structuring agent for gas on a wide range of scales in a galaxy disk. Most of the galactic gas could be fractal as a result of this turbulence, producing both clouds with their power-law mass spectra (Elmegreen \\& Falgarone 1996; Stutzki et al. 1998) and the low density intercloud medium (Elmegreen 1997).  This paper applies a power spectrum analysis to HI emission from the Large Magellanic Cloud using the combined Australia Telescope Compact Array (ATCA) interferometric survey of HI by Kim et al. (1998a,b, 1999) and a single dish survey done with the Parkes Multibeam by Staveley-Smith et al. (in preparation). The combined data set is described by Kim et al. (2000). Because the HI is more structured in the LMC than the SMC as a result of supershells (Meaburn 1980, Kim et al. 1999) and other effects of star formation, we examine both local and global power spectra from one dimensional cuts in right ascension and declination, as well as power spectra in the azimuthal direction. We also determine the more conventional 2D power spectrum and Delta-variance for the whole LMC. The results are compared with power spectra in the same field from foreground Milky Way gas. A model fractal medium is used to comment on the implications of the observations.  One of the results of this work is that the structure of HI in the LMC is not a simple fractal like that obtained from fractal Brownian motion models (e.g., Stutzki et al. 1998).  The geometry of the high temperature emission is distinctly different from the geometry of the low temperature regions.  Moreover, the extremes in HI brightness are too large to be reproduced by a realistic isothermal turbulence model. Rather, the ISM looks like a self-similar superposition of shells and filaments, separated by roundish holes. The contrast between the emission regions and the holes seems to require at least two phases of gas, such as the warm and cool phases of HI.  In this respect, the LMC resembles computer simulations by Wada, Spaans, \\& Kim (2000).  These simulations reproduce the large shells and the amorphous structures that are between the shells.  Similar models and a more detailed study of the energy input from overlapping shells were made by Scalo \\& Chappell (1999a,b). ",
        "conclusions": "The structure of HI in the LMC shows the same scale-free character in two-dimensional power spectra as it does in the SMC (Stanimirovic et al. 1999), but it steepens at short wavelengths in the one and two-dimensional power spectra, and in the azimuthal power spectra, perhaps as a result of a transition to a relatively thick disk on that scale. Such steepening has been predicted for such a transition from thin to thick lines of sight (Lazarian \\& Pogosyan 2000), but its importance as a potential indicator of extragalactic disk thickness was not previously recognized.  The power spectrum of the whole galaxy resembles the power spectra of individual channel maps in the regions where the channel maps trace the velocity centroids of the emission.  This implies that individual channels contain essentially the same structure as the whole spectral line at the spatial position where the centroid of the line has the same velocity as the channel map.  Narrow velocity slices do not come from a thinner region than the whole spectral line, for example, and the small scale structure in the ISM is not limited to any particular part of the line profile.  This conclusion is based only on our study of very large regions and does not necessarily apply to all lines of sight. Expanding shells should deviate from this trend, of course, because the small near- and far-side caps should appear in different parts of the spectral line than the large projected circumference.  The character of the HI structure was also investigated using positive and negative images to compare the high and low emission regions, and using unsharp masks with four different resolutions.  On all scales between $\\sim30$ and $\\sim 400$ pc, the emission tends to be filamentary, and the regions between the emission tend to be round. This is what might be expected from a conglomerate of expansion sites, although the nature of the high pressure that drives each expansion is not indicated. For example, the high pressure could be stellar in origin, or it could come from turbulent fluctuations.  Fractal models of a face-on galaxy with a log-normal probability distribution function for local density and a Gaussian distribution of density on the line of sight reproduce the observations fairly well. The models get the power spectrum of the LMC, the peak-to-peak amplitude of the emission variations, and the probability distribution function for pixel values.  The models suggest that most of the spatial structure in the HI is in a relatively thin midplane layer, perhaps $\\sim50$ pc thick. This implies that any thicker HI layer that surrounds the cloudy disk has relatively little structure; i.e., most of the LMC clouds are in a cool component of the HI that has a relatively low velocity dispersion and a low scale height.  The models include a log-normal distribution for local density in order to simulate compressible isothermal turbulence.  The model can be tuned to the observations by adjusting the Mach number of the turbulence, which enters into the Gaussian dispersion of the log-normal, and the thickness of the disk.  The best fits require unrealistically large Mach numbers, so other models were made having additional HI structure from a simulated phase transition between warm and cool HI components.  These models could be tuned to fit the LMC data with a more realistic Mach number.  Positional fluctuations in the power spectra are slightly larger at larger scales, suggesting that significant energy is put into the interstellar medium on kiloparsec scales, probably in connection with the well-known supershells. However, the positional fluctuations are still fairly large on smaller scales, amounting to 0.2-0.3 dex, so there are either additional energy sources on small  scales, or quick and local cascades of supershell energy to smaller scales.  A power-law distribution of shell sizes (Oey \\& Clarke 1997) may provide the range of scales necessary.  The power spectra of HI emission at large scales in the LMC are slightly shallower than the power spectra of foreground Milky Way emission. Presumably the foreground emission is thick on the line of sight, relative to the largest transverse wavelength, as is the short wavelength emission from the LMC. The similarity of the power spectra for thick emission is remarkable considering the factor of $\\sim200$ difference in physical scales.  We would like to acknowledge the encouragement and advice of Dr. Agris Kalnjas at Mount Stromlo Observatory in Australia. Helpful comments on the Delta variance by Drs. Ossenkopf and Mac Low, and on the power spectra of turbulent gases by Dr. Lazarian, are also appreciated. This work was supported by NSF Grant AST-9870112 to B.G.E."
    },
    "0010/astro-ph0010217_arXiv.txt": {
        "abstract": "We present recent near-infrared detections of circumstellar disks around the two old PMS Herbig stars HD\\,141569 and HD\\,100546 obtained with the HST/NICMOS2 camera. They reveal extended structures larger than 350--400\\,AU in radius. While the HD\\,100546 disk appears as a continuous disk down to 40\\,AU, the HD\\,141569 environment seems more complex, splitted at least into two dust populations. As a convincing example, the full modeling of the disk surrounding HR\\,4796, another old PMS star, is detailed and confronted with more recent observations. ",
        "introduction": "The presence of massive disks with large gas to dust ratios around T\\,Tauri stars is now well established (see a recent review by Dutrey 1999). Most recent near-IR and millimeter resolved images reveal optically thick disks in Keplerian rotation around single or multiple systems (e.g. GG Tau, Guilloteau et al.\\,1999). The dust appears to be the direct remnant of the primordial nebula and in a theoretical point of view, Weidenschilling et al.\\,(1993) have shown that planetesimals may form during this stage.  It is also known that a large fraction of Main-Sequence (MS) stars are surrounded by material heated by their central star (see a recent review by Lagrange et al.\\,2000). Among these Vega-like stars, the $\\beta$ Pictoris circumstellar disk has been for a long time the only one imaged till recently.  Submillimetric observations resolved for the first time dust emission around a few isolated MS stars (e.g. Greaves et al.\\,1998).  Vega-like stars bear optically thin disks with small gas to dust ratios and continuously replenished in smallest grains by collision and/or evaporation among larger bodies (Backman et Paresce\\,1993).  We have also direct (e.g.  55\\,Cnc, Butler et al.\\,1997) or indirect ($\\beta$ Pictoris: e.g. Beust \\&\\ Morbidelli\\,\\,2000) evidences for planets in these systems.  To better understand the full evolutionary scenario which lead to planetary systems formation, one must answer the following questions: what are the time-scales for dissipating disks? How does it depend on the spectral type and multiplicity? What are the dynamical dominant processes? How do gas and dust couple?  What is the origin and nature of the dust?  And finally, how does it correlate with planet formation? To do so, one must study circumstellar disks of intermediate ages which represent the missing link between embedded young disks and evolved ones. Full spectral energy distributions (SEDs) give precious constraints on the chemical composition of the grains (e.g. HD\\,100546, Malfait et al.\\,1998 with ISO) but only qualitative informations on the dust distribution itself. Resolved data are needed on these objects.  In section 2, we present HST/NICMOS2 images of two Herbig (Vega-like?) stars: HD\\,141569 and HD\\,100546. We then discuss in section 3 how we can infer valuable constraints on the dust composition and distribution as well as on the dynamics of the system by combining all available data (SED and multi-$\\lambda$ images) through a consistent model. The case of HR\\,4796 will be detailed. ",
        "conclusions": "Further modeling, and especially dynamical modeling, is necessary to improve our knowledge of disks at different stages. In the case of HR\\,4796, Wyatt et al.\\,(1999), Kenyon et al.\\,(2000) and Klahr et al.\\,(2000) have proposed models either to explain the ring-shaped structure nor brightness asymmetries. Further studies of HD\\,141569 and HD\\,100546 will allow to evidence which dynamical processes are generic or particular to such old PMS stars. But actually, very few observations of these circumstellar environments are available.  As a new example, SED fitting of the HD\\,141569 disk using the radial dust distribution derived from the 1.6\\,$\\mu$m images has predicted the presence of an inner dust population (Augereau et al.\\,1999b), recently confirmed by Fisher et al.\\,(2000). \\begin{figure} \\plotfiddle{augereaujc3.ps}{90pt}{90}{55}{55}{250}{-58} \\caption{\\small Simulated thermal images of the HR\\,4796 disk convolved with the N band observed PSF from Telesco et al.\\,(2000). To allow direct comparison, the contours levels are strictly the same as those of Figure 1a from these last authors. The white crosses indicate the position of outer ring.} \\end{figure}"
    },
    "0010/astro-ph0010021_arXiv.txt": {
        "abstract": "\\noindent We have performed the first calculations to follow the evolution of all stable nuclei and their radioactive progenitors in a finely-zoned stellar model computed from the onset of central hydrogen burning through explosion as a Type II supernova. Calculations were done for 15 $M_{\\odot}$, 20 $M_{\\odot}$, and 25 $M_{\\odot}$ Pop I stars using the most recently available set of experimental and theoretical nuclear data, revised opacity tables, and taking into account mass loss due to stellar winds. Here results are presented for one 15 $M_{\\odot}$ model. ",
        "introduction": "\\begin{figure}[t] \\centerline{\\includegraphics*[bb=50 50 792 550,height=18.83pc]{pf_mod.ps}} \\caption{\\label{rau:fig} Production factors in the ejecta of a 15 $M_{\\odot}$ star relative to solar abundance vs.\\ nuclear mass number.} \\end{figure} Stars above $\\sim10$ $M_{\\odot}$ are responsible for producing most of the oxygen and heavier elements found in nature.  Numerous studies have been devoted to the evolution of such stars and their nucleosynthetic yields, e.g.,\\ \\cite{rau:WW95,rau:TNH96}. However, our knowledge of both the input data and the physical processes affecting the evolution of these stars has improved dramatically in recent years.  Thus, it became worthwhile to improve on and considerably extend the previous investigations of pre-- and post--collapse evolution and nucleosynthesis. We present the first calculation to determine, self-consistently, the complete synthesis of all stable nuclides in any model for a massive star. Due to the limited space, in this report we mainly focus on giving an outline of our investigations. For further details on the calculations and results the reader is referred to the full papers \\cite{rau:rhhw00,rau:heg00}. ",
        "conclusions": "Here we only summarize the important results for one 15 $M_{\\odot}$ Pop I star. The production factors of this non-rotating model are shown in Fig.\\ \\ref{rau:fig} as an example of our results. The model included mass loss and the rate set of \\cite{rau:HWW00} below Ne. Note, however, that though this mass is a numerically typical case of a Type II or Ib supernova, the average nucleosynthetic yield of massive stars is the result of populations of different stars each of which has its own peculiar yields which must be combined to result in a solar-like abundance pattern. Other stars and rate sets will be discussed in \\cite{rau:rhhw00}.  \\subsection{Stellar structure}  The revision of the opacity table and the introduction of mass loss generally leads to smaller helium core sizes which tend to also decrease the mass of the carbon-oxygen and the silicon core. Note, however, that the absolute values of these core masses depend on the uncertainties, in particular, of the mixing processes in the stellar interior, such as semiconvection, overshooting, and rotation.  The change in the weak rates \\cite{rau:LM00}, important after central oxygen burning, leads to a $2-3\\,\\%$ higher electron fraction per nucleon, $Y_{\\rm e}$, at the time of core collapse in the center of the star and the ``deleptonized core'' tends to comprise less mass \\cite{rau:heg00}.  More important for the core collapse supernova mechanism might be the $30-50\\,\\%$ higher densities of the new models between the region of $m=1.5-2$ $M_{\\odot}$ \\cite{rau:heg00}, which may result in a correspondingly higher ram-pressure of the infalling matter.  \\subsection{Intermediate and heavy element nucleosynthesis}  A strong secondary s-process contribution appears between iron and a mass number of $A=90$.  Above $A=100$ the s-process in our 15 $M_{\\odot}$ star is very weak, but it becomes notably stronger in stars with more massive helium cores that perform helium burning at higher entropies. Furthermore, the strength of the s-process is found to be very sensitive to the ($\\alpha$,n)--($\\alpha$,$\\gamma$) rates and branching on $^{22}$Ne which is experimentally not well determined (see also \\cite{rau:ray00}). Second only to the well-known strong dependence of the stellar structure on the $^{12}$C($\\alpha$,$\\gamma$) rate, it becomes another important candidate for further laboratory study.  The proton-rich heavy isotopes above $A=123$ can be well produced by the $\\gamma$-process occurring during implosive and explosive oxygen and neon burning. The proton-rich isotopes around $A=160$ and those between $A=100$ and $A=123$, however, are underproduced by a factor of $3$ to $4$ with respect to $^{16}$O. The isotope $^{180}$Ta seems to show a strong overproduction by the $\\gamma$-process. However, in the figure only the totally produced $^{180}$Ta is shown. The surviving yield can be brought down to a more consistent production level by accounting for the distribution between ground and isomeric states \\cite{rau:rhhw00}.  The expected r- or n-process production by the supernova shock front passing through the base of the helium shell is not significant in any of our model stars, not even at $A=130$. We observed some redistribution of isotopes at the base of the helium shell around $A=123$ but this did not show the characteristics of a typical r-process nor was it important compared to the total yield of the star.  Summarizing, we have presented the first calculation to follow the complete s-process through all phases of stellar evolution and the $\\gamma$-process in the whole star through the presupernova stage and subsequent supernova explosion. This research was supported, in part, by DOE (W-7405-ENG-48), NSF (AST 97-31569, INT-9726315), the Alexander von Humboldt Foundation (FLF-1065004), and the Swiss NSF (2124-055832.98)."
    },
    "0010/astro-ph0010203_arXiv.txt": {
        "abstract": "I review the optical and near-infrared spectral characteristics of the extremely cool dwarfs grouped under the new classification of type L. These include both very low-mass stars and brown dwarfs, and we discuss the likely temperature range. Finally, we consider the frequency of binaries amongst very low-mass dwarfs, and the implications for formation mechanisms. ",
        "introduction": " ",
        "conclusions": ""
    },
    "0010/astro-ph0010459_arXiv.txt": {
        "abstract": "s{ Can the submillimeter counts and background be produced by applying a locally derived extinction correction to the population of Lyman break galaxies? We investigate the submillimeter emission of two strongly lensed Lyman break galaxies ($\\sou{MS}{1512{+}36}$-cB58 and $\\sou{MS}{1358{+}62}$-G1) and find that the procedure that is used to predict the submillimeter emission of the Lyman break galaxy population overpredicts the observed $850\\mum$ fluxes by up to a factor of 14. This result calls for caution in applying local correlations to distant galaxies. It also shows that large extinction corrections on Lyman break galaxies should be viewed with skepticism. It is concluded that the Lyman break galaxies may contribute to the submillimeter background at the 25 to 50\\% level. The brighter submillimeter galaxies making up the rest of the background are either not detected in optical surveys, or if they are detected, their submillimeter emission cannot be reliably estimated from their rest-frame ultraviolet properties. } ",
        "introduction": " ",
        "conclusions": "These results demonstrate that attempts to produce the submillimeter counts and background based on an extinction correction applied to the LBG population~\\cite{AdelbergerSteidel00} are fraught with considerable uncertainty, and fail in the case of the two lensed LBGs discussed here. While our sample is small, the results argue against the validity of the low redshift $\\beta$-FIR/UV correlation in the case of LBGs. Even if the local correlation were valid for high redshift galaxies as well, it still would not be able to produce the brighter submillimeter galaxies. These objects have luminosities putting them in the class of the ultraluminous infrared galaxies (ULIGs), which do not follow the $\\beta$-FIR/UV correlation, as shown by recent HST-STIS data of a sample of nearby ULIGs. Since these galaxies already account for $\\sim50\\%$ of the submillimeter background, it is not possible for the LBGs to produce a dominant fraction (let alone all) of the submillimeter background. A more likely situation is that the LBGs account for 25 to 50\\% of the submillimeter background as indicated by the faint structure in the HDF at $850\\mum$,~\\cite{Peacocketal00} but that the dominant part of the submillimeter background is made by a small number of ULIGs, which are either not detected in LBG surveys, or if they are detected, cannot be reliably corrected for extinction, in the same way that local ULIGs cannot be extinction corrected based on UV data.  In summary therefore, these results support the view that the brighter $850\\mum$ galaxies making at least 50\\% of the submillimeter background, form a population which cannot be reliably reproduced by extinction corrections applied to the LBG population."
    },
    "0010/astro-ph0010173_arXiv.txt": {
        "abstract": "We present optical observations of two remarkable new AGN discovered by the Two-Micron All Sky Survey. Both are classified as QSOs based on their optical spectra, near-IR colors, and near-IR and [\\ion{O}{3}]$\\lambda$5007 luminosities, but their optical polarizations are among the highest seen for non-blazar AGN; approaching 15\\% for 2MASSI J151653.2+190048. The polarized light spectrum for each object exhibits broad Balmer emission lines, but lacks the narrow lines that are evident in the total light spectrum. This is most pronounced for the Type-1.5 object 2MASSI J165939.7+183436, where broad lines dominate only in polarized light. The polarization properties of these AGN suggest that dust near the nucleus at least partially obscures the AGN and that material probably intermixed with the narrow line-emitting gas scatters nuclear light into our line of sight. These QSOs illustrate the variety of highly polarized AGN that have been missed by traditional optical search techniques, and demonstrate that such objects are exposed by surveys in the near-IR. ",
        "introduction": "The success of orientation-based models in explaining much of the variety in the observed properties of active galactic nuclei (AGN) \\citep[see e.g.,][]{anton93,hines95,wilkes95} has underlined the importance of obscuration located around these objects. In this class of models, a dusty torus surrounds the AGN and blocks our view of the nuclear region for high-inclination lines of sight. Not only can obscuration account for the existence of so-called ``Type-2'' objects dominated by narrow emission lines, but also their tendency toward higher polarization due to scattering within the narrow-line region (NLR), the presence of broad emission lines in polarized flux spectra, and their relatively large values of $L_{\\rm IR}/L_{\\rm OPT}$.  This appreciation of orientation effects has been extremely useful in establishing the equivalence (``unification'') of several apparently disparate AGN types, but it has also called attention to possible large biases in some samples of AGN \\citep[see e.g.,][]{rowan77}. Traditional UV/optical survey techniques \\citep[e.g.,][]{sch83, mrk67} are likely to miss AGN obscured by dust located either near the nucleus or in the body of the host galaxy. Depending on the covering factor of the obscuring material, the space density of radio-quiet AGN could be a factor of two or more greater than current surveys suggest \\citep[see e.g.,][]{low88, law91,ho97}.  Infrared surveys offer the advantage of being less susceptible to extinction by dust, plus they are sensitive to the more or less isotropic re-emission of the absorbed soft X-ray--optical radiation.   This reprocessed component appears as a broad peak in the spectral energy distribution around 25--100~$\\mu$m. The discovery of strongly polarized broad lines in ``hyperluminous'' infrared galaxies (HIGs: \\citealt{kleinman88,rowan91,cutri94}) found by the {\\it Infrared Astronomical Satellite\\/} proved that obscured QSOs can be detected in the IR \\citep{wills92, hines95, goodrich96, hines99}, but {\\sl IRAS\\/} lacked the sensitivity to acquire more than a handful of such AGN. Unfortunately, no sensitive hard X-ray or mid-IR surveys are currently planned that would cover a large part of the sky and be suitable for determining the fraction of obscured-to-unobscured radio-quiet AGN. However, \\citet{cutri00} have discovered a large number of previously unknown, primarily low-$z\\/$, red AGN in the Two Micron All Sky Survey (2MASS; \\citealt{skrutskie97}) using a simple near-infrared color-selection criterion. That is, objects at $\\vert{\\rm b}\\vert > 30^{\\circ}$ with $J-K_s > 2$.  In this {\\it Letter\\/} we present optical observations of two objects from the \\citet{cutri00} sample that illustrate the variety of AGN being found by 2MASS. The high polarization induced by small-particle scattering observed for these objects confirms that near-IR surveys do indeed find obscured AGN. ",
        "conclusions": "A population of obscured AGN with high optical polarization is being revealed by 2MASS.  The importance of this new population is perhaps best illustrated by comparison to the 114 members of the PG sample, where the most strongly polarized is PG~1114+445 at only $P \\sim 2$\\% \\citep{berr90}. Like the 2MASS QSOs studied here, the emission lines of PG~1114+445 basically share the polarization of the continuum \\citep*{smith93}, so scattering of continuum and BLR light into our line of sight is also favored as the polarizing mechanism.  Yet the very modest polarization of PG~1114+445 is accompanied by optical and near-IR colors ($J-K = 1.9$; $B-K = 3.8$) that are only slightly bluer than observed for \\jft . This suggests that the optical color selection criterion chosen for the PG survey effectively marks a boundary between blue QSOs that are essentially unpolarized and redder QSOs where the scattered flux is not necessarily swamped by the direct, unpolarized light of the nucleus.  High polarization in QSOs is also associated with absorption features intrinsic to the AGN \\citep[see e.g.,][]{schmidt99}. In this vision of the AGN unification scheme, broad absorption-line regions arise through the ablation of high-velocity clouds off the surface of the obscuring torus, but these are visible -- and a BALQSO appears -- only if our line of sight to the nucleus just skims that surface \\citep*{weymann91,voit93}.  Large scattering angles for the reflected light coupled with partial obscuration of the nucleus is thought to account for the tendency of BALQSOs to be more highly polarized than for unobscured QSOs. Ultraviolet observations test the viability and constrain the geometry of these models, and the highly-polarized 2MASS AGN are an important sample in which to test these predictions. X-ray observations will also be indispensable for determining column densities and the nature of the obscuring material.  Compared with current optical samples, the 2MASS AGN sample contains many QSOs with very different optical properties, including highly polarized objects. While the obscured-to-unobscured ratio and space density of radio-quiet AGN in the local universe remain to be determined, it is already clear that 2MASS will provide much more unbiased estimates than currently exist."
    },
    "0010/astro-ph0010490_arXiv.txt": {
        "abstract": "XMM-Newton observations of the low luminosity, radio quiet quasar Markarian 205 have revealed a unique iron K emission line profile. In marked contrast to the broad and redshifted iron K line commonly seen in \\asca\\ observations of Seyfert 1 galaxies, we find that a substantial amount of the line flux in Mrk 205 occurs above the neutral line energy of 6.4~keV. Furthermore, we find that the iron line emission has two distinct components, a narrow, unresolved neutral line at 6.4 keV and a broadened line centred at 6.7 keV. We suggest that the most likely origin of the 6.7 keV line is from X-ray reflection off the surface of a highly ionised accretion disk, whilst the 6.4~keV component may arise from neutral matter distant from the black hole, quite possibly in the putative molecular torus. Overall this observation underlines the potential of \\xmm\\ for using the iron K line as a diagnostic of matter in the innermost regions of AGN.  \\begin{keywords} galaxies: active -- quasars: individual: Markarian 205 -- X-rays: galaxies \\end{keywords} ",
        "introduction": "Since iron K$\\alpha$ emission was found to be a common feature in the X-ray spectra of AGN, it has been recognised as potentially a very important probe of matter in the central nucleus. Observations with the \\ginga\\ X-ray satellite first established the iron K$\\alpha$ line (at 6.4 keV) as a dominant feature in the hard X-ray spectra of many broad-line Seyfert 1 galaxies (Pounds \\et 1990, Nandra \\& Pounds 1994). The line was often accompanied by a flattening of the X-ray continuum (the Compton scattering hump) above the line energy. A widely favoured interpretation of these spectral features, as primary hard X-rays being `reflected' by Compton-thick matter - evidence for the putative accretion disk in AGN - was strengthened when \\asca, with its higher resolution solid-state detectors, was able to resolve the iron K emission line. The line profile of the bright Seyfert 1 galaxy MCG -6-30-15 (Tanaka \\et 1995) was found to be broadened (v$\\sim0.3$c), with a strong asymmetry red-wards of 6.4 keV; the extreme width and redshift of the iron line being attributed to the strong gravity and high velocities of the matter in the innermost regions around the putative massive black hole. Subsequent observations of a sample of Seyfert 1s with \\asca\\ (Nandra \\et 1997) revealed that such line profiles were apparently common in many nearby Seyfert 1s.  In contrast, the information on the more luminous, but generally fainter, quasars has remained inconclusive. {\\it Ginga} detected iron K emission from only a few quasars (Williams \\et 1992, Lawson \\et 1997), whilst although \\asca\\ increased the number with detected iron lines (Reeves \\et 1997, Reeves \\& Turner 2000), the data were not sufficient to resolve any of the line profiles. Despite this, a divergence in iron K line properties between the low luminosity Seyfert 1 galaxies and higher luminosity quasars has become apparent, with the line energy being closer to 6.7 keV, rather than 6.4 keV, and the line strength falling with increasing luminosity (see Iwasawa \\& Taniguichi 1993, Nandra \\et 1997b). The sensitivity of the \\xmm\\ EPIC instruments now offers the exciting prospect of observing the iron K emission of quasars in detail and hence studying the flow and physical state of matter in the nuclei of these most powerful of objects.  This paper presents the discovery, by \\xmm, of an unusual iron K$\\alpha$ line profile in the quasar Markarian 205, observed during the Cal-PV (Calibration and Payload Verification) phase. Mrk 205 is a nearby ($z=0.071$), {\\it low luminosity} quasar (M$_{V}=-23$, L$_{X}=2\\times10^{44}$~erg/s, L$_{BOL}\\sim5\\times10^{45}$~erg/s; Williams \\et 1992; Reeves \\& Turner 2000) and is radio-quiet (Rush \\et 1996). It is located only 0.7' away from the nearby ($z=0.00468$) spiral galaxy NGC 4319 and is viewed through the outer disk of this galaxy (Bahcall \\et 1992, Bowen \\& Blades 1993). Mrk 205 is also unusual in that it has a very weak UV bump (McDowell \\et 1989), suggesting that the thermal disc emission is hidden in the EUV. We now present the \\xmm\\ observations of Mrk 205. Values of $H_0=50$~km~s$^{-1}$~Mpc$^{-1}$ and $ q_0 = 0.5 $ are assumed and all fit parameters are given in the quasar rest-frame. ",
        "conclusions": "Observations with \\xmm\\ have revealed a remarkable iron K line profile in the low luminosity quasar Mrk 205. The broad iron line component at 6.7 keV is inconsistent with the relativistic profile expected from the inner accretion disc (Fabian \\et 1989), unless the disc material is highly ionised. The narrow line at 6.4 keV probably arises from Compton scattering off distant cool material; we may infact be observing the reflected X-ray emission from the putative molecular torus in Mrk 205."
    },
    "0010/astro-ph0010345_arXiv.txt": {
        "abstract": "We use a hydrodynamic simulation of an inflationary Cold Dark Matter model with a cosmological constant to predict properties of the extragalactic X-ray background (XRB). We focus on emission from the intergalactic medium (IGM), with particular attention to diffuse emission from warm-hot gas that lies in relatively smooth filamentary structures between galaxies and galaxy clusters. We also include X-rays from point sources associated with galaxies in the simulation, and we make maps of the angular distribution of the emission. Although much of the X-ray luminous gas has a filamentary structure, the filaments are not evident in the simulated maps because of projection effects. In the soft (0.5-2 keV) band, our calculated mean intensity of radiation from intergalactic and cluster gas is $2.3\\times10^{-12} \\ergdeg$, $35 \\%$ of the total soft band emission. This intensity is compatible at the $\\sim 1\\sigma$ level with estimates of the unresolved soft background intensity from deep ROSAT and {\\it Chandra} measurements. Only 4\\% of the hard (2-10 keV) emission is associated with intergalactic gas. Relative to AGN flux, the IGM component of the XRB peaks at a lower redshift (median $z \\sim 0.45$) and spans a narrower redshift range, so its clustering makes an important contribution to the angular correlation function of the total emission. The clustering on the scales accessible to our simulation ($0.1-10$ arcmin) is significant, with an amplitude roughly consistent with an extrapolation of recent ROSAT results to small scales. A cross-correlation analysis of the XRB against nearby galaxies taken from a simulated redshift survey also yields a strong signal from the IGM. Our conclusions about the soft background intensity differ from those of some recent papers, which have argued that the expected emission from gas in galaxy, group, and cluster halos would exceed the observed background unless much of the gas is expelled by supernova feedback. We obtain reasonable compatibility with current observations in a simulation that incorporates cooling, star formation, and only modest feedback. A clear prediction of our model is that the unresolved portion of the soft XRB will remain mostly unresolved even as observations reach deeper point-source sensitivity. ",
        "introduction": "It is now well established that the Cosmic X-ray Background (XRB) results almost entirely from the integrated X-ray emission of many discrete sources. Deep ROSAT surveys have shown that a large fraction of the soft ($0.5-2$ keV)  XRB is produced by active galactic nuclei (AGN), i.e. quasars and Seyfert 1 galaxies (e.g.,~Hasinger et al. 1998; Schmidt et al.~1998). Until recently, the nature of the sources producing the energetically dominant, hard ($2-10$ keV) XRB was largely unknown. {\\it Chandra} has now imaged the hard (2-10 keV) X-ray sky at high resolution and resolved essentially all of the XRB in this band (down to a flux of $4 \\times 10^{-15} \\ergscm$; Mushotzky \\etal 2000; see also Brandt et al. 2000). The optical follow-up campaigns have shown that the newly discovered XRB sources include (1) optically `faint' galaxies, which may be either quasars at high redshifts or obscured AGN, and (2) apparently `normal' galaxies.  However, a non-negligible fraction of the sky brightness, particularly at soft energies, may still be contributed by diffuse emission (mainly Bremsstrahlung) from intergalactic gas. Cosmological structure formation theories have been used to estimate the radiation emitted by gas in galaxy clusters (e.g., Blanchard \\etal 1992). In the last few years, attention has also focused (Cen \\etal 1995; Ostriker \\& Cen 1996; Dav\\'{e} \\etal 2000, hereafter D00) on the large fraction of baryonic matter at low redshift predicted by these theories to  be in the form of a ``warm-hot'' ($10^{5}-10^{7}$K) intergalactic medium (WHIM). As Cen \\& Ostriker (1999, hereafter CO99) have emphasized, detection of the faint, soft-band X-ray emission from this currently unobserved gas would help reconcile measurements of the baryon density in the local universe (e.g., Fukugita, Hogan \\& Peebles 1998) with the predictions of cosmological nucleosynthesis (Burles \\& Tytler 1998). In this paper, we use a cosmological hydrodynamic simulation to predict the intensity and angular distribution of X-ray emission from intergalactic gas. Our approach includes this relatively low density filamentary component, as well as clusters and groups. We also include a simple prescription for emission from AGN in the simulation.  Hydrodynamic simulations are routinely used to examine the expected X-ray emission from galaxy clusters in a cosmological context (see, e.g., Frenk \\etal 1999 and references therein). The lower density gas in structures between clusters in such simulations is morphologically more complicated, being distributed in the usual filaments and sheets. One can ask what observational signatures of X-ray emission are expected from this gas. For example, are the filaments visible in projection  on the sky? Experience with filamentary structure in the galaxy distribution, which only stands out clearly in 3-dimensional redshift maps (e.g., de Lapparent, Geller \\& Huchra 1986), suggests that they will be difficult to detect. Nevertheless, some evidence has been found for diffuse filamentary emission in ROSAT deep pointings (Scharf \\etal 1999). Images of individual filaments made from simulations by Pierre, Bryan \\& Gastaud (2000) suggest that the {\\it XMM-Newton} telescope should be able to detect them, at least when they are considered in isolation. An alternative approach, which we undertake here, is to make simulated sky maps using the cosmological models and measure statistical properties of the X-ray flux, which can eventually be compared to observations.  Another important question is the integrated intensity of the background.  Several recent papers have claimed that the expected emission from gas in groups of galaxies or galaxy halos exceeds the observed diffuse soft X-ray background by at least an order of magnitude (Pen 1999; Wu, Fabian, \\& Nulsen 1999).  They suggest that much of the gas in these groups and galaxies halos must be expelled or inflated by supernova feedback, decreasing its density and hence its X-ray luminosity. The amount of supernova feedback required is quite extreme, approaching 1 keV per baryon. However, these these calculations are not based on self-consistent simulations including cooling and galaxy formation. CO99 have carried out a large Eulerian hydrodynamic simulations which does include these processes, incorporating supernova feedback much milder than that advocated by Pen, and Wu, Fabian \\& Nulsen. CO99 find that the intensity of the 0.7 keV background predicted by their simulation is in fact consistent with observations. In this paper, we use a different numerical technique to CO99, but also only include very modest supernova feedback It is important to see whether our simulation gives similar results for the mean XRB intensity.  The layout of the paper is as follows. In \\S2, we present the hydrodynamic simulation and cosmological parameters used. We examine the diffuse X-ray emission from the IGM in \\S3, and study the relationship between X-ray emissivity and gas density. In \\S4, we describe our procedure for associating X-ray luminosities with some galaxies in the simulation. Simulated sky maps are generated as described in \\S5, and their clustering is analyzed in \\S6. We summarize and discuss our results in \\S7. Finally, in  Appendix A, we explain how and why we account for the multiphase nature of the gas when calculating its X-ray emissivity, and carry out some tests of our procedure. ",
        "conclusions": "We have examined the distribution of X-ray emission from intergalactic gas in a hydrodynamic simulation of a CDM model dominated by a cosmological constant. We also selected a certain fraction of galaxies in the simulation to have active nuclei, and calculated the X-ray emission from these AGN using a simple model with free parameters tuned to fit observed source counts. We projected the simulation volume, including dimming and redshift evolution to make artificial sky maps.  Some of our main conclusions drawn from an analysis of these maps are that:  (a) The mean total XRB intensity in the soft ($0.5-2$ keV) band, is $6.58 \\times 10^{-12} \\ergdeg$, with $35\\%$  being generated by the diffuse IGM: clusters, filaments and groups. $50 \\%$ of this IGM emission comes from below a redshift of $z=0.45$, and the XRB coming from $z>1.5$ is almost entirely due to AGN. The mean intensity is in reasonable agreement with the observational value ($6.0 \\pm 1.5 \\time 10^{-12} \\ergdeg$). The fraction of the total due to the IGM is slightly overpredicted by the simulations, as observationally less than $\\sim 25 \\%$ has yet to be resolved into discrete sources.  (b) In the hard ($2-10$ keV) band, only $4 \\%$ of the emission comes from the IGM, with the median redshift of XRB emission being $z=1.5$. The total intensity is in agreement with observations, as is the IGM fraction ($\\lesssim 40 \\%$ observationally).  (c) Because of projection, it is difficult to see any evidence for the filamentary emission, although this structure is obvious in thin slices taken through the simulation. Diffuse emission from relatively low densities does, however, cover much of the sky, and the median soft-band intensity of map pixels is  $1.7\\times10^{-13} \\ergdeg$.  (d) The clustering measured from the maps is strong on the small scales we are able to simulate ($<20$ arcmin), and fairly close to an extrapolation of ROSAT All Sky Survey results to these scales.  (e) The angular cross-correlation of the simulated X-ray sky with simulated bright galaxy catalogs also yields a strong signal.  The procedure that we have presented here for making XRB predictions probably represents the best that can be achieved using the hydrodynamic simulation available to us at present. It will be subject to many possible improvements once more and larger simulations can be used. One aspect that should be tested if we are to achieve more precise results is the effect of box replication on the maps. We are using the same simulation, repeated along the line of sight in the maps we have made here. This should not have any adverse effect as long as clustering on the box scale is linear and as long as the contribution of matter clustering on scales larger than the box is not significant. Our present box of size $50 \\hmpc$ is certainly smaller than one would like, although the non-linearities on the scale of the box evident in Figure \\ref{slice} are not necessarily severe, because only a tiny fraction of the box at these low redshifts contributes to the maps. In the future it will become possible to run convergence tests of box size on the results, and also use different phase simulations along the line of sight. Our results on the small scales we have concentrated on here should not change significantly. For example, Springel et al.\\ (2000) give evidence that Sunyaev-Zel'dovich map fluctuations have converged in a similar box size to ours.  Another reason to use a larger simulation volume is so that we do not under represent rare objects such as rich galaxy clusters, which will account for a large fraction of the emission. Because of this, the estimate of the XRB intensity from our relatively small simulation is more likely to be an underestimate than an overestimate. The maps we have made in this paper are not useful for studying the high luminosity tail of the emission, or individual large clusters.  Higher mass resolution would help us to represent the separate hot and cold gas phases more faithfully. This would improve the situation that results when hot IGM particles are close to clumps of cold galactic gas and are therefore assigned unrealistically high densities. We have investigated this problem and its effects on the X-ray emission in detail in Appendix A. The solution we have adopted with the present simulation is to separate the cold dense particles from the hot IGM after the simulation was run, and recalculate the densities, before calculating the X-ray emission. We have tested this technique in Appendix A to show that it is robust and gives the correct result. In the future, a different approach to calculating  the SPH densities could help, and if it were incorporated into the running of the simulation it may have some effect on galaxy formation by removing some artificially enhanced cooling.  Simulations with higher mass resolution would also be very useful to gauge the effect on the total X-ray flux. With finite resolution, one will miss small objects, and smooth out the cores of larger ones, which will yield an underestimate of the mean X-ray emissivity. That this should not be severe problem is indicated by the study of D00, who show that a substantial fraction of the X-ray emitting gas remains diffuse, with little variation between  simulations that use widely varying numerical techniques and have very different mass and spatial resolutions. We have also carried out a modest resolution test, described in \\S A.4 below. Although we have only one relatively large group at high resolution, we find that its luminosity is only $75\\%$ larger than the mean for comparable low resolution groups, even though the mass resolution is 8 times better. Lower mass groups make very little contribution to the average XRB.  The question of how much differing levels of feedback from galaxies affects the IGM X-ray emission is something we have not investigated in this paper. It has been shown in other simulations (e.g., Pierre et al.\\ 2000; Springel et al.\\ 2000) that non-gravitational heating can potentially drive the IGM out of small groups of galaxies and add to the more diffuse component. Some sort of strong feedback is also often invoked as being necessary for the observed cluster temperature-luminosity relationship to be reproduced (e.g., Kaiser 1991; Metzler \\& Evrard 1994; Bryan \\& Norman 1998). The simulation we use here includes star formation, and a mechanism for returning the energy of supernovae as thermal feedback to the IGM (see Katz \\etal 1996 for details). Because of the high spatial resolution of the SPH code, this feedback energy is deposited in very dense regions, and it is quickly radiated away, having only a minor effect on the IGM dynamics. Despite the modest role of feedback, a large fraction of gas with soft XRB temperatures is diffuse, having not yet fallen into virialized structures, and is heated by shocks. As shown by D00, this component contains a large fraction of the baryons ($\\sim 30\\%$ by mass are in this WHIM, with temperatures between $10^5$ and $10^7$ K).  In our model, as in that of CO99, much of the very soft XRB is produced by this diffuse WHIM gas, rather than by galaxy groups and halos. CO99 showed that in their simulation, one quarter (7 $\\kevkev$) of the XRB at 0.7 keV is produced by this gas. We have calculated the mean intensity in our simulations in a band centered around 0.25 keV, and found it to be 13 $\\kevkev$.  Observationally, different analyses (Cui \\etal 1996, Wang \\& Ye 1996, Warwick \\& Roberts 1998) find that after subtracting the AGN contribution in the very soft X-rays, at most $\\sim 4-15 \\kevkev$ can be in an IGM component.  This is fairly close to our simulation predictions, although probably a bit lower.  Several authors have recently argued that supernova feedback of about 1 keV per baryon is required to prevent the gas in galaxy halos and groups from over-producing the diffuse soft XRB.  Pen (1999) reached an estimate of $230 \\kevkev$ at 0.25 keV, using a hydrodynamical simulation without cooling and assuming all the baryons were gaseous with an emission-weighted temperature of $1.5 \\times 10^6 \\, {\\rm K}$, typical of galactic halos. Wu, Fabian, \\& Nulsen (1999) used a semi-analytic model to estimate a background of 35 to $60 \\kevkev$ at 0.25 keV, in the absence of supernova feedback.  This calculation allowed the gas temperature to vary as a function of group circular velocity, an improvement on the Pen (1999) calculation; the resulting background was dominated by groups of galaxies.  Our simulations, by contrast, demonstrate that extreme supernova feedback is not required to match the observed intensity of 0.25 keV emission. There are several reasons why our conclusions differ from those of Pen and Wu \\etal (see also the discussion in Dav\\'{e} \\etal 2000).  First, as mentioned above, only 30\\% of the gas is in the WHIM.  As Bryan (2000) also points out, the efficiency of forming galaxies is higher in groups than in clusters, which removes much of the gas in these lower mass systems. The papers cited above assumed that essentially all of the baryons in any virialized halo were in a hot X-ray emitting medium. Second, the gas in these papers is assumed to be isothermal in the potential of a cuspy dark matter halo, so that most of the emission is radiated at the center of the halo.  In our simulation, we find instead that the X-ray-emitting gas typically has a core with a radius of several hundred kpc (see Figure 7 of Dav\\'{e} et al.\\ 2000, which is based on this simulation).  The strong emission at the center is thus eliminated.  This core may be somewhat influenced by our numerical resolution, but the analytic model by Bryan (2000) also finds such a core.  Third, in our simulations much of the energy released in gravitational clustering emerges in the form of atomic lines, principally Ly$\\alpha$, rather than in X-rays (Fardal \\etal 2000).  As our XRB predictions are slightly higher than the observational limits, it is possible that a modest increase of feedback in the simulations would improve the agreement.  We will report on a study of the X-ray luminosity function of galaxy groups, which are most sensitive to feedback effects, in future work.  On the observational side, maps of extragalactic soft X-ray emission are extremely difficult to make, mainly because of emission and absorption by our galaxy. The ROSAT All Sky Survey is dominated by galactic thermal emission over much of the sky, and even in regions close to the galactic poles, there is uncertainty in the contribution of this component to the measured X-ray fluctuations (see, e.g., Soltan \\etal 1996). It is possible that realistic theoretical maps such as those we have presented here can play a role in testing the procedures used to remove galactic foreground contamination, just as they do in the study of the Cosmic Microwave Background (e.g., Bouchet \\& Gispert 1999).  Ultimately, one can hope to study cosmology with the small scale structure of the XRB. The contribution from various species of AGN is certainly complicated, but because the AGN emission is widely spread in redshift, the clustering of the soft XRB seems to be dominated (at least in the model we have studied) by the IGM. The physical processes involved in the emission are relatively simple, although the question of feedback and the effect it will have on small scale structure must be studied. The new X-ray satellites, {\\it Chandra} and {\\it XMM-Newton}, should soon measure clustering accurately on the scales for which we have made predictions in this paper. Among the many further analyses that can be made are cross-correlations of the XRB with Sunyaev-Zel'dovich and weak-lensing maps, as well as measuring the cross-correlation with galaxies as function of redshift. We also hope that simulated skymaps such as ours will help advance methods for detecting emission from the WHIM and thus help to complete the census of local baryons.  \\bigskip"
    },
    "0010/astro-ph0010423_arXiv.txt": {
        "abstract": "{The cluster of galaxies S\\'ersic~159$-$03\\ was observed with the XMM-Newton X-ray observatory as part of the Guaranteed Time program.  X-ray spectra taken with the EPIC and RGS instruments show no evidence for the strong cooling flow derived from previous X-ray observations.  There is a significant lack of cool gas below 1.5~keV as compared to standard isobaric cooling flow models.  While the oxygen is distributed more or less uniformly over the cluster, iron shows a strong concentration in the center of the cluster, slightly offset from the brightness center but within the central cD galaxy. This points to enhanced type Ia supernova activity in the center of the cluster. There is also an elongated iron-rich structure extending to the east of the cluster, showing the inhomogeneity of the iron distribution.  Finally, the temperature drops rapidly beyond 4\\arcmin\\ from the cluster center. ",
        "introduction": "The visible mass in clusters of galaxies is dominated by hot diffuse gas.  Due to the high temperature of the gas it is predominantly visible in the X-ray band.  X-ray spectroscopy is the key tool to understand the physics of this gas and its role in the structure and evolution of the cluster.  Since the gas is extended, spatially resolved spectroscopy is required.  Until recently medium-resolution spectroscopy using Gas Scintillation Proportional Counter (GSPC) or Charge-Coupled Device (CCD) technology was limited to a spatial resolution of a few arcminutes (ASCA and BeppoSAX).  Therefore only in a handfull of nearby clusters the core could be resolved spatially.  While other instruments aboard the Einstein and Rosat satellites had better spatial resolving power, they either lacked bandwidth or spectral resolving power.  High spectral resolution data of clusters have only been obtained for the bright core of the Virgo cluster with the Einstein FPCS detector (Canizares et al. \\cite{canizares}).  With XMM-Newton it is now possible to combine high-resolution spectroscopy of moderately extended sources using the Reflection Grating Spectrometer (RGS) with high-sensitivity, medium (CCD-type) spectral resolution imaging with the European Photon Imaging Camera (EPIC) on spatial scales down to a few arcseconds.  Here we report the XMM-Newton observation of the rich cluster of galaxies \\object{S\\'ersic~159$-$03}, discovered by S\\'ersic (\\cite{sersic}), also named Abell S~1101.  The cluster shows within the central 2\\arcmin\\ a cooling flow of 230~M$_{\\sun}$/year (Allen \\& Fabian \\cite{allen}), centered on the dominant cD galaxy \\object{ESO 291$-$9}.  The only redshift measurement of the cluster ($z$=0.0564) is from this central galaxy (Maia et al.  \\cite{maia}).  Using $H_0$=50~km\\,s$^{-1}$\\,Mpc$^{-1}$ and $q_0=0.5$ we then have a luminosity distance of 343~Mpc and an angular size distance of 307~Mpc.  We adopted a galactic column density of $1.79\\times 10^{24}$~m$^{-2}$ (Dickey \\& Lockman \\cite{dickey}, using NASA's w3nH tool). ",
        "conclusions": "Our results show that the central part of S\\'ersic~159$-$03\\ shows very little temperature structure: the average temperature in the innermost part is only 10~\\% lower than the temperature of the hot gas outside the core. From imaging data with little spectral information a large mass deposition rate has been deduced in the past (e.g. 288~M$_{\\sun}$/year based upon Einstein IPC observations (White et al. \\cite{white}); 230~M$_{\\sun}$/year with the Rosat PSPC (Allen \\& Fabian \\cite{allen}). These values will be partly biased by the strong abundance gradient in the core (Fig.~\\ref{fig:figfe}). Our results with both EPIC and RGS are consistent with such a large mass deposition rate, but only if the emission measure distribution has a low-temperature cut-off around 1.4~keV. The RGS spectra give upper limits of 1~\\% of the emission measure of the hot gas at any temperature below 1.5~keV. This upper limit translates into a 0.1--10~keV luminosity of any cool gas that is less than 20~\\% of the predicted luminosity of a 230~M$_{\\sun}$/year cooling flow without a low-temperature cut-off. The absence of cool gas is similar to what has been found in hotter clusters (A~1835: Peterson et al. \\cite{peterson}; A~1795: Tamura et al. \\cite{tamura}). We refer to those papers for a further discussion of this effect.  Another interesting feature found from our high-resolution RGS spectrum is the small O/Fe ratio in the core of the cluster: 0.34$\\pm$0.17. This is significantly smaller than the O/Fe ratio of 1.0$\\pm$0.3 derived from EPIC spectral fitting in the outer parts of the cluster. The oxygen abundance is mainly determined through the strong O~VIII Ly$\\alpha$ line. This line is subject to resonance scattering. We estimate the optical depth of this line to be $\\sim$0.5, which yields a relative line flux depression of no more than 15~\\% in the core. We conclude that there is a real decrease in the O/Fe ratio towards the center of the cluster. However the absolute oxygen abundance does not differ significantly between the core (0.19$\\pm$0.07, as derived from the RGS) and the outer parts (0.19$\\pm$0.06, as derived from EPIC), it is merely the iron abundance that increases towards the center. Since oxygen is almost totally produced by type II supernovae and iron mostly by type Ia supernovae, we conclude that the core of the cluster has been relatively rich in type Ia supernovae, while the type II supernovae are more or less uniformly distributed over the cluster. The oxygen from type II supernovae may have been ejected in protogalactic winds (Larson \\& Dinerstein \\cite{larson}). The dominance of type Ia supernovae in the center may be due to the dominance of elliptical galaxies with an old stellar population in the center of the cluster. Due to ram pressure stripping these galaxies can loose their metals to the intracluster medium (Gunn \\& Gott \\cite{gunn}).  The structure seen in the Fe-L equivalent width map (Fig.~\\ref{fig:ew}) in the central cD galaxy (within 30\\arcsec\\ from the core) indicates that the iron is distributed inhomogeneously in the core. A possible explanation might be a recent merger of a gas-rich galaxy in the NE part of the cD galaxy. Iron is not expected to diffuse over large distances during the life-time of a cluster. Therefore the elongated iron-rich structure seen in the EW direction extending out to at least 3\\arcmin\\ (0.3~Mpc) towards the east probably reflects the initial star formation in the early evolution phase of the cluster. The additional amount of iron in this structure is consistent with the total amount of interstellar iron from a solar-composition galaxy with $\\sim 10^{10}$~M$_{\\sun}$\\ of gas. Rosat PSPC images show indications for an extended, elongated structure out to at least half a degree to the east of the core, with a width of a few arcminutes. The characteristic density in this structure is of the order of 100--200~m$^{-3}$, similar to the average cluster density in our outermost annulus (Fig.~\\ref{fig:figdens}). Perhaps these structures are associated with a supercluster; unfortunately the region around S\\'ersic~159$-$03\\ is poorly studied, so we cannot confirm this.  A possible cause of the strong temperature drop near the outer part of the cluster might also be associated with the transition from cluster to supercluster.  Temperature drops by a factor of 2 or more and an associated metallicity gradient, on slightly larger spatial scales than in S\\'ersic~159$-$03\\ have been found e.g.  in the A 3562/Shapley supercluster (Kull \\& B\\\"ohringer \\cite{kull}; Ettori et al.  \\cite{ettori}).  Numerical models of cluster mergers (Ricker \\cite{ricker}) show that after the merger there can be large-scale temperature gradients in the outer parts of a cluster. This is due to a post-merger accretion shock caused by gas falling back after the merging of the cores.  The temperature can drop by an order of magnitude at 10 core radii (about 5\\arcmin\\ in S\\'ersic~159$-$03), qualitatively similar to what we find."
    },
    "0010/astro-ph0010615_arXiv.txt": {
        "abstract": "We present spectral analysis of early observations of the Type~IIn supernova 1998S using the general non-local thermodynamic equilibrium atmosphere code {\\tt PHOENIX}.  We model both the underlying supernova spectrum and the overlying circumstellar interaction region and produce spectra in good agreement with observations. The early spectra are well fit by lines produced primarily in the circumstellar region itself, and later spectra are due primarily to the supernova ejecta. Intermediate spectra are affected by both regions.  A mass-loss rate of order $\\dot M \\sim 0.0001-0.001$\\msol~yr$^{-1}$ is inferred for a wind speed of $100-1000$~\\kmps. We discuss how future self-consistent models will better clarify the underlying progenitor structure. ",
        "introduction": "SN 1998S was discovered on Mar. 3 UT by Zhou Wan \\citep{IAUC6829} as part of the Beijing Astronomical Observatory (BAO) Supernova Survey \\citep{IAUC6612}.  The discovery was confirmed by the Katzman Automatic Imaging Telescope (KAIT) during the Lick Observatory Supernova Search \\citep{IAUC6627,flipper00}. SN 1998S is located in NGC 3877, a spiral galaxy classified as SA, with a heliocentric velocity of 902~\\kmps\\ \\citep{UGC73} and a Galactic extinction of $A_B=0.01$~mag \\citep{burheilred}.  Filippenko \\& Moran \\citep{IAUC6829} obtained a high-resolution spectrum of SN 1998S on Mar. 4 with the Keck-1 telescope and classified SN 1998S as a Type~II supernova (SN~II) on the basis of broad H$\\alpha$ emission superposed on a featureless continuum.  Further spectra were obtained at the Fred L.~Whipple Observatory (FLWO) \\citep{Garn98S00} and a campaign to monitor SN~1998S in the UV from the \\emph{Hubble Space Telescope} (\\emph{HST}) was mounted by the Supernova INtensive Study (SINS) team. Three epochs have been observed with \\emph{HST} --- March 16, March 30, and May 13. SN~1998S is a Type~IIn supernova \\citep[SN IIn;][]{schlegel2n90}, a classification which shows  wide variations in the spectra \\citep{filarev97}, but includes narrow lines on top of an underlying broad-line supernova spectrum. This has been taken as strong evidence that the supernova ejecta were interacting with a slow-moving circumstellar wind \\citep{leon98S00}, probably in a fashion similar to (but possibly more extreme than) that of SN~1979C and SN~1980K \\citep{lentz98Saas99,bao98s00}.   In order to get an initial understanding and to confirm the basic picture of SNe~IIn as strong circumstellar interacters, a set of parameterized models of SN 1998S was examined with the fully relativistic, NLTE, multi-purpose, expanding atmosphere code, {\\tt PHOENIX}, \\citep[cf.][and references therein]{hbjcam99}. {\\tt PHOENIX} solves the spherically symmetric radiation transport along with the NLTE rate equations and the condition of radiative equilibrium (including deviations due to time dependence of the deposition of non-thermal gamma rays).  A modification of {\\tt PHOENIX} is underway in order to treat self-consistently both the underlying supernova spectrum and the overlying circumstellar interaction region; here we treat the two regions separately. Figure~\\ref{fig:diagram} shows a schematic representation of SN~1998S.  Some of the important coupling between the two regions is not included in the calculations and therefore not all of the features observed can be expected to be reproduced by the synthetic spectra. However, our models serve to confirm the basic picture of a SN~IIn as a Type~II supernova that interacts strongly with a near-constant velocity wind.  We are able to identify important physical effects that need to be included in future simulations.  While our models are spherically symmetric, \\citet{leon98S00} have shown that the spectra of SN~1998S are significantly polarized, which could be due to asymmetry in the outermost SN ejecta, the circumstellar medium (CSM), or both. \\citet{gerardetal00} suggest that dust and CO are likely to have formed in the SN ejecta while \\citet{Fassia98S00} argue that the early dust is likely to come from the CSM. ",
        "conclusions": "We have shown that a simple model of an ordinary Type~II supernova atmosphere interacting strongly with a radiatively accelerated wind reasonably well reproduces the observed line-widths and many of the observed features in both the UV and the optical spectra. This model is robust in that it works well at both very early times and more than a month after the explosion. This confirms the general picture of SNe~IIn as being the core collapse of massive stars that have experienced a significant mass-loss epoch and thus are surrounded by a circumstellar medium with which the supernova ejecta interact. As expected from our models \\citep{lentz98Saas99}, SN~1998S has been detected about 600 days after explosion at 6~cm \\citep{IAUC7322}. Although SN~1998S is about 5 times less luminous than SN 1988Z, further monitoring of the radio light curve will be very interesting and will help determine the mass-loss rate. From the light curve \\citet{Fassia98S00} find that the mass of the ejected envelope was quite low and the wind was weaker than that of SN~1988Z. SN~1998S may well be more closely related to SN~1979C and SN~1980K.  In future work we will develop a more self-consistent model of the supernova circumstellar interaction, and will be able to constrain mass-loss rates and total mass loss which are of great interest for the theory of stellar evolution."
    },
    "0010/astro-ph0010565_arXiv.txt": {
        "abstract": "Two continuum subtracted $\\mathrm{H}\\alpha$ HST frames of \\object{M101} are used to determine the positions, angular sizes and absolute fluxes of 237 \\ion{H}{ii} regions using a semi-automated technique. From these we have constructed the luminosity and diameter distribution functions.  We repeat this process on the images after artificially reducing the linear resolution to that typically obtained with ground based imaging. We find substantial differences in the luminosity function and diameter distribution.  The measured internal properties, such as central surface brightness and radial gradient are dominated by the PSF at linear resolutions less than roughly 40\\,pc FWHM. From the ground such resolutions are currently only obtainable for the nearest galaxies.  Further support for the dominant role played by the seeing is provided by simple analytical models. We also study the clustering properties of \\ion{H}{ii} regions and their effect on the luminosity function by construction of a Minimal Spanning Tree (MST). We find evidence for two regimes of clustering of the \\ion{H}{ii} regions and diffuse emission. These intrinsic clustering properties in combination with the spatial resolution typically obtainable from ground based observations might be responsible for the break in the \\ion{H}{ii} region luminosity function which is usually found at $\\log\\!L_{\\mathrm{H}\\alpha}=38.6$\\,erg/s, suggesting two different regimes of star formation in late type spiral galaxies.  From the high resolution HST data we find a luminosity function slope of $\\alpha=-1.74\\pm0.08$.  We also observe a flattening  at luminosities $\\log\\!L<36.7$\\,erg/s.  For the diameter distribution we find a characteristic scale of $D_0=29.2$\\,pc from an exponential fit. However, a scale free power law with index $\\beta=-2.84\\pm0.16$ provides a better fit to the data. ",
        "introduction": "\\label{sec1}  As tracers of recent massive star formation, \\ion{H}{ii} regions are indispensable.  The radiation emitted by the hot gas carries the signature of the young hot stars to which it gave birth. Complementing the detailed spectroscopic work that has been done, are studies on the statistical properties of \\ion{H}{ii} regions. These global investigations are aimed at gaining an understanding of the connection between the large scale dynamics and the star formation properties of galaxies (e.g. interactions and mergers, bars and spirals, resonances and rings).  Most of this work has focused on the luminosity and diameter distribution of the regions imaged over full galactic disks in the line of $\\mathrm{H}\\alpha$.  A long standing difficulty with making such measurements has been in the identification of the regions.  At low resolutions the blending together of regions can be significant.  An initial estimate of this effect was made by Kennicutt et al. (\\cite{ken89}), who found that blending effects became important at linear resolutions less than 300\\,pc.  However, at the luminosity limits now obtainable with modern CCDs, a study of resolution effects has not been made.  For the diameter distribution of \\ion{H}{ii} regions Van den Bergh (\\cite{van81}) found  the functional form of an exponential law. Hodge (\\cite{hod87}) confirmed this form for a sample of spiral galaxies and derived a dependence of characteristic size $D_0$ on absolute magnitude of the galaxy. A wide range of of $D_0$ between 45\\, and 560\\,pc was observed for the different galaxies.  Nevertheless, in other studies a power law form fit the diameter distribution better than the exponential function (Kennicutt \\& Hodge \\cite{ken80}, Elmegreen \\& Salzer \\cite{elm99}).  Likewise, Knapen (\\cite{kna98}) found a value of $D_0$ for M100 that is 1.5 to 2.2 times smaller than found in previous determinations at lower resolutions.  The reliability of available diameter measurements is highly questionable.  A systematic study of the luminosity function (LF) of the \\ion{H}{ii} regions in disk galaxies was carried out by Kennicutt et al. (\\cite{ken89}), using a sample of 30 spiral and irregular galaxies.  They found the LF could be represented by a power law with index $\\alpha=-2\\pm0.5$. A flattening at luminosities below $\\log\\!L_{\\mathrm{H}\\alpha} = 37$\\,erg/s was found and  explained in terms of  a transition from single ionizing stars to ionizing stellar clusters (McKee \\& Williams \\cite{McK97} and Oey \\& Clarke \\cite{oey98}).  The LF slope was also found to be systematically steeper for earlier Hubble types and  the properties of regions in spiral arms differed to those between arms. Theoretical explanations for these variations were given by Oey \\& Clarke (\\cite{oey98}), who showed that evolutionary effects and maximum number of ionizing stars per cluster can account for such differences. Some galaxies also exhibited a break in the LF at about $\\log\\!L_{\\mathrm{H}\\alpha} = 39$\\,erg/s with a steeper slope at the high luminosity end.  Other surveys were carried out by Banfi et al. (\\cite{ban93}) with a sample of 22 spirals located mainly in the virgo cluster and by Delgado \\& Perez (\\cite{del97}) with 27 spiral galaxies.  In the later survey 25\\% of objects were identified as having double slope LFs.  More detailed studies have also been performed on single objects. Scowen et al.(\\cite{sco92}) have studied M101 and found an index of $\\alpha=-1.85\\pm0.05$ for a power law fit to the LF. Rand (\\cite{ran92}) examined M51, Walterbos \\& Braun (\\cite{wal92}) M31 and Wyder et al. (\\cite{wyd97}) M33. Also an example of a barred galaxy, NGC~7479, was studied by Rozas et al. (\\cite{roz99}), where evidence for a difference in the properties of the regions located in the bar from that of the overall disk were found.  Studies of the $\\mathrm{H}\\alpha$ emission in grand design spirals were carried out by Cepa \\& Beckman (\\cite{cep89}, \\cite{cep90}), Knapen et al. (\\cite{kna93}) and continued in a series of four grand design spirals by Rozas et al. (\\cite{roz96a}, \\cite{roz96b}, \\cite{roz98}). The LFs in all four cases showed a clear slope transition at $\\log\\!L_{\\mathrm{H}\\alpha}=38.6$\\,erg/s accompanied by a local maximum or ``glitch''.  In addition to the LF break, they also found a break at the same luminosity in the surface brightness gradient of the \\ion{H}{ii} regions. The gradients of less luminous regions were about constant, whereas the gradients of more luminous regions were steeper and growing with luminosity.  Yet another break has been reported at this luminosity in the central surface brightness (Beckman, priv. comm.). These features have been interpreted by Beckman et al. (\\cite{bec98}) as representing a physical change in the \\ion{H}{ii} regions at this characteristic luminosity. Specifically, they contend that a transition from ionization to density bounding occurs at $\\log\\!L_{\\mathrm{H}\\alpha}=38.6$\\,erg/s. They have also proposed the use of this feature as a standard candle.  Related studies of the diffuse ionized interstellar medium (DIG) have been carried out by Walterbos \\& Braun (\\cite{wal94}), Ferguson et al. (\\cite{fer96}) and Greenawalt et al. (\\cite{gre97}). The diffuse emission was found to be spatially associated with the \\ion{H}{ii} regions and exhibited a variety of different morphologies.  The fraction of emission coming from this diffuse component was estimated at 20--50\\% (Ferguson et al. \\cite{fer96}). Leaking photons from \\ion{H}{ii} regions, turbulent mixing and shock ionization have been suggested as possible ionization mechanisms.  This paper presents statistics on the \\ion{H}{ii} regions in M101 from HST observations.  The very high resolution of the images will allow us to investigate in detail the role played by resolution. We will find that the linear resolution required to perform reliable measurements of central surface brightness, radial gradients, and diameters is quite stringent and can be accomplished from the ground without the aid of adaptive optics for only the nearest galaxies.  The effects of blending on the luminosity function will also be found to be significant, with the diffuse emission making an important contribution.  Also examined is the intrinsic clustering characteristics of the regions and diffuse emission and the role this plays in shaping the behavior of the blending.  Evidence for a possible break at a characteristic clustering regime will be presented, which when combined with blending effects may provide a mechanism for producing the LF discontinuity along with it's relative stability in luminosity.  In Sec.~\\ref{sec2} of this paper we present the data and reduction methods. A description of the methods used to measure the properties of the \\ion{H}{ii} regions is given in Sec.~\\ref{sec3}.  The results and comparisons to some simple analytic models are then given in Sec.~\\ref{sec4}, followed in Sec.~\\ref{sec5} by a discussion of their implications. ",
        "conclusions": "\\label{sec5}  We have shown in the previous sections that the basic properties of \\ion{H}{ii} regions determined from high resolution HST frames differ significantly from even high quality ground based measurements. Only a few previous studies have had the necessary linear resolution to make reliable measurements of their statistical properties, such as size and luminosity distributions (Walterbos \\& Braun \\cite{wal92}; Wyder et al. \\cite{wyd97}).  We simulate the effects of lower resolution on our HST frames by artificially smearing the data to typical ground based resolutions. The diameter distribution from high resolution HST observations exhibits a totally different range of values compared to those determined from lower resolution data. We find a diameter distribution which is more consistent with a power law than with the often used exponential function. This puts into doubt the concept of a characteristic size for \\ion{H}{ii} regions.  The luminosity function can be fit by a single slope power law above a luminosity of $\\log\\!L=36.8$\\,erg/s.  Below this value the distribution is flat. Unfortunately, the existence of a break at higher luminosities can not be reliably determined through direct examination of the luminosity function derived from the \\object{M101} data because of the small number of regions with luminosities above the expected location of the break. In order to make such a determination, greater coverage of \\object{M101} will be needed to increase the statistics, particularly at the high luminosity end.  The LF of the artificially smeared data (Fig.~\\ref{fig06}{\\bf a}) is flatter in the mid-luminosity range. A clear identification of a high luminosity break can also not be reliably established. A close comparison of the frames at the two different resolutions has given us important insight into the influence of resolution on the measured \\ion{H}{ii} region properties.  We find that as the resolution is reduced regions which are spatially close together are blended  into a single observed ``region'', along with a significant amount of diffuse emission. This process leads to the shifting of regions from low luminosity bins to higher bins.  A luminosity of $\\log\\!L\\approx38.5$\\,erg/s appears to mark an upper limit for this luminosity shifting effect.  The blending of regions alone is not sufficient to explain this limiting luminosity, but requires the inclusion of the diffuse emission.  While still not fully understood, it appears that the intrinsic properties of the diffuse emission clouds, such as their size, surface brightness, and spatial association with the \\ion{H}{ii} regions, play an important role. Modeling of the LF with the  inclusion of  the various effects of degraded resolution (Sec.~\\ref{sec4.1}) would be helpful in determining the nature of the high luminosity break and its possible relation to the diffuse emission.  The idea that the high luminosity break is a result of the blending of \\ion{H}{ii} regions and the masquerading of diffuse regions as genuine ionization sources differs from the hypothsis that it is due to a transition in the physical state of the \\ion{H}{ii} regions from ionization bounded to density bounded (Rozas et al. \\cite{roz96a}; Beckman et al. \\cite{bec98}). Still a connection might be given by the nature of the diffuse emission. If the diffuse emission close to the \\ion{H}{ii} regions is only ionized by the nearest source, then it may be considered as being associated with the regions. When a region and its associated diffuse emission are thrown together by the effects of resolution, then the blended region may be seen as being ionization bounded when sufficient diffuse gas is included. On the other hand a region may be seen as density bounded when the amount of diffuse gas included is insufficient with respect to the luminosity of the internal ionizing source. In this scenario the amount of diffuse gas included in a blended region is controlled by the resolution and hence to some extent so would be the position of the high luminosity break in the \\ion{H}{ii} region LF.  With even good ground based data the definition of what is an \\ion{H}{ii} region is heavily influenced by the resolution. At high resolution one can see that these smeared ``regions'' really consist of a complex of associated structures, such as diffuse clouds, arclets, shells, and filaments. Hence the surrounding gas is far from being evenly distributed around the ionizing source.  With such a non-uniform topology it would be difficult to achieve a true ionization bounding, because the possibilities for a Lyman-continuum photon to escape unhindered are numerous. Such a situation is a poor representation of an idealized Stromgren sphere. It is also not clear to what degree external sources may contribute to the ionization of the diffuse component and particularly of the diffuse regions.  A direct examination of a sequence of artificially resolution degraded frames, and confirmed by the MST analysis, shows that the effects of smearing become particularly prominent around a critical linear resolution of about 40\\,pc FWHM and thereafter only worsens slowly. If the high luminosity break in the \\ion{H}{ii} region luminosity function is an artifact of resolution then this characteristic behavior may explain the stability of the break luminosity over a relatively large range of galaxy distances. Therefore the stability of the break luminosity is not unambiguous evidence against blending as a mechanism for producing the discontinuity.  Clearly, modeling of this mechanism will be required to determine its viability.  However, the importance of the diffuse emission should not be ignored and any model of the blending will need to take this into account along with the characteristics of the clustering.  The underlying reason for this behavior seems to be the clustering characteristics of the \\ion{H}{ii} regions and diffuse emission. The MST edge-length distribution exhibits a power law form with a flattening at about 25\\,pc. The length at which this transition occurs may be the primary factor influencing the location of the high luminosity break. The physical reason for the flattening at this value of edge-length is not clear, but may be related to a typical maximal fragmentation length within a single precursor gas-cloud. The power law slope would then represent the distribution of distances between the precursor gas-clouds. Another possibility could be related to two different mechanisms of star formation as described by Scowen et al. (\\cite{sco96}). Whereas the power law edge-length distribution might correspond to the fragmentation of the ISM induced by the passage of a spiral arm density wave, the flat part may represent typical spatial sizes of local subsequent star formation. In this case the patches of diffuse emission might be interpreted as expanding HII region remnants. If the break in the LF is interpreted as due to these clustering properties, its frequent appearance derived from ground-based data may suggest a similarity of two different regimes of star formation in M101 and other late-type spirals.  There is an indication that the clustering characteristics between the two observed fields differ.  While the edge-length distributions show similar slopes, the value at which  the power law flattens out differ by about 30\\,pc.  It is not clear if this difference has a physical origin or is merely statistical. Such a difference is not found when we compare the luminosity functions or diameter distributions for the different fields. The flattening of the luminosity function is found for both fields at $\\log\\!L=36.8$, which corresponds to a diameter of about 30\\,pc. If the difference is indeed physical in nature, then one must explain how the properties of the fragmentation regimes can differ without affecting the other star formation properties.  Our simple analytic model of the resolution effects on the profiles can successfully reproduce the general behavior of the surface brightness gradients seen at lower resolutions. While the profiles of the lower luminosity regions ($\\log\\!L\\la38.5$) are dominated by the point spread function, the profiles of higher luminosity regions which have spatial sizes greater than the seeing scale are expected to represent the true properties of the regions. We therefore interpret the break in the surface brightness gradient distribution as a resolution effect.  This simple model provides a  prediction on how the luminosity of the break changes with distance.  The measured central surface brightness of the resolved regions shows a large scatter and does not exhibit a well defined trend with luminosity. This is not true for the smeared regions which exhibit a nearly constant value up to $\\log\\!L=38.5$\\,erg/s. After this the central surface brightness increases steeply with increasing luminosity.  Our simple model again reproduces this behavior fairly well and predicts the transition luminosity will vary with distance.  In both cases the profile and the central brightness become seeing dominated at linear resolutions worse than 40\\,pc FWHM. Therefore this linear resolution marks the minimum resolution required for reliable determinations of region characteristics. This kind of resolution is currently only achievable from the ground for the nearest galaxies. Taking this into account the observed breaks at lower resolution do not seem to be reliable evidence for a transition from ionization bounded to density bounded \\ion{H}{ii} regions.  Concerning the question as to whether the high luminosity break in the LF can be used as a standard candle as proposed by Beckman et al. (priv. comm.), more observations will be required to validate the stability of the break luminosity with distance (resolution) and galaxy type.  A clearer understanding of its physical origin would greatly assist in such an effort. In our interpretation, the break is a result of the clustering characteristics of the \\ion{H}{ii} regions and diffuse emission. As seen in the MST analysis, the differences between the two fields we studied in \\object{M101} may be an indication that significant differences in the clustering properties exist even within a single galaxy.  Depending on the degree to which this were true, the utility of a global LF of a galaxy as a measure of its star formation properties would be compromised.  A further indication of variations in region clustering may be in the study of the barred galaxy NGC 7479 by Rozas et al. (\\cite{roz99}), who find differences between the bar \\ion{H}{ii} region population and that of the overall disk. Furthermore this galaxy shows a different behavior in comparison to the non-barred galaxies with respect to the break luminosity in the central surface brightness distribution. This could be a hint for a possible variation of \\ion{H}{ii} region clustering properties with galaxy type.  Clearly additional high resolution work is needed to both confirm and extend these results.  More sophisticated methods should be applied to the radial profiles in order to better characterize their properties. Even a rough determination of cluster age from broad band imaging would be useful for constraining the physical origin of the larger scale HI structures and their relationship to the star forming clouds.  With respect to possible further high resolution observations, obtaining better statistics is vital for making further progress on the luminosity and diameter distributions. To generalize the results, these investigations should be carried out for different galaxy types. A comparison between a barred and non-barred galaxy would be an important test of the stability of the high luminosity break as the dynamics of the two systems are quite different."
    },
    "0010/astro-ph0010086_arXiv.txt": {
        "abstract": "We construct the first X-ray selected sample of radio-loud quasars from the EMSS survey.  The X--ray selection allows us to explore the properties of radio--loud quasars 10--100 weaker than classical samples in the radio band.  There are no radio--loud quasars whose synchrotron peak reaches the UV--soft X--ray band at these (radio) flux levels (as occurs for BL Lac objects selected in the X--ray band), but they appear to have comparatively stronger optical--UV emission.  We suggest that this can be ascribed to a significant contribution from a quasi--thermal optical--UV component, emerging due to the comparatively weak non-thermal emission.  The detection of sources at low radio fluxes also shows the presence of a large population of steep spectrum quasars, and the lack of the predicted turnover in the quasar counts. ",
        "introduction": "To understand a population properties it is crucial to disentangle the intrinsic features from the selection induced ones. To this aim it is necessary to compare the characteristics of sources selected in different spectral bands. We have therefore considered radio--loud quasars -- which so far have been extracted from radio surveys -- and studied an X--ray selected sample. We are mainly interested in determining whether the relation between the Spectral Energy Distribution (SED) and the source power proposed by Ghisellini et al. (1998) and Fossati et al. (1998) -- a blazar sequence from High-peaked BL Lacs (HBL), to Low-peaked BL Lacs (LBL) to Flat Spectrum Radio quasars (FSRQ) -- % holds also for fainter X-ray selected quasars. The trend in the SED can be physically accounted for by an increase in the external radiation field along the sequence. ",
        "conclusions": ""
    },
    "0010/nucl-th0010051_arXiv.txt": {
        "abstract": "{We show that nuclei lying at the edge of stability can behave as nonlinear quantum amplifiers. A tiny change in the nucleon-nucleon interaction can trigger a much bigger change in the binding energy of these systems, relative to the few-cluster breakup threshold.}  \\begin{document} ",
        "introduction": "Recently, the structure and reactions of nuclei far from stability have received a lot of attention. Many of these systems possess interesting properties, such as, e.g., halos, skins, etc. Most of these unusual features are direct consequences of the very small binding energies of these nuclei, relative to the breakup into 2-3 clusters. We would like to show a hitherto unnoticed interesting property of these systems: at the edge of stability they may behave as nonlinear quantum amplifiers.  Lying very close to the breakup threshold, the constituent clusters of a nucleus far from stability are barely held together by the residual forces between them. As the energy moves closer to the breakup threshold, the nucleus becomes larger. The rate at which the nucleus moves toward breakup, as a function of its size, is determined by the rate at which the (attractive) residual interaction between the clusters vanishes, as a function of their distance. One may imagine situations (not necessarily in nuclear physics) where the residual interaction drops to zero roughly as fast as the few-cluster binding energy, while the size is increased. If this happens, then a tiny change in the basic interaction can cause only a similarly small response in the binding energy. If, however, the residual interaction between the clusters goes to zero much more slowly than the binding energy does, then tiny changes in the basic interactions can get substantially amplified in the binding energy \\cite{amp}. We demonstrate this effect through the example of the $0^+_2$ state of $^{12}$C. ",
        "conclusions": "The strong sensitivity of the $0^+_2$ resonance energy to small changes in the N-N interaction has a spectacular consequence on the stellar production of carbon and oxygen. Since both steps of the triple-alpha reaction are governed by narrow resonances, the triple-alpha rate can be given \\cite{Rolfs}, in a very good approximation, as \\begin{equation} \\label{3alpha} r_{3\\alpha} = 3^{\\frac{3}{2}} N_{\\alpha}^3 \\left(\\frac{2 \\pi \\hbar^2}{M_{\\alpha} k_{\\rm B} T}\\right)^3 \\frac{\\omega \\gamma}{\\hbar} \\exp \\left(-\\frac{\\varepsilon}{k_{\\rm B} T}\\right), \\end{equation} where $M_{\\alpha}$ and $N_{\\alpha}$ is the mass and the number density of the $\\alpha$ particle, respectively, $T$ is the temperature of the stellar plasma, $\\varepsilon$ is the resonance energy of the $0^+_2$ state, relative to the $3\\alpha$ threshold, and $\\omega \\gamma$ is the resonance strength. Since $r_{3\\alpha}$ depends exponentially on $\\varepsilon$, even small variations in $\\varepsilon$ can cause large changes in $r_{3\\alpha}$. We calculated the $r_{3\\alpha}$ rates for the resonance energies shown in Table \\ref{tab1}, and used these rates in a contemporary stellar model code \\cite{Helmut} to see how much the tiny changes in the N-N force can influence the synthesis of carbon and oxygen \\cite{prev}. We performed calculations for low-mass, medium-mass, and massive stars. The resulting carbon and oxygen abundances, relative to the standard abundances, are shown in Fig.\\ \\ref{fig1} as a function of the change in the strength of the MHN and MN forces. \\begin{figure}[!t] \\vspace*{-0.1cm} \\begin{center} \\leavevmode \\hbox to\\hsize{\\hss\\epsfxsize=330bp\\epsfbox{fig.ps}\\hss} \\end{center} \\vspace*{-0.7cm} \\caption[]{The change of the carbon ($\\triangle$) and oxygen ($\\Diamond$) mass abundances ($X$) through variations of the strength of the strong interaction. They are shown in panels a, b, and c for stars with masses of 20, 5, and $1.3M_\\odot$, respectively, in units of the standard values $X_{\\rm stand}$. The variations of the strength of the strong interaction are given for the two effective N-N forces MHN and MN. The dashed curves are drawn to guide the eye. } \\label{fig1} \\vspace*{-0.5cm} \\end{figure} One can easily understand the qualitative behavior of the results by noticing that a stronger force leads to a smaller $\\varepsilon$, thus to a larger $r_{3\\alpha}$, which results in a more effective triple-alpha process, relative to the $^{12}{\\rm C}(\\alpha,\\gamma){^{16}}{\\rm O}$ burning. A similar but reversed reasoning holds for a weaker force.  The really spectacular feature of Fig.\\ \\ref{fig1} is that the tiny changes in the N-N force can cause enormous changes in the carbon and oxygen abundances. At the very root of this behavior lies the nonlinear amplification phenomenon discussed above. We can say that a 0.5\\% change in the N-N force would lead to a situation where there is virtually no carbon or oxygen present, which would make carbon-based life impossible. As the strength of the N-N force is connected, through the pion mass, to the quark masses and ultimately to the vacuum expectation value of the Higgs field, our results can in principle be used to give constraints on some fundamental parameters of the Standard Model \\cite{Jeltema}."
    },
    "0010/hep-ph0010266_arXiv.txt": {
        "abstract": " ",
        "introduction": "In the scenario of inflationary cosmology, which can explain the unanswered problems left by the standard big bang scenario, all the particles at present are thought to be created in the reheating process following the exponentially inflating de Sitter phase. That process is viewed as the decay of the inflaton field, which begins to oscillate around the minimum of the inflaton potential. The inflaton itself is treated as a coherently oscillating classical field, and its oscillation is damped by its decay, whose decay products fill the universe to realize the hot universe. A several yeas ago, it was pointed out by Linde et al.\\cite{KLS1} that light bosons could be produced more drastically than the perturbative reheating, when some conditions are satisfied on their coupling to the inflaton and the amplitudes and frequency of the inflaton oscillation. This process is called {\\it preheating}, to distinguish from the perturbative reheating.\\par The preheating is nothing but particle creation in the oscillating background. Although particle creation or annihilation generally occurs in an time-dependent background, the number of created particles grows exponentially when the background is periodic in time. This is a kind of parametric resonance. This process of particle creation is far from equilibrium, and the state after that has large quantum fluctuation composed of low frequency modes, whose relaxation process to the equilibrium is also out of equilibrium. Hence there are chances for the matter-antimatter asymmetry of the universe to be generated in these eras, if $C$ and $CP$ are violated during these processes. In fact, the preheating has been argued in connection to generation of the baryon asymmetry of the universe (BAU)\\cite{Shapsh}, focusing on the fact that the low-frequency quantum fluctuations of the created particles are so large that the sphaleron processes may be enhanced. Besides this possibility, it is natural to think that the baryon number or other charges were generated during the preheating when all the particles in the universe were created. We investigate this possibility to generate a charge asymmetry in the presence of $C$ and $CP$ violation in the oscillating background.\\par We study the particle creation when $C$ and $CP$ is violated in the interaction of quantum scalar fields with the background. We extend the analysis by Linde et al.\\cite{KLS2} to a system of multicomponent complex scalar fields. The interaction with the oscillating background is approximated by the successive scattering with some potential, while the effects of $CP$ violation are evaluated perturbatively. In \\cite{KLS2}, the number density of the created particles during the preheating was expressed in terms of the wave function in the inflaton background, and the results were reproduced by the picture of successive scattering approximation. It is, however, difficult to distinguish the positive-frequency wave from the negative-frequency wave in the oscillating background, so that one cannot define a charge, to which the contributions from the antiparticles have opposite sign relative to the particles, in terms of wave functions. Then the treatment based on the successive Bogoliubov transformations will be useful. In Section~2, we generalize the field theory of a real scalar in the oscillating background studied in \\cite{KLS2} to the case of multicomponent complex scalars and formulate the descent equations for the Bogoliubov coefficients. We apply the method to a simple toy model and find some charges to be generated, in Section 3. The final section is devoted to discussions. ",
        "conclusions": "Since all the particles at present are thought to be created after the reheating process in the inflationary scenario and the process is far from equilibrium, it is natural to consider that the asymmetry between matter and antimatter was also generated in that era. We formulated charge generation during the preheating, which is viewed as a particle creation in the oscillating background.\\par If the scale factor of the FRW universe changes from $R_{\\rm preheating}$ to $R_{\\rm th}$ just after thermalization, the charge-to-entropy ratio will be given by \\begin{equation} {{j_1}\\over s} = {{8\\sqrt2 \\pi\\,m^3\\,q^{3/4}\\int_0^\\infty d\\k\\,\\k^2\\,j_{1k}}\\over {{{2\\pi^2}\\over{45}}g_{*S}T_{\\rm th}^3}} \\left({{R_{\\rm preheating}}\\over{R_{\\rm th}}}\\right)^3, \\end{equation} where $g_{*S}$ counts the degrees of freedom contributing to the entropy and $T_{\\rm th}$ is the temperature just after thermalization. So far, we have not specified what the charge is, but we expect it to be conserved after the preheating process. One may think it as the baryon or lepton number, regarding the $\\chi$-field as the scalar partner of the quarks or leptons. If the preheating occurs before the electroweak phase transition, the generated lepton number is converted to the baryon number through the sphaleron process. When the charge is the weak hypercharge, the equilibrium state after thermalization with the hypercharge excess will become baryon-rich state, just as predicted in the theory of electroweak baryogenesis. Anyway, our next subject will be to construct a realistic model, including at least two complex scalars, which admit the charge generation based on the mechanism studied here and to evaluate the baryon asymmetry produced by that model.\\par Although we have considered the coherently oscillating field as the inflaton, the formalism may be applicable to the Affleck-Dine scalar, which is coherently rotating in the internal space with a constant frequency if we neglect the damping due to the expansion of the universe. The AD scalar, which is the classical squark or a slepton field, is thought to decay perturbatively to quarks and leptons to leave $B$ or $L$. This situation is similar to that of the conventional reheating process and one may expect that there is also a process similar to the preheating in the AD mechanism."
    },
    "0010/astro-ph0010222_arXiv.txt": {
        "abstract": "The interaction rates of dark-matter halos and subhalos, including collisions and mergers, are computed using high-resolution cosmological \\nbody\\ simulations of the $\\Lambda$CDM model. Although the number fraction of subhalos of mass $>2\\times10^{11}\\hsolmass$ is only $\\sim10\\%$, we find that the interaction rate of such subhalos is relatively high because they reside in high density environments. At low redshift, the subhalo collisions dominate the total collision rate, and even at $z=3$ they are involved in more than 30\\% of all collisions. About 40\\% of the ``major\" mergers (those of mass ratio $>0.3$) are between subhalos. Therefore subhalo interactions must be incorporated in models of structure formation.  We find that the collision rate between halos in physical density units, is $\\propto (1+z)^\\delta $, with $\\delta = 3-4$, in agreement with earlier simulations and most observational data.  We test previous analytic estimates of the interaction rates of subhalos based on statistical models, which could be very inaccurate because of the small number of subhalos and the variation of conditions within small host halos. We find that, while such statistical estimates may severely overestimate the rate within hosts $< 10^{13} \\hsolmass$, typical of high redshifts, they are valid for larger hosts regardless of the number of subhalos in them. We find the \\citeN{makino-hut} estimate of the subhalo merger rate to be valid for hosts $\\ga 6\\times10^{11}\\hsolmass$ at all redshifts.  The collision rate between subhalos and the central object of their host halo is approximated relatively well using the timescale for dynamical friction in circular orbits. This approximation fails in $\\sim 40\\%$ of the cases, partly because of deviations from the assumption of circular orbits (especially at low redshift) and partly because of the invalidity of the assumption that the subhalo mass is negligible (especially at high redshift). ",
        "introduction": "\\label{sec:intro}   Interactions among dark matter halos and the galaxies that reside in them are inevitable in any cosmological scenario of hierarchical structure formation and have important observational implications. Halo interactions cause tidal stripping and harassment \\cite{moore:98}. They promote the exchange of angular momentum (e.g., \\citeNP{barnes-efstathiou}), trigger star bursts (e.g., \\citeNP{mihos:94a,mihos:94b}), mix different stellar populations \\cite{larson-tinsley:78}, produce tidal tails \\cite{toomre:72,dubinski:99,springel-white}, and change the galaxy morphology (e.g., \\citeNP{oemler:92}). The tidal stripping and harassment that occur during these interactions enrich the intergalactic medium with processed elements \\cite{kolatt-od}, disrupt existing galaxies, build new ones, and heat up the galaxy interior matter.  Galaxy interactions are observed directly (e.g., \\citeNP{int-groups,int-faint,int-clusters,int-mds}) and indirectly by statistical means and through their outcomes \\cite{abraham:97,conselice:00,kwu-phd}. The interaction rate rises as function of redshift \\cite{le-fevre:00,patton:00,patton:98,carlberg:1990,zepf+koo} (but cf. \\citeNP{carlberg:00}) and the density of the environment. Since the interaction rate is roughly proportional to the number density squared times the relative velocity, groups and clusters of galaxies show a higher specific interaction rate (per object) than the specific rate in the field.   The details of the physical processes that take place during and as a result of these interactions are not well understood. Gas dynamics, star formation and supernova feedback hinder a direct understanding of how such interactions influence the process of galaxy formation in general. In order to overcome this difficulty, semi-analytic investigations \\cite{kwg:93,cafnz,sp:99,spf:00} use simplified models for the merger rate of substructure (including the role of dynamical friction) in order to explore the possible implications of these events. These simplified models should be tested and improved by comparison with realistic cosmological simulations.  But even before addressing the additional physics involved in galaxy interactions, the halo gravitational interaction rate alone is too complicated to calculate analytically. The finite halo size and density profile, the many body interactions, the varying background potential all have their impact on the interaction rate.  The first stage of the theoretical exploration of gravitational halo interactions introduces the impulsive approximation \\cite{spitzer:58}, to estimate the effect of a close passage of two gravitationally bound systems. This approximation models the heating up of each individual system by the transfer of gravitational potential energy to internal kinetic energy. It also models the disruption by stripping and the contraction of the remaining debris. The impulsive approximation has been tested \\cite{richstone:75,DLS:80} in \\nbody\\ simulations where the effects of close encounters on mass exchange and profile evolution were addressed. The main objective of these investigations was to test mass loss due to tidal stripping. The conclusion was that although the impulsive approximation works well even beyond ``fast\" encounters, it is invalid for encounters with small impact parameters. Later, using two-halo \\nbody\\ experiments, \\citeN{augilar:85} found that the impulsive approximation works only if the minimal distance between the passing halos is bigger than $\\sim 10$ effective (de Vaucouleurs) radii. This clearly prevents the use of this approximation for close encounters and collisions.  The evolution of many-galaxy systems (rich clusters) was addressed by \\citeN{merritt:83} with the inclusion of close encounters and tidal stripping, but not merging. Merritt investigated the question of mass segregation between subhalos and the host cluster, and found that the implied dynamical friction time is longer than would be expected without the presence of close encounters and tidal stripping.  Merging of dark matter halos was investigated on the basis of simulations by \\citeN{makino-hut} (MH), who parameterized equal-mass mergers as a function of impact parameter and relative velocity for different initial density profiles.  MH also explored the parameter space of very close passages.  Recently, many authors have performed detailed simulations of individual multi-halo systems (``clusters\") \\cite{moore:96,tormen:97,tormen+b+w:97,tormen:98,ghigna:1998,van-den-bosch:99,sensui:99,okamoto-habe:99,moore:99a}. These investigations focused on issues like tidal stripping, harassment, and orbital parameters. They led to various insights about the role interactions play within clusters, understanding how morphology changes due to fast encounters, how anisotropic orbits affect the final density profile, quantifying the dynamical equilibrium state in clusters, and quantifying the mass fraction of subhalos with respect to the diffuse matter. However, systems the size of rich and poor galaxy groups have been neglected until now. The present work is the first detailed investigation of subhalo encounters within a cosmological volume simulation.  Previously, the only relevant cosmological modeling of halo interactions has been via analytic methods such as the Press-Schechter (PS) formalism \\cite{ps:74} and its extended versions (EPS) (\\citeNP{lc:93,lc:94}). These theoretical predictions have been compared to \\nbody\\ results, with surprisingly good agreement (up to a factor of $\\sim 2$) (e.g., \\citeNP{gross:suites,slkd:00}). Unfortunately EPS theory only accounts for mergers between distinct, virialized halos, and does not consider any other form of interaction. In particular EPS does not take into account non-merger interactions and does not provide the joint multiplicity-mass function of the progenitors (cf. \\citeNP{sk:99}). In addition, EPS theory neglects the spatial distribution of the existing halos, environmental dependence, and, most importantly, the interactions between subhalos within a given encompassing virialized host halo. All of these cases are extremely relevant to galaxy formation and evolution. For example, the results of MH were used by Somerville et al. (2000) to include in semi-analytic models the process of subhalo mergers, which they considered to be related to the Lyman break galaxies observed at high redshift ($z\\ga3$). In \\citeN{kolatt-lbg} we emphasize the role these collisions play, without appealing to MH but rather as identified in the high-resolution simulations studied in more detail here.  One may think that the shortcomings of the PS/EPS formalism may be overcome by using EPS to model isolated systems and then applying kinetic theory to virialized systems and substructure (cf. Somerville et al. 2000). However, this approach is questionable -- first because of the cross-talk between isolated halos and subhalos, and then because of the limitations of the kinetic theory.  In the kinetic theory within the framework of statistical mechanics one predicts the collision rate among many bodies in a given system. When we calculate collision or merger rates in a gravitating many body system, the simple approach of kinetic theory is no longer valid. Several important differences prevent us from applying the conventional methods of statistical mechanics:  \\begin{enumerate}  \\item The gravitating systems under consideration are not necessarily in equilibrium. The velocity distribution function may be anisotropic, may deviate from the usually assumed Maxwellian distribution for relaxed systems, and may evolve in time. Moreover the number of ``bodies\" (i.e., halos) is not constant; halos may merge or dissolve and other new halos may enter the system or just pass through it in the course of time.  \\item The ``bodies\" are not equal, but exhibit a large range of masses. The mass function (and thus the cross section distribution) is complicated and evolves in time and space.  \\item The ``bodies\" are not point-like or rigid bodies, nor do they retain their identity; ``bodies\" may shrink, evaporate, grow, deform, etc. \\label{item:par_evol}  \\item Gravitating systems exhibit ``gravitational focusing\" \\cite{BT}, i.e., an increase of the cross section due to the gravitational attraction between the interacting bodies.  \\item Most of the gravitational systems under consideration are embedded within a background potential (e.g., individual galaxy halos in groups or clusters). This potential cannot in general be neglected. Moreover, the interplay between the cluster or group potential and its subhalos (see item \\ref{item:par_evol}) evolves in time.  \\item The effect of the background potential sometimes adds anisotropy to the orbits -- for example spiraling-in of satellite halos within a host halo due to dynamical friction \\cite{BT}, or radial orbits along filaments and to the center of deep potential wells.   \\end{enumerate}  For the reasons outlined above, at present the only way to measure the halo interaction rate and other characteristics is to resort to high-resolution \\nbody\\ simulations. There are some minimum requirements for any simulation used to tackle such a problem. One needs a big enough dynamical range in spatial and mass resolution for the compilation of a statistical sample of the host/subhalo systems within a cosmological volume. The mass ratio between subhalos and hosts can range from $\\sim 1$ to three or four orders of magnitude. The evolution of number density of the host halos and their subhalos is strongly dependent on the cosmological background and its ability to provide a fresh supply of progenitors throughout the host evolution. In addition, spatial resolution is of paramount importance since substructure tends to erase within the host unless the resolution allows its clear identification. Another necessary condition is the storage and analysis of many time-steps of the simulation to determine the temporal behavior and interaction of the substructure.  The simulations we use in the current analysis \\cite{klypin:overcoming} meet all of these requirements, and in doing so represent an important advance in simulation technology. We have designed, tested, and used a new hierarchical halo finder (\\citeNP{bullock_thesis,profiles} hereafter BKS+) specifically for the goals of this paper.  In this paper we focus on the DM component alone and analyze the interactions between halos and within host halos -- those that have at least one other halo within their virial radius. We make no discrimination by host or subhalo masses, though we sometimes segregate the results into clusters and groups. A major goal of this paper is to assess various estimates for the time scales in virialized systems that consist of distinct constituents.  In section \\ref{sec:simu} we briefly discuss the \\nbody simulations and describe our strategy to find and model the dark matter halos. We shall then define the hierarchy we build for halos within the simulation and the various criteria for interactions. Section \\ref{sec:stat_view} prepares the ground for the interaction rate calculation by discussing the multiplicity function of host halos and its sensitivity to mass resolution. The multiplicity function plays an important role in various analytical estimates for the collision rate. In \\S{\\ref{sec:rate} we shall present the collision rate of halos, and the progenitor mass spectrum. A comparison to various approximations for collision timescales is carried out in \\S\\ref{sec:evaluation}, and we devote \\S\\ref{sec:df} to a similar evaluation of the dynamical friction timescale. We finally summarize our findings in \\S\\ref{sec:conc}. ",
        "conclusions": "\\label{sec:conc}   We investigated the interaction rate of subhalos, which should be an important ingredient in the process of galaxy formation. This study is needed because virialized systems of objects evolve differently than unvirialized systems in which the objects are only weakly correlated. Also, the observed large fraction of galaxies in groups suggests that virialized systems of two galaxies or more are very common, and such high-density environments tend to be preferentially selected for observation. We thus computed the collision and merger rates within virialized systems in N-body simulations, and used the results to evaluate useful semi-analytic approximations. The large sample of halos of a range of masses and multiplicities enabled a comprehensive statistical analysis, and the especially designed halo finder and collision identifier allowed a proper quantitative study of substructure evolution. Given the resolution limits of our simulations and halo finder, we could account for only the two most massive members of a group like the Local Group, but since these two members commonly dominate the group dynamics, the mass resolution is adequate for our purpose.  We found that the comoving collision rate of subhalos is a substantial fraction to the total collision rate of halos in the $\\Lambda$CDM simulation. At $z\\sim 4$ the fraction of collisions that involve substructure is $\\sim 0.3$, and this fraction grows to $\\sim 0.75$ at low redshifts (Fig. \\ref{fig:rate_sub} (top)).  In physical coordinates, this corresponds to a collision rate $\\propto (1+z)^{3-4}$ (Fig. \\ref{fig:rate_sub_phys}). The evolution of the subhalo collision rate follows that of the total collision rate, namely a sharp rise between $z=4$ and 2, followed by a decline between $z=1$ and 0.4. (Fig. \\ref{fig:rate_sub_sub}). Even though the exact evolution is affected by the specific cosmology simulated and the results are valid for a specific mass range limited by the resolution of the simulation, the general behavior of a rise followed by a decline should be robust. We find that the collision rate of subhalos lags behind the total rate by about $1-2$ Gyr. It also exhibits a different behavior at low redshift, when many large virialized systems exist and contribute a significant fraction of the collision rate and a rise in the merger rate.  The mass function of progenitors that participate in subhalo mergers is somewhat different from that of distinct halo collisions (Fig. \\ref{fig:mratio_sub}), amplifying small differences in the tails of the mass functions for subhalos and distinct halos. The fraction of ``major\" interactions among those which involve at least one subhalo is only $\\la 0.5$, unlike the $\\sim 0.6$ fraction of major interactions among distinct halos. The moral is that accurate theoretical predictions of interaction rates should not be solely based on the halo mass function as derived from approximations such as Press-Schechter or Extended-PS, which ignore substructure.  We found that the Makino-Hut formula can serve as a reasonably good approximation for subhalo merger rates, as long as host halos more massive than $\\sim 6 \\times 10^{11}\\hsolmass$ are considered. At lower masses, the MH formula overestimates the merger rate by an order of magnitude. At high redshift, where less massive halos are abundant, the MH formula ceases to be a good description for the merger rate of the subhalo population.  Predictions of the subhalo collision rate from kinetic theory are found to recover the actual rate for host halos more massive than $\\sim 10^{13.5} \\hmsun$, but fail to do so, by one or two orders of magnitude, for less massive host halos, depending on their multiplicity. We presented a correction formula to match kinetic theory predictions with the simulated halos, using either the mass of the host, or the number of subhalos it contains.  For interactions between subhalos and their hosts, we examined the accuracy of the dynamical friction timescale as estimated for circular orbits. In general, the estimate based on the NFW density profile is acceptable for about half the cases. It significantly overestimates the collision time in about $40\\%$ of the cases, and by more than an order of magnitude in $\\sim 20\\%$ of the cases. One reason for this failure is that, at least at early times, the halos tend to be of low concentration and the high-mass host halos are of low abundance, making the assumption of $M_{\\rm sub}/M_{\\rm host} \\ll 1$ invalid in many cases. The deviation from circular orbits is clearly another reason for the failure of the dynamical-friction approximation, as demonstrated at the most recent output time of the simulation. We find that, by some coincidence, the estimate based on the approximation of halos as isothermal spheres is acceptable at high redshift. This is good news for semi-analytic models that commonly use this approximation. At low redshifts, on the other hand, the concentration tends to be high, and the SIS predictions become poor. The moral is that, in order to mimic substructure dynamics in a cosmological context, one needs to appeal to accurate simulations, or use approximations that were derived from such simulations. We have reasons to assume that these approximations, once expressed properly in terms of the host-halo mass, can be robust and not too sensitive to the particular cosmology. Proper collision rates can then be incorporated into galaxy formation scenarios and connect them to the evolution of large-scale structure."
    },
    "0010/astro-ph0010014_arXiv.txt": {
        "abstract": "We develop a general formalism for analysing parameter information from non-Gaussian cosmic fields.  The method can be adapted to include the nonlinear effects in galaxy redshift surveys, weak lensing surveys and cosmic velocity field surveys as part of parameter estimation. It can also be used as a test of non-Gaussianity of the Cosmic Microwave Background.  Generalising Maximum Likelihood analysis to second-order, we calculate the nonlinear Fisher Information matrix and likelihood surfaces in parameter space.  To this order we find that the information content is always increased by including nonlinearity. Our methods are applied to a realistic model of a galaxy redshift survey, including nonlinear evolution, galaxy bias, shot-noise and redshift-space distortions to second-order.  We find that including nonlinearities allows all of the degeneracies between parameters to be lifted. Marginalised parameter uncertainties of a few percent will then be obtainable using forthcoming galaxy redshift surveys. ",
        "introduction": "The estimation of cosmological parameters from galaxy redshift surveys, Cosmic Microwave Background (CMB) surveys, weak lensing surveys and velocity field surveys has become a major industry in cosmology, due to the data explosion the field is currently undergoing. To date the majority of the analysis has considered the problem of parameter estimation from the large-scale, linear regime. However much of the important information on parameters lies on smaller scales, where nonlinear gravitational clustering has distorted the initial field.  In the case of galaxy redshift surveys additional nonlinearity may arise form the bias relationship between galaxies and matter, while in the case of the CMB the original pattern is still evident, allowing one to test the temperature fluctuations for evidence of primordial non-Gaussianity.   The industry standard for parameter estimation in the linear regime is the multivariate Gaussian Likelihood method. Within this framework the uncertainty on parameters can be estimated from the second derivatives of the likelihood function, the Fisher Information matrix.  Tegmark, Taylor \\& Heavens (1997) were the first to study the Gaussian Fisher Information matrix as a general tool for characterising the information content of cosmological surveys, while Heavens \\& Taylor (1997) demonstrated its importance for the design of galaxy redshift surveys. Tegmark (1997) used the Gaussian Fisher matrix to investigate the accuracy of parameter estimation from the Sloan Digital Sky Survey in some detail, but did not calculate the effects of nonlinear clustering, redshift-space distortions, or nonlinear biasing. Jungman et al. (1996) applied a Fisher analysis to the CMB, while Hu \\& Tegmark (1999) used it to estimate parameter uncertainties in weak lensing surveys.   While likelihood parameter estimation in the linear, Gaussian regime has been extensively studied, until recently little work has been done on the non-Gaussian generalisation. Amendola (1996) and Rocha et al. (2000) have extended the likelihood methods to include non-Gaussian contributions by expanding the likelihood function in a series of cumulants. Amendola considered the effects of degradation of parameter uncertainties due to the non-Gaussian broadening of the likelihood function, while Rocha et al. applied their likelihood to estimate residual non-Gaussianity in the CMB.   In the analysis of galaxy redshift surveys, Matarrese Verde \\& Heavens (1997) have investigated a Gaussian likelihood estimator for the bispectrum as a tool for extracting the mean density parameter and galaxy bias from galaxy redshift surveys in the nonlinear regime. This builds on the work of Fry (1994) who suggested that the bispectrum could be used to break the linear degeneracy between the bias parameter and the density parameter, $\\Omega_m$.  Verde et al. (1998) developed this further to include redshift-space distortions. Scoccimarro et al. (1998) and Scoccimarro, Couchman \\& Frieman (1999) similarly investigated the bispectrum, including the effects of redshift-space distortions and galaxy bias.  Scoccimarro et al. (2000) have recently estimated the bias parameters from the combined IRAS-QDOT and 2Jy and 1.2Jy redshift surveys.   In this paper we develop a new method for exploring the estimation of cosmological parameters from scales going into the nonlinear regime. We expand the general non-Gaussian multivariate Probability Distribution Function (PDF) and derive the non-Gaussian Fisher matrix. Since the most relevant non-Gaussianity will be due to nonlinear evolution of the density field, we use second-order perturbation theory to find the dependance of the higher-order moments on cosmological parameters and estimate the Fisher matrix to second order.  We illustrate these methods using a realistic model galaxy redshift survey, but note that they can be similarly applied to a  2-d weak lensing, or 3-d cosmic velocity survey.   The paper is organised as follows. In Section 2 we describe the expansion of the multivariate probability distribution function in both the discrete and continuum limits. In Section 3 we develop the formalism for calculating the Fisher information matrix in the non-Gaussian regime.  We introduce a new function, the parameter entropy, to study the likelihood surface in parameter space in Section 4 and calculate a general expression for non-Gaussian fields. These methods are illustrated in Section 5 where we apply them to a model galaxy redshift survey, including the effects of shot-noise, redshift-space distortions and galaxy bias. We show how nonlinear effects change the estimation of parameters in two surveys; the 2-degree Field galaxy redshift survey (2dF; Colless 1996) and the Sloan Digital Sky Survey (SDSS; Gunn 1995).  Finally, we present our conclusions in Section 5. Three appendices deal with technical issues. ",
        "conclusions": "\\label{conclusions}  We have presented a method for including higher-order moments in the likelihood functions of cosmological fields in such a way that the parameter dependencies of the non-Gaussian terms may be used for estimating cosmological parameters. This non-Gaussian correction generalises likelihood analysis for application to fields that either contain intrinsic non-Gaussianity, or have become non-Gaussian due to nonlinear gravitational evolution. Such a generalisation is fundamentally important for cosmology since most of the interesting fields show non-Gaussian properties. Our method has the advantage that we attempt to deal with the probability distribution of the field explicitly, allowing the natural combination of linear and nonlinear regimes. In this respect our work differs significantly from existing techniques that separate out the nonlinear regime and apply a Gaussian PDF to the higher-order moments themselves (Matarrese et al. 1997, Verde et al. 1998, Scoccimarro et al. 1999, Scoccimarro et al. 2000).  We have presented a general formalism for calculating the non-Gaussian Fisher matrix and the shape of the likelihood around its maximum, the parameter entropy function. A central result of this paper has been to show that to lowest order, the dominant effect of adding nonlinearity, or more generally a non-Gaussianity, is to increase the the parameter information. This effect dominates over the degradation of parameter information due to the nonlinear evolution of the shape of the likelihood function.    Applying our analysis of nonlinearity to a simple model for a galaxy redshift survey, including the effects of shot-noise, redshift-space distortions and galaxy bias, we have found that nonlinear effects may be extremely useful for placing tight constraints upon cosmological parameters. Of crucial importance is the fact that even at second-order, degeneracies can be broken so that all of the cosmologically interesting parameters may be estimated independently. While going to higher-order may place tighter constraints on the nonlinear bias function through estimation of the series coefficients $b_n$, essentially redshift surveys offer up a great deal of their information without the need to turn to progressively more intricate perturbative calculations. In addition, this analysis suggests that galaxy redshift surveys can be used in isolation to determine cosmological parameters. This greatly enhances their power as parameter estimation can be compared from independent surveys, such as the CMB, cosmic velocity fields and weak lensing surveys and combined further to reduce uncertainties.    Our analysis relies upon a multivariate generalisation of the Edgeworth approximation for the probability distribution function.  As such one should be careful to determine the range of applicability of second-order perturbation theory, and the series expansion of the likelihood function.  It is well known from studies of 1-point distributions (Bernardeau \\& Kofman 1995, Juszkiewicz et al. 1995 Gazta\\~{n}aga et al. 1999, Taylor \\& Watts 2000, Watts \\& Taylor 2000) that the Edgeworth PDF has limitations -- displaying unphysical features when the variance or skewness becomes too high. However, these effects do not automatically render the Edgeworth a bad approximation. Comparison with N-body simulations in redshift-space (Watts \\& Taylor 2000) has shown that peak of the distribution remains a good fit even when the extremities begin to behave badly. Fortunately we have found that a good deal of information comes from the nonlinear terms at relatively low-$k$ and have shown that the degeneracies between parameters are broken even for modest values of wavenumber. This is good news from a computational point of view as well since the number of modes available for analysis grows like $k^3$. While this is beneficial in some respects, it can also lead to computational problems if the number of modes gets too high. In his case data compression methods may need to be incorporated (Taylor et al. 2000).  In a forthcoming paper (Watts \\& Taylor, in preparation) we rigorously test our model on N-body simulations in order to accurately determine the limitations, and address the specific issues associated with application to real data sets."
    },
    "0010/astro-ph0010478_arXiv.txt": {
        "abstract": "We estimate the two- and three-dimensional power spectra, $P_2(K)$ and $P_3(k)$, of the galaxy distribution by applying a maximum likelihood estimator to pixel maps of the APM Galaxy Survey. The analysis provides optimal estimates of the power spectra and of their covariance matrices if the fluctuations are assumed to be Gaussian. Our estimates of $P_2(K)$ and $P_3(k)$ are in good agreement with previous work but we find that the errors at low wavenumbers have been underestimated in some earlier studies.  If the galaxy power spectrum is assumed to have the same shape as the mass power spectrum, then the APM maximum likelihood $P_3(k)$ estimates at $k \\le 0.19 h {\\rm Mpc}^{-1}$ constrain the amplitude and shape parameter of a scale-invariant CDM model to lie within the $2\\sigma$ ranges $0.78 \\le (\\sigma_8)_g \\le 1.18$ and $0.05 \\le \\Gamma \\le 0.38$.  Using the Galactic extinction estimates of Schlegel, Finkbeiner and Davis, we show that Galactic obscuration has a negligible effect on galaxy clustering over most of the area of the APM Galaxy Survey.   \\vskip 0.2 truein ",
        "introduction": "In this paper we analyse the three-dimensional power spectrum of galaxy clustering using the APM Galaxy Survey (Maddox \\etal~ 1990a, b, c). The APM Galaxy Survey is a two-dimensional catalogue of galaxies complete to a magnitude limit of $b_J = 20.5$ and covering an area of approximately $12$ percent of the sky. The survey has been used to estimate the angular two-point correlation function  $w(\\theta)$ and the angular power spectrum $P_2(K)$, which are related to their three-dimensional analogues $\\xi(r)$ and $P_3(k)$ via simple integral equations (Limber 1953; Groth and Peebles 1977; Baugh and Efstathiou 1994). Recovering the three-dimensional power spectrum from angular statistics therefore requires stable numerical techniques for inverting these integral equations.   Baugh and Efstathiou (1993) described a technique for recovering the three-dimensional power spectrum from measurements of the angular correlation function. The three-dimensional power spectrum was parameterized by a set of amplitudes $P_3^i$ (or `bandpowers') over bands of wavenumbers centred at wavenumber $k_i$. The integral equation relating $w(\\theta)$ to $P_3(k)$ was solved using Lucy's (1974) iterative deconvolution technique. A similar technique was applied by Baugh and Efstathiou (1994) to recover $P_3(k)$ from estimates of the two-dimensional power spectrum $P_2(K)$ and by Baugh (1996) to recover $\\xi(r)$ from $w(\\theta)$. These investigations show that  Lucy's algorithm can provide a stable inversion. However, it is difficult to derive a reliable covariance matrix for the recovered estimates of $P_3^i$. Baugh and Efstathiou (1993, 1994) derived estimates of the errors by computing the scatter in the $P_3^i$ derived from four nearly equal areas of the APM survey. However, since the number of zones is small,  these error estimates are crude and cannot be used to fit theoretical models with any confidence.  Recently, Dodelson and Gazta\\~naga (2000) have described a method of inverting $w(\\theta)$ to recover $P_3^i$ that employs a Bayesian prior to contrain the smoothness of the inversion. This method can return a covariance matrix for $P_3^i$, but requires an estimate of the covariance matrix of the input estimates of $w(\\theta_i)$ and a model for the Bayesian prior. Eisenstein and Zaldarriaga (1999) present another inversion technique using singular value decomposition (see {\\it e.g.} Press \\etal~ 1992). Their method also recovers the covariance matrix for $P_3^i$ but requires estimates of $w(\\theta)$ and its covariance matrix as inputs.  The purpose of this paper is two-fold. Firstly, to assess the effects of Galactic extinction on large scale clustering in the APM Survey using the extinction model of Schlegel, Finkbeiner and Davis (1998, hereafter SFD) based on the COBE/DIRBE and IRAS maps. Secondly, to apply to the APM Survey modern maximum likelihood (ML) techniques similar to those used to estimate the power spectrum of the cosmic microwave background (CMB) anisotropies (Bond, Jaffe and Knox 1998; de Bernardis \\etal~ 2000; Hannay \\etal, 2000). With the increase in computer power over the ten years since the APM survey was completed, it is now feasible to perform a direct ML estimate of the angular power spectrum over wavenumbers extending into the non-linear regime.  This provides an optimal estimate (under certain assumptions) of the power spectrum and its covariance matrix in a conceptually straightfoward way, avoiding the need for estimators of $P_2(K)$ or $w(\\theta)$ that require a model of the true power spectrum. (See {\\it e.g.}  Hamilton, 1997a, b; Tegmark 1997, Kerscher \\etal~ 2000, and references therein for a discussion of estimators of $P_2(K)$ and $w(\\theta)$). An additional advantage of ML methods is that it is as easy to compute bandpower estimates of the three-dimensional power spectrum $P_3^i$ (and its covariance matrix) as it is to estimate the two-dimensional power spectrum. The inversion from two to three dimensions can therefore be done with the same computer code and without the need for any assumptions other than that the underlying fluctuations obey Gaussian statistics.  The outline of this paper is as follows. Section 2 describes the method and applies it to Gaussian realizations of the APM Survey. In Section 3, we use the SFD dust maps to show how the two-dimensional power spectrum is affected by Galactic extinction. A model for the mean distribution of galaxies with redshift $dN(z)/dz$ is constructed using data from the 2dF Galaxy Redshift Survey and this is used to compute the two- and three-dimensional power spectra by ML. Constraints on theoretical models are discussed in Section 4 and our conclusions are summarized in Section 5. ",
        "conclusions": "In this paper we have tested and applied a maximum likelihood method to estimate the 2d and 3d power spectra of the galaxy distribution from a two-dimensional catalogue with a known redshift distribution. The methods are similar to those applied to estimate the power spectrum $C(\\ell)$ of the CMB anisotropies. However, applied to galaxy clustering, the ML method provides an optimal way of estimating the three-dimensional power spectrum and its covariance matrix directly from the 2d data. This provides a simple alternative to inverting the integral equations relating the 2d power spectrum, or angular correlation function, to the 3d power spectrum.  We have investigated the effects of Galactic extinction on the APM survey using the extinction maps of Schlegel, Finkbeiner and Davis (1998). Galactic extinction is shown to have little  effect on the power spectrum over the APM area with $\\delta < - 20^\\circ$. Eliminating the small regions below this declination limit where the extinction exceeds $0.2$ magnitudes depresses the power spectrum of dust still further so that it has a neglible effect on the power spectrum of galaxy clustering.  Our results show that the ML power spectra are in good agreement with previous estimates of the 2d and 3d power spectra of the APM Survey (BE93, BE94; Maddox \\etal, 1996; Eisenstein and Zaldarriaga, 1999; Dodelson and Gazta\\~naga 2000). The ML method produces reliable estimates of the covariance matrices of the 2d power spectra.  In agreement with Eisenstein and Zaldarriaga, we conclude that the errors on the 3d power spectrum have been underestimated in earlier papers (BE93, BE94) and that there is no evidence for a peak, or turnover, in the APM galaxy power spectrum at $k \\simlt 0.1 h {\\rm Mpc}^{-1}$. By fitting a scale invariant CDM model to the 3d power spectrum at wavenumbers $k \\le 0.19 h {\\rm Mpc}^{-1}$ we find the amplitude and shape parameter lie within the ranges $0.78 \\le (\\sigma_8)_g \\le 1.18$ and $0.05 \\le \\Gamma \\le 0.38$ at the $2\\sigma$ level. Including the scalar spectral index $n_s$ as a parameter significantly weakens the constrains. Nevertheless, compatibility with the APM data requires that CDM models with $\\Gamma \\simgt 0.5$ have spectral indices $n_s \\simlt 0.8$ at the $2\\sigma$ level.  The methods described here have applications to other 2d surveys, for example, the forthcoming Sloan Digital Sky Survey (SDSS, see {\\it e.g.}  Margon 1999). It is also possible that a pixel based ML method, as described here, may prove useful for the analysis of 3d galaxy redshift surveys such as the 2dFGRS and the SDSS. The likelihood distributions derived here may have some applications in parameter estimation studies using CMB and other data ({\\it e.g.} Jaffe \\etal, 2000; Lange \\etal, 2000). The likelihood distributions plotted in Figures 9 and 11 are available from the authors on request.   \\bigskip  \\noindent {\\bf Aknowledgements:} S.J. Moody thanks PPARC for the award of a research studentship. We also thank the 2dFGRS team for allowing us to use some of their data prior to publication."
    },
    "0010/astro-ph0010634_arXiv.txt": {
        "abstract": "These lecture notes intend to form a short pedagogical introduction to the use of {\\it typical} type Ia-Supernovae (hereafter SNIa) as standard candles to determine the energy density of the universe. Problems of principle for taking SNIa as cosmological probes are pointed out, and new attempts at solving them are indicated including the empirical width-luminosity relation (WLR) and its possible explanations. \\\\ Finally, the observations of SNIa at high redshift carried out by two major teams are briefly reviewed and their interpretation as evidence for an accelerating universe is also rapidly discussed ",
        "introduction": "A supernova is a very powerful event: a star which suddenly brightens to about 10$^9$-10$^{10}$ L$_\\odot$. There are two different types of supernovae (see Table 1): SNII and SNIa, such as: \\begin{itemize} \\item SNII are core-collapse induced explosions of short-lived massive stars \\\\ (M$_\\star \\, > $ 8 M$_\\odot$) which produce more $O$ and $Mg$ relative to $Fe$. \\item SNIa are thermonuclear explosions of accreting white-dwarfs in close binaries which produce mostly $Fe$ and little $O$; the exact companions stars of white-dwarfs are generally not identified but must be relatively long-lived stars. \\end{itemize} There are also particular supernovae: $SNIb$ and $SNIa$ with most properties of $SNII$ except their spectra which have no $H$ line (such as SNIa) but they believed to be core-collapse supernovae (such as SNII). \\\\  \\begin{table*}[h!tbp] \\renewcommand{\\arraystretch}{.90} \\centering \\begin{tabular}{||l||c|c||}\\hline \\hline {\\bf Type} & {\\bf SNIa}  & {\\bf SNII}  \\\\ \\hline \\hline & & \\\\ {\\bf Hydrogen} & no & yes \\\\ \\hline & & \\\\ {\\bf Optical spectrum} & Metal lines & P Cyg lines \\\\ & deep 6180 $\\AA$ & Balmer series \\\\ \\hline & & \\\\ {\\bf Absolute luminosity} & $\\sim$ 4$\\times 10^9$ L$_{\\odot}$ & $\\sim$ 10$^9$ L$_\\odot$ \\\\ at max. light & small dispersion  & large dispersion \\\\ & standard candles ? & \\\\ \\hline & & \\\\ {\\bf Optical light curve} & homogeneous & heteregeneous \\\\ \\hline & & \\\\ {\\bf UV spectrum} & very weak & strong \\\\ \\hline & & \\\\ {\\bf Radio emission} & no detection & strong \\\\ & & slow decay\\\\ \\hline & & \\\\ {\\bf Location} & all galaxies & spirals and irregulars\\\\ \\hline & & \\\\ {\\bf Stellar population} & old & young\\\\ \\hline & & \\\\ {\\bf Progenitors} & WD in binary systems & massive stars\\\\ \\hline \\hline \\end{tabular} \\label{tab:geom} \\end{table*}       For some time astronomers had focused on all types of supernovae to measure Hubble's constant. Now, with the use of Hubble Space Telescope (HST) observations of Cepheids in the host galaxies of these supernovae, there is a consensus that $H_o$ is in the range of 60-70 km s$^{-1}$ Mpc$^{-1}$ -see Branch (1998) for a review. \\\\ But very recently, two observational groups working independently to use SNIa as standard candles to measure distance versus redshift presented evidence that the Hubble expansion has been accelerating over time. In effect, these two independent groups (Perlmutter et al. 1998, Schmidt et al. 1998) had to develop not only a method of discovering SNIa at high redshift by careful observations -with large format CCDs, large apertures- and scanning of plates but they also had to determine if these SNIa could be used as standard candles. This recent and important progress has been possible because an empirical relation (Phillips 1993) between the duration of the peak phase of a SNIa's light curve and its luminosity ({\\it broader is brighter}) has been taken into account by both teams. \\\\ This is precisely the work of these two groups that we are trying to present, in an educational manner, in these lessons. In this scope, in section 2, we summarize our current understanding of SNIas: their possible progenitors; some models of their explosions; observed optical spectra of SNIas, $\\gamma$-rays from SNIas, observed optical light curves and finally the empirical width- luminosity relation (WLR) which is also called the {\\it Phillips Relation}. In section 3, we give a cosmological background by recalling what is the ($\\Omega_\\Lambda$, $\\Omega_M$)-plane and the luminosity distance. In section 4, we present the observations of SNIa at high redshift by the two major teams: the {\\it Supernovae Cosmology Project} (SCP) and the {\\it High z supernova search team} (HZT). In section 5, we discuss some aspects of their results and finally give a conclusion. ",
        "conclusions": "After a brief review of type Ia supernovae properties (Section 2) and of cosmological background (section 3), we presented the work done by two separate research teams on observations of high-$z$ SNIa (Section 4) and their interpretation (Section 5) that the expansion of the Universe is accelerating. If verified, this will proved to be a remarkable discovery. However, many questions without definitive answers remain. \\\\ What are SNIa and SNIa progenitors ? \\\\ Are they single or double degenerate stars ? \\\\ Do they fit Chandrasekhar or sub-Chandrasekhar mass models at explosion ? \\\\ Is the mass of radioactive $^{56}Ni$ produced in the explosion really the main parameter underlying the {\\it Phillips relation} ? \\\\ If so, how important are SNIa evolutionary corrections ? \\\\ Can systematic errors be negligible and can they be converted into statistical errors ? \\\\ Can a satellite, like SNAP, provide large homogeneous samples of very distant SNIa and help determine several key cosmological parameters with an accuracy exceeding that of planned CMB observations ? \\vskip2mm \\noindent Meanwhile, cosmologists are inclined to believe the SNIa results, because of the preexisting evidence for {\\it dark energy} that led to the prediction of accelerated expansion. \\vskip2mm \\noindent {\\bf Acknowledgements}\\\\ The authors gratefully acknowledge Stan Woosley for helpful suggestions. Part of the work of D.P. has been conducted under the auspices of the {\\it Dr Tomalla Foundation} and Swiss National Science Foundation."
    },
    "0010/astro-ph0010128_arXiv.txt": {
        "abstract": "We report here a factor of 5.7 higher total CO flux in Arp~244 (the ``Antennae'' galaxies) than that previously accepted in the literature (thus a total molecular gas mass of 1.5$\\times$10$^{10}$ \\ms), based on our fully sampled CO(1-0) observations at the NRAO 12m telescope. Currently, much of the understanding and modeling of the star formation in Arp~244 has been derived from a much lower molecular gas mass. It is imperative to reconsider all as the high molecular gas mass might provide sufficient fuel for ultraluminous extreme starburst in Arp~244 once the merging advances to late stage.  Our observations show that the molecular gas peaks predominately in the disk-disk overlap region between the nuclei, similar to the far-infrared (FIR) and mid-infrared (MIR) emission. The bulk of the molecular gas is forming into stars with a normal star formation efficiency (SFE) $L_{\\rm IR}/M({\\rm H}_2) \\approx 4.2 \\ls/\\ms$, same as that of giant molecular clouds in the Galactic disk. Additional supportive evidence is the extremely low fraction of the dense molecular gas in Arp~244, revealed by our detections of the HCN(1-0) emission, which traces the active star-forming gas at density $\\approxgt 10^4$ cm$^{-3}$.  Using the high-resolution BIMA plus the NRAO 12m telescope, full-synthesis CO(1-0) images, and our VLA continuum maps at 20cm, we estimate the local SFE indicated by the ratio map of the radio continuum to CO(1-0) emission, down to kpc scale. Remarkably, the local SFE stays roughly same over the bulk of the molecular gas distribution. Only some localized regions show the highest radio-to-CO ratios that we have identified as the sites of the most intense starbursts with SFE$\\approxgt 20~\\ls/\\ms$. Here we have assumed that the 20cm emission is a fairly good indicator of star formation down to kpc scale in Arp~244 because of the well-known correlation between the FIR and the radio continuum emission. These starburst regions are confined exclusively in the dusty patches seen in the HST optical images near the CO and FIR peaks where presumably the violent starbursts are heavily obscured. Nevertheless, recent large-scale star formation is going on throughout the system (\\eg, concentrations of numerous super-star clusters and MIR ``hotspots''), yet the measured level is more suggestive of a moderate starburst (SFE$\\approxgt 10~\\ls/\\ms$) or a weak to normal star formation (SFE$\\sim 4~\\ls/\\ms$), not necessarily occurring at the high concentrations of the molecular gas reservoir. The overall starburst from the bulk of the molecular gas is yet to be initiated as most of the gas further condenses into kpc scale in the final coalescence. ",
        "introduction": "The ``Antennae'' galaxies (Arp~244, NGC~4038/39, VV~245), the nearest IR-luminous and perhaps the youngest prototypical galaxy-galaxy merger (the first in Toomre (1977) merger sequence), certainly reclaimed its fame from the recent releases of the HST/WFPC2 images (Whitmore \\etal 1999) and the Chandra X-ray images (Fabbiano \\etal 2000), and the observations at essentially all available wavelengths from radio to X-ray (\\eg,  Hummel \\& van der Hulst 1986; Neff, \\& Ulvestad 2000; Hibbard, van der Hulst, \\& Barnes 2001 in preparation; Bushouse, Telesco, \\& Werner 1998; Nikola \\etal 1998; Mirabel \\etal 1998; Vigroux \\etal 1996; Read, Ponman, \\& Wolstencroft 1995; Fabbiano, Schweizer, \\& Mackie 1997). The HST images reveal over one thousand bright young star clusters that are thought to have formed in recent bursts of star formation. H$\\alpha$ imaging also shows the most recent locations of the star-forming giant HII regions and their velocity fields as well~(Rubin, Ford, \\& D'Odorico 1970; Amram \\etal 1992; Whitmore \\etal 1999). Soft X-ray~(Fabbiano \\etal 1997) and radio continuum~(Hummel \\& van der Hulst 1986) images may hint of previous star formation sites currently seen as supernova remnants. In addition, mid-infrared (MIR)~(Mirabel \\etal 1998; Vigroux \\etal 1996), far-infrared (FIR)~(Evans, Harper, \\& Helou 1997; R. Evans 1998, private communications; Bushouse \\etal 1998) and submillimeter (Haas \\etal 2000) images indicate that the most intense starburst takes place currently in an off-nucleus region that is inconspicuous at optical wavelengths. The star-forming molecular gas in Arp~244 is perhaps least understood, however; we show in this paper that the total molecular gas mass accepted in the literature for the last decade~(Sanders \\& Mirabel 1985; Stanford \\etal 1990) has been underestimated by nearly a factor of 6.  The importance of gas during galaxy-galaxy merging far exceeds its mass proportion, as demonstrated by the sophisticated numerical simulations~(\\eg, Barnes \\& Hernquist 1996, 1998). At a distance of 20 Mpc ($H_0=75$\\kms~Mpc$^{-1}$, \\eg, van der Hulst 1979; Mihos, Bothun, \\& Richstone 1993; Mirabel \\etal 1998), the total IR luminosity (8--1000 $\\mu$m, Sanders \\& Mirabel 1996) of Arp~244, measured from the {\\it IRAS} four-band fluxes given in Soifer \\etal (1989), is $L_{\\rm IR}=6.2\\times 10^{10} \\ls$. Using the most recent remeasured total {\\it IRAS} fluxes (as opposed to simply the ``point source'' in Soifer \\etal 1989) in the Revised Bright Galaxy Sample (D.B. Sanders 2000, private communication; Sanders, Mazzarella, Kim, \\& Surace 2001 in preparation), the same IR luminosity is obtained. Therefore, strictly speaking Arp~244 is not a luminous infrared galaxy (LIG, $L_{\\rm IR}\\approxgt 10^{11} \\ls$) unless a Virgocentric flow distance of 29.5 Mpc is used (Sanders \\etal 2000, but we use 20 Mpc throughout this paper), which leads to $L_{\\rm FIR}=1.0\\times 10^{11} \\ls$ and $L_{\\rm IR}=1.3\\times 10^{11} \\ls$. A much smaller molecular gas mass of $2.6 \\times 10^9$~\\ms~ (Sanders \\& Mirabel 1985), currently accepted in the literature, leads to a higher global ratio of $L_{\\rm IR}/M({\\rm H}_2) \\sim 24~\\ls/\\ms$. This ratio is often referred as the star formation efficiency (SFE) since the FIR emission, tracer of current star formation rate, has been normalized by the molecular gas mass available to make stars. With this high yield of young stars per unit molecular gas mass and a lower molecular gas content, most of the molecular gas will be depleted in $\\sim 10^8$ years, as noted by Sanders \\& Mirabel (1985) and Stanford \\etal (1990). There would be no chance for Arp~244 to become an ultraluminous infrared galaxy (ULIG, $L_{\\rm IR}\\approxgt 10^{12} \\ls$) in the late stage of the merging. Numerical modeling (Mihos \\etal 1993), based on this lower gas mass, indeed predicted that Arp~244 would not join the rank of ULIGs.  ULIGs, with the IR luminosity comparable to the bolometric luminosity of QSOs, are the most luminous galaxies in the local universe. They are believed to be powered mainly by starbursts (e.g., Smith, Lonsdale, \\& Lonsdale 1998; Genzel et al. 1998) taking place predominantly in the {\\it extreme starburst} regions of a characteristic size of $\\sim 100$ pc and $L_{\\rm IR} \\sim 3\\times 10^{11} \\ls$ (Downes \\& Solomon 1998), produced as a result of the merging of molecular {\\it gas-rich} spiral galaxies (Sanders \\& Mirabel 1996). Alternatively, the dust-enshrouded active galactic nuclei (AGN) may still be responsible for significant energy output in some ULIGs (Sanders \\etal 1988; Veilleux, Sanders, \\& Kim 1999; Sanders 1999). Now, for a total molecular gas mass of $1.5\\times 10^{10} \\ms$, based on our fully sampled CO(1-0) observations obtained at the NRAO \\footnote{The National Radio Astronomy Observatory is a facility of the National Science Foundation operated under cooperative agreement by Associated Universities, Inc.} 12m, many previous conclusions and speculations about Arp~244 need to be revised. After all, the initial gas content, particularly the molecular gas --- the fuel for star formation, will probably be the most important factor in determining whether a merging pair of spiral galaxies reaches the peak of ultraluminous extreme starburst phase since all ULIGs are still gas-rich with $\\sim 10^{10}$ \\ms~ molecular gas mass (Solomon \\etal 1997), most likely concentrated in the kpc scale disks/rings around the merging nuclei (Downes \\& Solomon 1998; Scoville, Yun, \\& Bryant 1997; Sakamoto \\etal 1999; cf. Evans, Surace, \\& Mazzarella 2000).  With a molecular gas mass comparable to that of ULIGs and an early/intermediate merging stage, Arp 244 is perhaps an example of what an ULIG might have looked like a few hundred million years ago. Arp 244 may be a snapshot in the evolution of a typical gas-rich merger into an ULIG system. Thus, it is crucial to understand how, where and when the starbursts initiate or have occurred in such an on-going merger. This becomes especially imperative in Arp~244 given the advantage of its close-up distance and therefore the better linear resolution available. Nonetheless, the strongest starburst site revealed from the Infrared Space Observatory ($ISO$) MIR images (Vigroux \\etal 1996; Mirabel \\etal 1998; Xu \\etal 2000) seems to be offset from the peak emission in all the 60, 100, and 160 $\\mu$m FIR maps of the limited resolution, obtained with the Kuiper Airborne Observatory (KAO, Evans, Harper, \\& Helou 1997; Bushouse \\etal 1998). In order to best locate the sites of the intense star formation, high-resolution imaging in the FIR is ultimately required.  Since there is an excellent correlation between the FIR and the radio continuum emission (\\eg, Helou, Soifer, \\& Rowan-Robinson 1985; Condon \\etal 1990; Condon 1992; Xu \\etal 1994; Marsh \\& Helou 1995), the FIR emission can thus be approximately scaled according to the radio continuum emission with the high resolution achievable by the Very Large Array (VLA) observations. We have therefore, obtained the VLA radio continuum images to compare with our high-resolution full-synthesis Berkeley-Illinois-Maryland Association (BIMA) plus the NRAO 12m CO maps. This is because the star formation occurs within giant molecular clouds (GMCs), especially the dense cores and starburst can be better characterized by the elevated SFE, i.e., the FIR to CO ratio, approximated here by the radio-to-CO ratio. Therefore, local SFE across over the merging system can be approximately measured using the ratio map of the radio continuum to the CO emission to locate the most intense starburst sites of the highest radio-to-CO ratios.  We here briefly report our various observations in \\S 2 detailing the single-dish CO mapping in Arp~244. The BIMA interferometry CO and the VLA 20cm continuum observations will be presented elsewhere. The observational results and analysis are summarized and compared in \\S 3. Section 4 discusses the starburst properties of Arp~244 and the implications of our observations. Finally, we conclude our main results. ",
        "conclusions": "We summarize our main results and present our concluding remarks in the following:  1. Our fully sampled CO(1-0) map (FWHM$\\sim 55''$) at half-beam spacing reveals a factor of 5.7 larger total CO flux in Arp~244 than the previously accepted. This is mainly due to the enormous CO extent of Arp~244, which is much larger than the single pointing $55''$ beam used in the old observation. In general, single beam observations of merging systems may well systematically underestimate CO content in the cases where the merging systems are much larger than the single beam, unless a map of extensive coverage of the merging systems is made. Most of CO emission in Arp~244 indeed peaks at the disk overlapping region with a broad velocity spread of 600~\\kms. We obtain a total molecular gas mass of 1.5$\\times 10^{10}\\ms$ for Arp~244 using a standard CO-to-H$_2$ conversion factor.  2. HCN(1-0) emission has been detected in Arp~244. HCN observations suggest that there is only a small amount of dense molecular gas in Arp~244 compared to the total molecular gas content. The fraction of the dense molecular gas is only comparable to that of normal spiral galaxies and is much lower than that of luminous and ultraluminous infrared galaxies (LIGs/ULIGs).  3. We detected extended CO emission far away (nearly two full beams) from the major CO concentrations, such as those revealed by the interferometers. The peak CO emission in the overlap region has an average gas surface density near $\\sim 200~\\ms$~pc$^{-2}$ over $\\sim 5$~kpc --- the scale probed by the $55''$ beam. Even the resolved interferometry CO images only reveal that the highest gas surface density is about 10$^3~\\ms$~pc$^{-2}$, still order of magnitude lower than that of ULIGs. This is consistent with the small fraction of dense gas found from the HCN measurements. Weak CO emission from the tidal dwarf galaxy at the tip of the southern tail has also been tentatively detected and the estimated molecular gas mass is possibly larger than $\\sim 2\\times 10^8$~\\ms.  4. The excellent correspondence of the radio continuum images with the FIR maps is evident in Arp~244, which also implies the tight correlation between the two emission. We use the high resolution VLA 20cm image to approximately represent the FIR emission distribution in Arp~244, and compare with the CO images of comparable resolution to estimate the local star formation efficiency (SFE), which is defined as the FIR to molecular gas mass ratio, and is now indicated by the radio-to-CO ratio. SFE is low throughout the entire system with a global SFE$=4.2~\\ls/\\ms$, just same as that of the Galactic disk GMCs. Only some confined regions in the overlap, inconspicuous in optical observations, are the sites of the current powerful starbursts with SFE~$\\approxgt 20~\\ls/\\ms$, comparable to that of LIGs/ULIGs. The VLA radio continuum images also reveal a very similar morphology as that of the molecular gas distribution.  5. Although some localized starbursts are going on in the confined regions in the overlap, the bulk of the molecular gas is forming into stars with a low SFE, just same as that of normal spiral galaxies. The large amount of total molecular gas, yet only a small amount of dense molecular gas and a low SFE, revealed from our observations, combined with the theoretical predictions of models of gas-rich mergers, seem to suggest that Arp~244 has the ultraluminous extreme starburst potential when merging proceeds into late stage. Arp~244 appears to be a snapshot of an ULIG in its early stage of making.  6. Multi-wavelength comparison together with numerical models appear to indicate that Arp~244 has recently passed the peak of large scale active star formation. Globally, the bulk of the molecular gas in Arp~244 is currently not in a starburst phase. The confined bona fide starburst sites of the highest SFE appear to be offsets from the peaks of the CO, radio continuum, FIR, and MIR emission. The future ultraluminous extreme starburst phase can possibly be reached once the CO concentrations in the overlap merge with the double-nucleus gas concentrations in the final coalescence.  7. In conclusion, our newly determined 5.7 times larger total molecular gas mass in the ``Antennae'' galaxies lowers the global SFE to be comparable with that of GMCs in the Milky Way disk. Comparing the CO images with other observations, including our own VLA 20 cm continuum imaging, an overall relatively low molecular gas density and a low SFE across over the bulk of the molecular gas distribution have been revealed, which is in sharp contrast with most LIGs/ULIGs. The strongest starburst sites are currently confined in small regions heavily obscured in the dusty patches. We conclude that the large reservoir of molecular gas is the sufficient fuel for Arp~244 to ultimately enter an ultraluminous extreme starburst phase as merging proceeds to the advanced stage. The ultraluminous phase is possibly achieved once almost all molecular gas condenses into $\\sim$kpc scale, with an order of magnitude increase in the gas surface density and more than ten-fold higher SFE."
    },
    "0010/astro-ph0010402_arXiv.txt": {
        "abstract": "Our numerical simulations of the Edgeworth-Kuiper belt objects gravitationally scattered by the four giant planets accounting for mean motion resonances reveal numerous resonant gaps in the distribution of the scattered population. ",
        "introduction": " ",
        "conclusions": ""
    },
    "0010/astro-ph0010291_arXiv.txt": {
        "abstract": " ",
        "introduction": "Emission from an active galactic nucleus (AGN) is known to exhibit apparently random variability, over a wide wavelength range from radio to X-rays and $\\gamma$-rays (e.g., Krolik et al. 1991; Edelson et al. 1996). Although the exact origin of such a variability is still unclear, it has been employed practically as a measure of the mass and size of the central black hole (BH) residing in the AGN. For example, Wandel \\& Mushotzky (1986) pointed out a strong correlation between the mass of the BHs in the AGNs estimated from optical emission lines, and the timescale of their X-ray variability in terms of the intensity  doubling time. Hayashida et al. (1998) showed that the more luminous AGNs, which may have more massive central BHs, have longer timescale of variability; they utilized power spectrum density (PSD), which is mathematically more exact than the doubling time but suffers from the window function (the Fourier transform of the observational sampling) convolved with the true power spectrum of the source. These results indicate that the intrinsic X-ray variability timescale of AGN is an important parameter, which roughly represents the mass of the central BH.  Among AGNs, high-luminosity ones including quasars (QSOs) have such a long timescale of variability, e.g. a few years, that the analysis of their variability timescale needs long observations. Such long observations in X-rays are generally limited to the all-sky monitoring of bright sources, including a limited number of QSOs (e.g., $>$ 30 mCrab for RXTE, where \"1 mCrab\" flux is $\\sim$ 3 $\\times$ $10^{-11}$ erg s$^{-1}$ cm$^{-2}$ in the 2--10 keV range).  RX~J0957.9+6903, a possible QSO, has been observed as many as 16 times over six years from 1993 to 1999 with ASCA, in which the total exposure time reaches 620 ksec. This is because the object lies close to the spiral galaxy M81, and was always within the same field-of-view (FOV) in the repeated ASCA observations of SN1993J. This object was first detected in the Einstein observation of M81, and was designated M81 X-9 (Fabbiano 1988; Elvis et al. 1992), although it is not associated with M81. Subsequently it was observed with  EXOSAT (Giommi et al. 1991) and ROSAT, and renamed RX~J0957.9+6903 (Bade et al. 1998). It was identified optically with a faint point-like object having an extreme blue continuum (Bade et al. 1998) in the wavelength range between 3400 $\\AA$ and 5400 $\\AA$. These optical properties suggest that RX~J0957.9+6903 is a QSO, even though its cosmological redshift is still unknown. On the other hand, its association with a possible supernovae remnant (SNR), at a separation of $\\sim$ 10$^{\\prime\\prime}$, was proposed from an H$_\\alpha$ observation (Miller et al. 1995).  In the present paper, we first study the 0.7--10 keV X-ray spectrum of RX~J0957.9+6903 and variability in section 3, and show them to be consistent with those of a QSO. We then analyze the long-term light curve of RX~J0957.9+6903 more quantitatively in section 4, and estimate its PSD utilizing the new method, developed by Iyomoto et al. (2000) and Iyomoto (1999) to study the long-term variation of the M81 nucleus. Because RX~J0957.9+6903 and the M81 nucleus have always been in the same ASCA FOV and observed under the same sampling window, we can directly compare the estimated PSD of these two objects, one being a promising QSO candidate while the other a typical low-luminosity AGN (LLAGN). ",
        "conclusions": "RX J0957.9+6903 has been observed 16 times with ASCA over 5.5 years. The average 0.7--10 keV GIS spectrum is represented by a power-law continuum of photon index 1.58 $\\pm$ 0.03, and an absorption column of $\\sim$ 1$\\times{\\rm 10}^{21}$ ${\\rm cm}^{-2}$. Spectral variation was insignificant except the 13th and 16th observation. The 2--10 keV flux varied by a factor of four through 6 years, and the short-term variation was 25 \\% or less on a time scale of about one day. These results, together with the optical observation, support the interpretation of RX~J0957.9+6903 as a QSO (Fabbiano 1988; Bade et al. 1998).  Through a forward-method analysis utilizing structure function, the PSD of RX J0957.9+6903 has been estimated to have a very low break frequency $f_{\\rm b}$, $1/f_{\\rm b}$ = $1600^{+\\infty}_{-1100}$ days, and the slope index $\\alpha$ = $-1.55 \\pm$ 0.2. From the mass-timescale relation for various BHs, the central BH mass of RX~J0957.9+6903 is estimated to be $> 10^{8}$ $M_{\\odot}$. The comparison of the PSDs between RX~J0957.9+6903 (QSO) and the M81 nucleus (LLAGN), estimated in the same way under the identical sampling window, indicates that the two objects have similar time scales of variation. This further supports that the central BH mass is not much different between QSOs and LLAGNs.  \\bigskip      \\clearpage"
    },
    "0010/astro-ph0010599_arXiv.txt": {
        "abstract": "We have discovered two recycled pulsars in relativistic orbits as part of the first high-frequency survey of intermediate Galactic latitudes. PSR J1157--5112 is a 44 ms pulsar and the first recycled pulsar with an ultra-massive ($M > $~1.14~\\msun) white dwarf companion. Millisecond pulsar J1757--5322 is a relativistic circular-orbit system which will coalesce due to the emission of gravitational radiation in less than 9.5 Gyr. Of the $\\sim$40 known circular orbit pulsars, J1757--5322 and J1157--5112 have the highest projected orbital velocities. There are now three local neutron-star/white-dwarf binaries that will coalesce in less than a Hubble time, implying a large coalescence rate for these objects in the local Universe. Systems such as J1141--6545 \\citep{klm+00a} are potential gamma-ray burst progenitors and dominate the coalescence rate, whilst lighter systems make excellent progenitors of millisecond pulsars with planetary or ultra-low mass companions. ",
        "introduction": "The first surveys of the Galactic plane at high frequencies for radio pulsars \\citep{cl86,jlm+92} uncovered a much younger and more distant population than the traditional 400 MHz surveys.  High frequency surveys are much less affected by the deleterious effects of scatter-broadening and dispersion smearing which make short period pulsars impossible to detect at large dispersion measures. Unfortunately, the sampling rates employed in these early surveys limited their sensitivity to millisecond pulsars (MSPs) \\citep{jb91}, and the spatial coverage was limited by the small beam of a radio telescope at high frequencies. No millisecond pulsars were discovered in these early high-frequency surveys. It was widely believed \\citep{ffb91} that MSPs were steep-spectrum objects, and that there was little point in searching for them at high frequencies.  As the population of MSPs became larger however, a series of spectral studies \\citep{lylg95,tbms98} provided evidence that spectra of millisecond pulsars were not too dissimilar from that of normal pulsars. In a simulation of the Galactic millisecond pulsar population, \\citet{tbms98} predicted that high frequency surveys undertaken with the Parkes multibeam system \\citep{swb+96} should detect a large number of millisecond pulsars. Motivated by this study we have conducted the first large-scale survey for millisecond pulsars at intermediate Galactic latitudes.  Our survey covered the region enclosed by $5\\degr < |b| < 15\\degr$ and $-100\\degr < l < 50\\degr$. Pointed observations of 265 seconds were made with the 64-m Parkes radiotelescope using the 21-cm multibeam receiver. The receiver consists of thirteen dual linear polarization feeds arranged in a sparse focal-plane pattern which allows full sky coverage with beams overlapping at the half-power points. This increases the rate of sky coverage by a factor of thirteen over a single-beam receiver and enabled us to conduct a large-scale survey in only 14 days of integration time. The system is sensitive to frequencies from 1230 MHz to 1530 MHz, with an average total system noise temperature of 24 K. The backend system was built at Jodrell Bank for the ongoing deep Galactic plane multibeam pulsar survey \\citep{lcm+00}. It includes twenty-six filterbanks, each with ninety-six channels and total bandwidth of 288 MHz, centred at a sky frequency of 1374 MHz. Filterbank outputs were summed in polarisation pairs and one-bit digitised with an integration time of 125 $\\mu$s. The large bandwidth, narrow channels, fast sampling, low system temperature and multiple beams facilitiated the discovery of a large number of pulsars, including millisecond pulsars, with greater time efficiency than any previous large survey. Data were recorded on magnetic tape for offline processing on the Swinburne supercluster, a network of 64 Compaq Alpha workstations, using standard techniques \\citep{mld+96}.  The survey is now complete. Full details of the survey will be provided in a subsequent paper (Edwards et al., in prep.). In all, we have discovered 58 new pulsars, of which one has a period of 44 ms and seven others have periods less than 20 milliseconds. In this {\\it Letter} we will discuss the two recycled pulsars for which we have phase-connected solutions, PSRs J1157--5112 and J1757--5322. ",
        "conclusions": ""
    },
    "0010/astro-ph0010550_arXiv.txt": {
        "abstract": "Eight luminous quasars with $ -30 < M_B < -27$ at z = 1.4 - 3.7 have been observed in the mid- and far-infrared using ISO. All the quasars have been detected in the mid-infrared bands of ISOCAM, while no far-infrared detections have been made with ISOPHOT. SEDs (Spectral Energy Distributions) from the UV to far-infrared have been obtained while supplementing ISO observations with photometry in the optical and near-infrared made from the ground within 17 months. The SEDs are compared with the MED (Mean spectral Energy Distributions) of low-redshift quasars with $-27 < M_B < -22$. It is shown that our far-infrared observations were limited by confusion noise due to crowded sources. ",
        "introduction": "The $IRAS$ sky survey was the first observation in which many quasars in the far-infrared were detected, and Neugebauer et al.(\\cite{neugebauer86}) presented $IRAS$ measurements of 179 quasars from 12 to 100 $\\mu m$. Combining $IRAS$ observations with data taken in other wavelengths, Sanders et al. (\\cite{sanders}) have given SEDs from $\\sim$ 0.3 nm to 6 cm of 109 quasars in the Palomar-Green (PG) survey (Green et al. \\cite{green}).  Compiling $Einstein$, $IUE$, and $IRAS$ data with a supplement of ground-based observations, Elvis et al. (\\cite{elvis}) have presented SEDs for a sample of 47 normal quasars, and derived the SEDs for radio-quiet and radio-loud quasars. Recently, Andreani et al.(\\cite{andreani}) have presented SEDs for 120 optically selected quasars at low- and high-redshift, combining sub-mm and mm observations with optical, near-infrared, $IRAS$, and radio observations.  The gross shape of quasars SEDs is characterized by two major features; the big blue bump shortward of 0.3 $\\mu m$ and the infrared bump between 2 and 200 $\\mu$m (Sanders et al. \\cite{sanders}; Elvis et al. \\cite{elvis}). It is generally considered that the blue bump is dominated by thermal emission from an accretion disk. The infrared bump is ubiquitous in both radio-quiet and radio-loud quasars.  The infrared emission in radio-loud quasars has generally been taken to be dominated by non-thermal emission (e.g., Impey \\& Neugebauer \\cite{impey}; Bloom et al. \\cite{bloom}; Neugebauer \\& Matthews \\cite{neugebauer99}).  The origin of the infrared emission in radio quiet quasars is generally attributed to thermal emission from heated dust. The rise from 1 $\\mu$m minimum toward 3 $\\mu$m, which is universally present in quasars, is naturally explained by the sublimation of dust grains at $\\sim$ 1,500 K (e.g., Kobayashi et al. \\cite{kobayashi}).  This is expected from the current unified models of active galactic nuclei, which have a dusty obscuring torus around a central source. Pier \\& Krolik (\\cite{pier}) computed the infrared properties predicted for the dust obscuring tori surrounding central sources by modeling free parameters of the inner radius of the torus to its thickness, the Thomson depths constraining the outer radius of the torus, and the flux of the nuclear radiation. The infrared emission in 1 - 10 $\\mu$m of PG quasars is well explained by their models, while the emission in the far-infrared is much greater than that predicted. Several models were proposed to account for the far-infrared emission of quasars; these are (1)warm dust in a distorted disk extending from 0.1kpc to more than 1kpc, which is heated directly by radiation from central source (Sanders et al. \\cite{sanders}), (2)dust clouds in the narrow line region heated by the radiation from the central sources and star-forming regions (Rowan-Robinson \\cite{rowan}), and (3)dust in the obscuring tori extending to 200 - 3kpc (Andreani et al. \\cite{andreani}). However, it should be noted that temporal variations of 10 $\\mu$m brightness of radio-quiet quasar PG 1535+547 found by Neugebauer \\& Matthews (\\cite{neugebauer99}) casts doubt on the thermal origin of the infrared emission in radio-quiet quasars.  High-redshift quasars have been widely recognized to provide unique probes of high-redshift star formation and galaxy evolution. Metallic abundance in high-redshift quasars is solar or higher metallicities out to $z > 4$. Comparing the UV-to-optical spectra of 186 quasars with $0 < z < 3.8$, Osmer et al. (\\cite{osmer}) concluded that there was no evidence for redshift-dependent spectral changes. Hamann \\& Ferland (\\cite{hamann}) analyzed N$_{\\mathrm{\\ V}}$/C$_{\\mathrm{\\ IV}}$ and N$_{\\mathrm{\\ V}}$/He$_{\\mathrm{\\ II}}$ broad emission ratios, and found that metallicities in the broad line gas at high redshift are 1 - 10 times solar. The ratio of UV Fe$_{\\mathrm{\\ II}}$/Mg$_{\\mathrm{\\ II}}$ of B 1422+231 at $z = 3.6$ shows the the host galaxy was already in the late-evolutionary phase of the Fe enrichment by SNe Ia at z = 3.6 (Kawara et al. \\cite{kawara96}; Yoshii et al. \\cite{yoshii}). These all suggest that the central part of host galaxies formed rapidly at very high-redshift $z \\geq 10$, and metallicity enrichment in the central part was already completed at $z = 4 - 5$.  Dust emission from high-redshift quasars provide another probe to study star formation and evolution of host galaxies, especially in outer regions. Up to date, many quasars at z $>$ 3 have been detected in the region from 350 $\\mu m$ to 1.3 mm (e.g., Andreani et al. \\cite{andreani93}; Chini \\& Kr\\\"{u}gel \\cite{chini94}; Dunlop et al. \\cite{dunlop}; Isaak et al. \\cite{isaak}; McMahon et al. \\cite{mcmahon}; Ivison \\cite{ivison}; Omont et al. \\cite{omont96a}; Hughes et al. \\cite{hughes}; Benford et al. \\cite{benford}; Carilli et al. \\cite{carilli00}). An extensive 240 GHz survey by Chini et al. (\\cite{chini89}) revealed that the majority of $z < 1$ quasars have dust masses about a few times 10$^7 M_{\\odot}$, comparable to normal spiral galaxies, thus suggesting that dust is heated by radiation from the central sources. On the other hand, dust masses $\\geq$ 10$^8 M_{\\odot}$ have been found in six of 16 quasars at $z > 4$ (Omont et al. \\cite{omont96a}), and CO emission was found in three of these (Ohta et al. \\cite{ohta}; Omont et al. \\cite{omont96b}; Guilloteau et al. \\cite{guilloteau97},\\cite{guilloteau99}; Carilli et al. \\cite{carilli99}). This may imply that dust emission in these host galaxies may be dominated by radiation from star-forming regions at high-redshift.  To study the distribution of dust in high-redshift quasars, mid- and far-infrared observations are indispensable. The emission from sublimated dust could be observed in the mid-infrared, thus probing that the innermost part of the obscuring tori, and the distribution of dust from obscuring tori to disks (or outer star-forming regions) could be obtained from far-infrared observations.  The $IRAS$ sky survey has only detected  several quasars at z $>$ 3 with exceptionally high luminosity in the far-infrared (Neugebauer et al. \\cite{neugebauer86}; Bechtold et al. \\cite{bechtold}; Irwin et al. \\cite{irwin}); F08279+5255 (APM 08279+5255) at z = 3.87, 2126-158 (PKS 2126-15) at z = 3.275, and 0320-388 at z = 3.12.  In hoping to detect more quasars, several groups have carried out mid- and far-infrared photometry using the Infrared Space Observatory ($ISO$; Kessler et al. \\cite{kessler}). The first report of the $ISO$ European Central Quasar Programme which observed 70 quasars between 4.8 and 200 $\\mu m$ including high-redshift quasars has been published by Haas et al. (\\cite{haas}), and the $ISO/NASA$ AGN Key Project has also observed 72 quasars and AGNs covering a range of redshift up to 4.7 (Wilkes et al. \\cite{wilkes}). As an $ISO$ open time program, we have executed mid- and far-infrared observations of eight quasars at $1.4 < z < 3.7$ using the raster mapping mode.  This paper presents the results of the photometry of these quasars supplemented with optical and near-infrared data taken on the ground. In the course of this work, mid- and far-infrared sources have serendipitously been discovered. These sources will be described in the forthcoming paper (Oyabu \\& Kawara \\cite{oyabu}). Throughout this paper, $H_0$ = $75km\\ s^{-1}\\ Mpc^{-1}$ with $q_0$ = $0$ is assumed. ",
        "conclusions": "\\begin{figure*}[htbp] \\resizebox{18cm}{!}{\\includegraphics{oyabuf3.eps}} \\hfill \\parbox[b]{18cm}{ \\caption{Restframe SEDs compared with the MED of low-redshift quasars compiled by Elvis et al. (\\cite{elvis}). Dashed lines show the 1 $\\sigma$ deviation (actually 68 Kaplan-Meier percentile) envelopes. Asterisks denote data taken by Richards et al.(\\cite {richards}) and Schneider et al.(\\cite {schneider}). The dot-dashed and dot lines for B 1422+231 shows the MED of radio-loud quasars from Elvis et al.(\\cite{elvis}) and the SED of 3C273 from Lichti et al.(\\cite{lichti}), respectively. Note that 3 $\\sigma$ upper limits correspond three times uncertainties listed in Table \\ref{tab3}. } \\label{fig:sed}} \\end{figure*}  All the quasars have been detected in the ISOCAM LW2 and LW3 or LW10 band. Combining optical and near-infrared observations, SEDs from the UV to mid-infrared have been obtained for eight quasars at z = 1.4 - 3.7. Fig. \\ref{fig:sed} shows the restframe SEDs ($L_{\\nu}$\\footnote{$L_{\\nu}$ is defined by $4\\pi D_L^{2} \\nu f_{\\nu}$ where $D_L$ denotes the luminosity distance to the object.}). The down arrows indicate the 3 $\\sigma$ upper limits obtained with the ISOPHOT observations. Note that 3 $\\sigma$ upper limits in Fig. \\ref{fig:sed}  correspond to three times uncertainties listed in Table \\ref{tab3}. The mean spectral distributions (MEDs) of low redshift quasars by Elvis et al.(\\cite{elvis}) are also shown by fitting to the $ISO$ $6.7\\mu$m flux density. Note that $6.7\\mu$m in the observing frame is approximately $1.25\\mu$m in the restframe. The dash lines indicate the 1 $\\sigma$ deviation (strictly speaking 68 Kaplan-Meier percentile) envelopes. The rise from the 1 $\\mu$m minimum toward 3 $\\mu$m is evident at least in $z = 3.5$ quasar PC 1548+4637. This rise is naturally explained by the sublimation of dust grains at $\\sim$ 1,500 K (e.g., Kobayashi et al. \\cite{kobayashi}) as expected from  the current unified models of active galactic nuclei.  Hence, our data suggest that the obscuring torus model could be applied to high-redshift quasars.  The original objective of this work was to study whether there was any evidence for redshift-dependent changes in SEDs from the UV to far-infrared. As shown in Fig. \\ref{fig:sed}, the data from the UV to mid-infrared do not deviate significantly from the MED of the low-redshift sample. However, this cannot be taken as evidence for no SED evolution in the UV to mid-infrared, because our sample is too small in number and too narrow in spectral coverage. Our far-infrared observations only provide 3 $\\sigma$ upper limits, and are not very helpful to check SED evolution in the far-infrared.  The far-infrared detection with $ISO$ have been severely limited by the confusion due to galaxies and local peaks of the IR cirrus (Kawara et al. \\cite{kawara98}; Herbstmeier et al. \\cite{herbstmeier}; Matsuhara et al. \\cite{matsuhara}).  Simulating the 90 and 170 $\\mu$m source counts  in the Lockman Hole, Kawara et al. (\\cite{kawara00a}, \\cite{kawara00b}) concluded that the effect of confusion due to crowded sources (presumably extragalactic) was significant in the flux range below 200 mJy, and 50\\% of 140 mJy sources at 170 $\\mu$m were left undetected due to confusion noise.  The expected fluxes of our sample quasars ranges from 7 - 170 mJy at 170 $\\mu$m.  Such confusion is clearly seen in Fig.~\\ref{map}.  There are a few sources scattering around the quasar, which hampers detection in the far-infrared. It should be noted that the brightest source in the PG 1630+377 field has 106 mJy and the one in PG 1715+535 has 190 mJy (Oyabu \\& Kawara \\cite{oyabu}).  It is clear that larger aperture telescopes like FIRST, which provide spatial resolution finer than ISO, are required for far-infrared studies of high-redshift quasars. In cases where similar aperture telescopes such as IRIS(ASTRO-F) and SIRTF are used, data sampling, namely a step size of raster mapping, should be carefully designed so that far-infrared images can be improved by deconvolution without violating the sampling theorem (Magain, Courbin, \\& Sohy \\cite{magain}).  \\acknowledgement We wish to thank the staff of Vilspa, CTIO, OAO, and the Kiso observatory for their assistance and hospitality.  We are grateful to T.Tsuji for his continuous support to this work. We thank the referee, P. Andreani, for valuable suggestions on how to improve this work."
    },
    "0010/astro-ph0010285_arXiv.txt": {
        "abstract": "Observations of core-collapse supernovae (SNe) have revealed the presence of extensive mixing of radioactive material in SN ejecta. The mixing of radioactive material, mostly freshly synthesized Ni, is not complete, which leads to a two-phase SN ejecta structure. The low-density phase consists of Fe bubbles, created by the energy input from radioactive Co and Ni, surrounded by compressed high-density metal-rich ejecta.  We report on the theoretical investigation of supernova remnant (SNR) dynamics with the two-phase SN ejecta. We first present 3-dimensional hydrodynamic simulations of a single Fe bubble immersed in an outer ejecta envelope, and compare the results with previous work on shock-cloud interactions.  We then consider randomly distributed Fe bubbles with an average volume filling fraction of $1/2$. We find that the presence of Fe bubbles leads to vigorous turbulence and mixing of Fe with other heavy elements and with the ambient normal-abundance gas. The turbulent energy can be an order of magnitude larger than in the case of smooth ejecta. A significant fraction of the shocked ejecta is found in narrow filaments and clumps moving with radial velocities larger than the velocity of the forward shock. Observational consequences of the two-phase ejecta on SNR X-ray spectra and images are briefly mentioned. ",
        "introduction": "The presence of large-scale mixing of ejecta in core-collapse supernovae (SNe) has been well established on both observational and theoretical grounds. The early, unexpected emergence of X-rays and $\\gamma$-rays shortly after the explosion of SN 1987A provided  dramatic evidence for mixing of radioactive Ni throughout the He core and H-rich envelope. Extensive mixing of Ni is also required to explain the light curve and spectral observations of SN 1987A (see McCray 1993 for a review of SN 1987A). From studies of light curves and spectra of numerous core-collapse SNe, it is now known that such large-scale mixing is nearly always present. Another line of evidence is provided by studies of meteoritic graphite grains which condensed in SN ejecta (Travaglio et al. 1999 and references therein).  Such grains apparently contained radioactive, short-lived isotopes such as $^{44}$Ti at the time of their formation, which implies extensive mixing of freshly-synthesized material deep in the SN ejecta with the C-rich layer at the bottom of the He core. This overwhelming observational evidence for mixing is consistent with our present understanding of the collapse and explosion of massive stars. The neutrino-driven Rayleigh-Taylor instability just outside the proto-neutron star leads to convective motions, strengthening the post-bounce shock and maybe even making the explosion possible in many cases (e.~g., Mezzacappa et al. 1998; Kifonidis et al. 2000). Further mixing occurs later during the SN explosion because of Rayleigh-Taylor instabilities generated at interfaces between stellar layers with different chemical composition (e.g., Fryxell, Arnett, \\& Mueller 1991; Hachisu et al. 1992; Kifonidis et al. 2000).  Large-scale turbulence generated during the explosion appears not to be sufficient to mix SN ejecta completely, but only macroscopically.  In Cas A, optical observations revealed the presence of ejecta with very different chemical abundances, such as O-, S-, Ar-, and Ca-rich ejecta knots, which were macroscopically mixed during the SN explosion as evidenced by the lack of spatial stratification expected in the absence of mixing (e.~g., Fesen \\& Gunderson 1996). The most recent IR and X-ray observations of Cas A provide further evidence for this macroscopic but not microscopic mixing (Arendt et al. 1999; Douvion et al. 1999; Hughes et al. 2000; Hwang et al. 2000).  Spectroscopic observations of numerous SNe also confirm the presence of clumpy ejecta (Spyromilio 1994). Detailed studies of SN 1987A lead to the same conclusion.  In SN 1987A, the inhomogeneities take a particularly interesting form: Fe-Ni bubbles (Li et al. 1993), inferred from observations of Fe, Co, and Ni lines.  Ni-rich ejecta clumps, transported outwards by turbulent motions, are heated by their own radioactive energy input, and expand in the ambient substrate of other heavy elements, forming low-density Fe bubbles. The SN structure is then expected to be like a Swiss cheese, with Fe bubbles occupying a substantial ($\\sim 0.5$) fraction of the ejecta volume (Li et al. 1993; Basko 1994).  These bubbles should lead to vigorous turbulence and mixing in young supernova remnants (SNRs) in general.  Most of ejecta in young SNRs can be seen in X-rays, where modern X-ray observatories such as {\\it Chandra} and XMM-{\\it Newton} have begun to provide high-quality information. Interpretation of these observations must be done in the framework of appropriate hydrodynamical models. But realistic simulations for young SNRs, which include the presence of Fe bubbles in SN ejecta, are lacking. We report here on such 3-dimensional hydrodynamical simulations (preliminary results based on 2-D simulations were reported by Borkowski et al. 2000). Our goal is to study the basic hydrodynamics of the interaction by investigating a single Fe bubble, and to explore how the presence of multiple Fe bubbles changes the global dynamics of young SNRs. ",
        "conclusions": "Our primary conclusion is that large-scale inhomogeneities associated with the presence of Fe bubbles in heavy-element ejecta lead to vigorous turbulence and mixing in young SNRs. The turbulent energy density may be increased by an order of magnitude when compared with homogeneous ejecta. This turbulence radically changes the spatial structure of the interaction region between the blast wave and the reverse shock. While the time-averaged dynamics and the propagation of the forward shock are well described by a self-similar solution \\cite{chevalier82,cbe92}, the reverse shock geometry and the structure of the shocked region are very different in the presence of the Fe bubbles. The reverse shock is no longer approximately spherically symmetric, and jet-like fingers of ejecta can even affect the location of the blast wave. Angle-averaged density profiles rise smoothly toward the remnant's interior, with no signs for the presence of a spatially distinct shell of ejecta.  Pressure and temperature variations of orders of magnitude are also present, unlike the case of homogeneous ejecta which can be described by well-defined density, pressure, and temperature profiles. Such large variations should dramatically influence X-ray emission, which depends strongly on temperature and density. In particular, low temperature, dense ejecta are particularly efficient in producing X-ray emission, and may dominate X-ray spectra. But this emission is generally produced by ejecta which suffered the least amount of deceleration, so that their radial velocities and proper motions may be even higher than that of the blast wave. This example demonstrates that standard SSDW solutions should be used with an extreme caution for young SNRs with inhomogeneous ejecta. The amount of turbulence generated by the presence of Fe bubbles in SNRs most likely varies from remnant to remnant because of large ($\\sim 100$) variations in the radioactive Ni yields in core-collapse SNe. In addition, the amount of turbulence might be lower for those SNRs which did not have enough time to enter the self-similar turbulent regime, because of long timescales necessary to achieve it. But unless mixing of heavy elements is somehow inhibited during a SN explosion, it should be clear that one-dimensional models are clearly not acceptable for remnants of core-collapse SNe, and that multidimensional hydrodynamical modeling is essential for understanding dynamics of heavy-element ejecta in such remnants.  It is also possible that Ni bubbles can be formed in Type Ia supernova explosions (Wang \\& Chevalier 2000), so that our results may be applicable there as well.  We expect that small-scale turbulence will have a strong effect on the synchrotron radio morphology of a young remnant.  In 2-D MHD simulations of the SSDW phase of a young Type Ia ($s = 0$) supernova remnant, Jun \\& Norman (1996) found that turbulent energy eventually rose to about 0.6\\% of the kinetic energy in the remnant, and that random magnetic energy rose as well, to a level of about 0.3\\% of the turbulent energy.  The random magnetic-field energy did not track the total turbulent energy precisely, but spatial locations of strong turbulence (Kelvin-Helmholtz unstable edges of Rayleigh-Taylor fingers) were also the locations of strong magnetic-field amplification.  In synchrotron visualizations, Jun \\& Norman (1996) also found that regions of strong magnetic field produced strong synchrotron emission, but their simulation used a relatively simple description of relativistic-electron acceleration and transport.  We expect that magnetic fields in a bubble simulation would also track locations of turbulence and vorticity, that is, should indicate bubble walls, and ought to be noticeable in synchrotron images.  In addition, the far higher levels of turbulent energy we find compared to smooth ejecta models such as Jun and Norman's should result in considerably higher magnetic energy densities and consequent higher synchrotron emissivities.  However, hydrodynamic simulations including magnetic-field tracking and particle acceleration and subsequent evolution will be necessary to make more definite predictions.  Our results are in qualitative agreement with the observed morphology of Cas A, undoubtedly the best example of a young SNR with ample evidence for vigorous mixing. Our simulations are able to reproduce filamentary ejecta emission  seen in {\\it Chandra} images. The ragged appearance of Cas A, including its ``jet'' feature, may also be a consequence of an interaction of bubbly ejecta with the ambient medium, and not necessarily the result of a strongly asymmetric explosion. It is also tempting to identify its rings of optical knots (Reed et al. 1995) with the most dense ejecta clumps at the boundaries of Fe bubbles. Our current simulations are however not suitable for a detailed modeling of the Cas A dynamics because of the evidence for the dynamically important shell of circumstellar matter in Cas A \\cite{cl89}. The circumstellar interaction, studied by us through one-dimensional hydrodynamical simulations \\cite{bsbs96}, must be now simulated in 3-D in the framework of bubbly ejecta. A quantitative study of spatial morphology of Cas A, coupled with a more detailed examination of spatial structures in hydrodynamical data sets, would also be useful.  X-ray observations should provide a most complete test of the two-phase model, because they probe the bulk of the shocked material. While modeling of X-ray emission based on multidimensional calculations clearly demands a separate effort, we briefly outline what major effects are expected. If most Fe initially resides in low-density shocked bubbles, while other heavy elements are located in the dense phase of SN ejecta, then Fe lines should generally be much weaker than lines from other abundant heavy elements. Fe lines indeed appear to be weaker than expected in Cas A \\cite{bsbs96,vkb96,favata97} when compared with strong Si, S, Ar, and Ca lines, based on nucleosynthetic yields of core-collapse SNe. Unlike in the optical and IR, freshly synthesized Fe was detected by {\\it Chandra} \\cite{hughes00, hwang00}, but its spatial distribution is different than that of Si- and S-rich ejecta. This means that not all Fe resided in low-density bubbles, either because of microscopic mixing of Fe at the bubbles' boundaries or because radioactive Ni was mixed to sufficiently high velocities where the bubble growth was inhibited by escape of $\\gamma$-rays from Ni clumps. Quantitative analysis of {\\it Chandra} Cas A data, coupled with hydrodynamical simulations and X-ray modeling, should provide a stringent test of the two-phase ejecta model with Fe bubbles.  As remnants become older, turbulent mixing will lead to a gradual strengthening of Fe lines with respect to other heavy elements. Even if Fe is microscopically mixed with other elements during the SNR evolution, its ionization age should lag behind that of other elements. In the next few years, a combination of new X-ray observations and sophisticated hydrodynamical and X-ray modeling should provide us with a vastly better understanding of how chemical elements are ejected in explosions of massive stars and how they are dispersed into ambient interstellar medium."
    },
    "0010/astro-ph0010416_arXiv.txt": {
        "abstract": "We present {\\sl XMM-Newton} X-ray spectra of the disc accreting cataclysmic variable OY Car, which were obtained during the performance verification phase of the mission. These data were taken 4 days after a short outburst. In the EPIC spectra we find strong iron K$\\alpha$ emission with weaker iron K$\\beta$ emission together with silicon and sulphur lines. The spectra are best fitted with a three temperature plasma model with a partial covering absorber. Multiple temperature emission is confirmed by the emission lines seen in the RGS spectrum and the H/He like intensity ratio for iron and sulphur which imply temperatures of $\\sim$7keV and $\\sim$3keV respectively. ",
        "introduction": "Cataclysmic variables (CVs) are close binary systems in which the secondary (usually a dwarf main sequence star) fills its Roche lobe and transfers material onto the white dwarf primary. In non-magnetic systems, this material forms an accretion disc around the primary. Some of these systems show `dwarf nova' outbursts on the timescale of weeks to months when the system brightens by several magnitudes lasting a few days to months. During an outburst, the material close to the white dwarf is optically thick, while in quiescence the gas is optically thin (Pringle \\& Savonije 1979, Narayan \\& Popham 1993, Popham \\& Narayan 1995).  X-ray observations of dwarf novae are hampered by their relatively low flux levels. With observations using large effective areas, such as {\\sl XMM-Newton} (Jansen et al, 2001), we can obtain high signal to noise X-ray spectra of dwarf novae for the first time. In this paper we present and analyse X-ray spectra of the dwarf nova OY Car.  In a companion paper (Ramsay et al 2001), we present and analyse the light curves of OY Car. ",
        "conclusions": "We have examined the X-ray spectra of OY Car obtained using {\\sl XMM-Newton}. These spectra have a much higher signal to noise and spectral resolution than previous X-ray spectra of dwarf nova. We find strong emission lines of various ionisation species. In fitting the EPIC spectra we require a multi-temperature plasma. This is confirmed by the line species seen in the RGS spectrum and also from the H/He-like intensity ratios of iron and sulphur. Adding a more complex absorber such as a warm absorber or a partial covering model to a neutral absorber significantly improves the fit to the data."
    },
    "0010/physics0010047_arXiv.txt": {
        "abstract": "After  the dynamo experiment in November 1999 \\cite{PRL1} had shown magnetic field self-excitation in a spiraling liquid metal flow, in a second series of experiments emphasis was placed on the magnetic field saturation regime as the next principal step in the dynamo process. The dependence of the strength of the magnetic field on the rotation rate is  studied. Various features of the saturated magnetic field are outlined and possible saturation mechanisms are discussed. ",
        "introduction": " ",
        "conclusions": ""
    },
    "0010/astro-ph0010620_arXiv.txt": {
        "abstract": "It is important to determine whether the observed bow shocks in the working surfaces of Herbig-Haro outflows has lead to a destruction of dust grains and consequently to a change in the gas phase metal abundances (say of Fe) in the cooling regions of HH bow shocks. Detailed studies are currently available for only 5 HH outflows (Beck-Winchatz et al.\\ 1996). This small number is due to the large observational and theoretical effort required to determine metal abundances in HH objects.  Information about metal abundances in more HH objects is badly needed. We therefore use a very approximate method. We introduce a ``characteristic number,'' $A_{\\rm me}$, whose definition is based only on the often observed line fluxes of [Fe II] 5159, [Fe II] 7155, [Ca II] 7291, as well as H$\\alpha$ and H$\\beta$.  These fluxes can easily be determined from existing observations.  We find a good correlation between $A_{\\rm me}$ and the Fe abundance for the 5 well-studied HH objects. We use this correlation to determine approximate values of the gas phase Fe abundance in 13 additional high excitation and in 3 additional low excitation HH objects.  The results are the following: Of the 16 high excitation HH objects studied, there are 6 which approximately agree with normal population I abundance (i.e., no depletion due to dust formation).  The remaining 10 show some (very modest) Fe gas phase depletion which, however, never gets larger than by a factor of 2.5.  This result is in agreement with our qualitative expectations that fast shocks efficiently destroy dust grains.  Of the 5 low excitation HH objects studied, there are 4 which show a normal population I abundance (strictly speaking, an even slightly higher abundance than this). This is completely unexpected. In low excitation objects, one might expect strong gas phase Fe depletion (showing the unchanged molecular cloud composition), unless the matter has previously gone through shocks of much higher shock velocities. We discuss this possible explanation and the question of whether low excitation HH objects have a different ``history'' than usually assumed. ",
        "introduction": "It has been generally accepted that the Herbig-Haro outflows (jets) from young stellar objects become optically visible through the shocks (bow shocks and jet shocks, Mach disks) at their working surfaces (Mundt \\& Fried 1983; Hartmann \\& Raymond 1984; B\\\"ohm \\& Solf 1985; Schwartz 1985; Mundt 1985; Hartigan, Raymond \\& Hartmann 1987; Reipurth 1989; Hartigan 1989; Solf \\& B\\\"ohm 1991; Reipurth \\& Heathcote 1993; Eisloeffel, Mundt \\& B\\\"ohm 1994; Raga 1995).  In other words, in HH objects we see matter (if it is observed in a bow shock) that originally comes from the molecular cloud and has recently gone through the shock and through an essential part of the cooling region of the shock.  Recently, it has also become clear that, in a number of Herbig-Haro (HH) objects, we can identify ``internal working surfaces'' and their bow shocks (Raga 1995; Reipurth \\& Heathcote 1997; Reipurth et al.\\ 1997; Raga \\& Noriega-Crespo 1998).  That is, in a number of HH objects (e.g., in HH 111 or in HH 34), the cloud matter has passed through multiple shocks.  In molecular clouds (including the boundary regions of the clouds, which are accessible to optical HH observations), we expect that the gas phase abundances of elements like Fe or Ca are considerably depleted (by a factor of on the order of 100) because of the formation of dust grains (van Dieshoeck et al.\\ 1993). If one takes into account the fact that refractory elements like Fe and Ca can return to the gas phase by the destruction of the dust grains in sufficiently strong shocks ($v = 150$ km s$^{-1}$; e.g., McKee et al.\\ 1987; Seab 1987; Draine 1995), we might expect that HH objects will show different Fe/H or Ca/H ratios that depend on the velocity of the shock through which the visible matter has passed before it is observed.  The (admittedly very naive) expectation would be that ``low excitation objects'' (slow shocks, $v \\la 50$ km s$^{-1}$), would have rather small gas phase Fe and Ca abundances because these elements remain in the dust grains.  In ``high excitation objects'' (fast shocks, $v > 85$ km s$^{-1}$), we might expect the dust grains to be destroyed, at least partially, by the shock.  The gas phase abundance of elements like Fe or Ca would be high (solar?).  This hypothesis inspired Beck-Winchatz, B\\\"ohm, \\& Noriega-Crespo (1994, hereafter BBN94; 1996, hereafter BBN96) to determine the Fe gas phase abundance for a few objects.  In both papers the authors decided that it would be most convenient (for the study of dust grain survival) to determine the Fe abundance (they derived the abundance ratio Fe/S and Fe/O).  The effort required for carrying out this program for a single HH object is sizable. It requires reliable line spectrophotometry with fairly high spectral resolution in a fairly large wavelength interval and the detailed study of the statistical equlibrium of Fe (and a comparison element, either O or S).  For this reason the program was carried out for only 5 HH objects, namely for HH 1, HH 7, HH 11, HH 43A, and HH 255 (Burnham's Nebula; BBN94, BBN96).  Many of their results are surprising and do not (in general) agree with our above mentioned expectations.  First, the Fe abundance for both HH1 and HH7 are essentially solar (with only 18\\% and 3\\% deviation, respectively).  This is surprising because HH 1 is the standard high excitation object, while HH 7 is one of the lowest excitation objects known (so we expect a low Fe abundance).  Second, the Fe abundance of the low excitation object HH 11 is (for unknown reasons) somewhat higher (by a factor of 1.68) than the solar abundance.  Finally, the authors detected some depletion of Fe in HH 43A (Fe/O about 38\\% and Fe/S about 48\\% of the solar value) and HH 255 (Fe/O about 71\\% and Fe/S about 36\\% of the solar value), though these are high excitation objects. In judging these results we have to be aware that HH 255 is a somewhat complicated HH object (see B\\\"ohm \\& Solf 1997; Solf \\& B\\\"ohm 1999), to which some of the standard assumptions (e.g., that the line emission is exclusively caused by a bow shock in the working surface of the outflow) may be only partly applicable.  Even if we exclude for the moment HH 255 from our consideration, it is clear that the earlier results do not agree with the simple expectations expressed above.  One way to explain why the low excitation objects (HH 7 and HH 11) show no Fe depletion is to assume that the shocked gas we observe has passed through a much faster shock at some earlier time.  In order to clarify this problem, it is important to determine the gas phase Fe abundance (and correspondingly the destruction of dust grains) in a larger number of HH objects.  However, as discussed above, this is very difficult to do in detail.  We therefore consider it presently justified to try to determine preliminary, (very) approximate values for the Fe abundance in a larger number of HH objects, even if they are considerably less reliable than the earlier determinations of BBN96.  First, we want to mention related studies by other authors.  Hartigan et al.\\ (1999, 2000) derived gas phase Fe abundance in HH objects which approximately agree with BBN94 and BBN96. In their 1999 paper, they studied the Fe II and [Fe II] spectrum for HH 47A in unusual detail, covering the bow shock and the Mach disk separately.  Their conclusion was that the gas phase Fe abundance is approximately solar.  It is also worth noting that there has been another recent study of the gas phase Fe abundance which gave a somewhat lower abundance (on the average by a factor of 3) for HH objects by Mouri \\& Taniguchi (2000).  They use a plane shock wave model to calculate the theoretical line ratio [Fe II] 8617/[O I] 6300 in considerable detail and compare this to observations of HH objects (which are not individually listed).  We agree (of course) that HH objects are caused by shock waves, but an attempt to explain in quantitative detail individual spectra by a (plane or bow shock) model leads to difficulties \\citep{ragaea96,bg97}.  For this reason, the approach of BBN94 and BBN96 was completely different.  BBN94 and BBN96 studied the HH objects individually and tried to introduce (where possible) observationally determined parameters (empirically determined densities, sizes, and electron temperatures of the line emitting regions in HH objects).  They used different methods to get information about the ionization.  In one method, for example, BBN96 determined the abundance ratio \\begin{equation} {{\\rm N(Fe)}\\over{\\rm N(O)}} = {{\\rm N[Fe\\: II] + N[Fe\\: III]}\\over {\\rm N[O\\: I] + N[O\\: II] + N[O\\: III]}} \\nonumber \\end{equation} (where N[X] is the number of [X] ions).  (This does not require detailed theoretical information about the ionization.)  They also based the derived abundance information on many lines (e.g., 42 [Fe II] lines in HH 1).  Since the approaches of \\citet{mt00} and BBN94/96 are completely different (one based on a detailed, but basically very simple model, the other on determining the empirical parameters for individual HH objects), we presently cannot say which is more correct.  This will depend on obtaining more insight into the detailed structure of individual HH objects.  However, our following approach has been to employ knowledge of individual objects, and consequently, we follow the work of BBN96.  We proceed as follows: In \\S \\ref{method}, we introduce a characteristic number which uses the line fluxes of only a very few lines which are observed in reasonably many HH spectra (which are not taken with the special purpose of determining the Fe abundance, see Raga, B\\\"ohm, \\& Canto 1996 for a reference to appropriate spectra). We show that the characteristic number is well correlated with the actual Fe abundance for the 5 HH objects for which the Fe abundance has been determined in detail (BBN96).  In \\S \\ref{application}, we use this correlation to determine approximately the gas phase Fe abundance in the other HH objects.  Finally, in \\S \\ref{conclusions}, we discuss the enigmatic results and suggest possible explanations. ",
        "conclusions": "}  Earlier detailed studies of the gas phase Fe abundance for a very few HH objects had led to the conclusion that there is no simple correlation between the Fe abundance and the HH shock velocity (BBN94, BBN96).  A correlation would have been expected if the molecular cloud matter enters the HH working surface and dust grains are destroyed in the HH bow shock by the usual processes (mostly by non-thermal sputtering, see e.g., McKee et al.\\ 1987; Draine 1995). Since the study of BBN96 is based on only 5 objects (HH 1, HH 7, HH 11, HH 43A, HH 255), and adding more objects to this list using a detailed method requires great effort, we have derived Fe abundances of 16 additional HH objects using a very approximate method.  We introduced a characteristic number, $A_{\\rm me}$ (Eqn. \\ref{fit}), which depends on the (frequently measured) line fluxes of [Fe II] 5159 and [Fe II] 7155, as well as on the [Ca II] 7291 line flux and the H$\\alpha$ and H$\\beta$ fluxes.  For the five HH objects (mentioned above) for which we have a detailed gas phase analysis, $A_{\\rm me}$ is correlated reasonably well with the gas phase Fe abundance. If we assume that this correlation is not only approximately correct for the 5 objects studied in detail, but for all HH objects for which we can determine $A_{\\rm me}$, then we can use this correlation to determine their gas phase Fe abundance.  For the high excitation HH objects, our results are the following.  In addition to the 3 high excitation objects for which the Fe gas phase abundances have been measured in detail (BBN96), we have now determined the Fe gas phase abundance for 13 more objects. Of these 16 objects, 4 show a gas phase Fe abundance between 1.0 and 1.44 times the solar abundance. Taking into account the possible errors of the abundance determination, this probably means that 4 of these objects have unchanged population I Fe gas phase abundances (no Fe bound in dust grains). The other 12 high excitation objects show some (but relatively small) Fe depletion which never gets larger than a factor of 2.5.  In general, the result looks reasonable. The high excitation HH objects show that, in some cases, Fe has gone back completely to the gas phase, and in other cases, some (but not too much) Fe may still be bound in dust grains, leading to moderate depletion of the gas phase abundance. We also find that similar depletion exists in different parts of the same object (e.g., in HH 2A and HH 2G or in HH 54B and HH 54C).  The gas phase Fe abundance for high excitation objects agrees qualitatively with the predictions given, e.g., by McKee et al.\\ (1987) that dust grains are destroyed by strong shocks. The limit of the depletion factor of 2.5 is interesting but has not been explained in detail.  Understanding the results for the low excitation HH objects seems to be more difficult, and some questions remain open. The first problem is that we could study only 5 objects. For these objects we find quite unexpected results. Four of them have an Fe gas phase abundance between 1.02 and 1.86 times the solar Fe abundances.  One may argue that our uncertainty (which is supposedly larger in the low excitation than in the high excitation objects) might permit these to be consistent with 1 solar Fe abundance.  However, we still have the problem that low excitation objects (for which we do not expect destruction of dust grains) apparently have all or most of their Fe in the gas phase.  It is also important that we find no correlation of the Fe abundance with shock velocity.  While this is not surprising for the high excitation objects, we would have (naively) expected a large difference between the low and high excitation objects.  Specifically, we would have expected a strong Fe depletion (perhaps by a factor of 100) in the low excitation objects, since their shocks are not strong enough to destroy grains \\citep{mckeeea87,draine95}.  The (perhaps) simplest explanation is that the material we have studied in low excitation objects has passed through a faster shock at earlier times. This may be a surprising result, but we cannot see any way to avoid it.  It is difficult to speculate about this process.  What would be required is that the matter that enters the low excitation bow shock (which we assume to be molecular cloud matter) has previously gone through a high excitation bow shock (in which the dust grains were destroyed).  The simplest assumption would be that the previous high excitation bow shock was generated by the same source star that generates the low excitation HH outflow.  This suggests that none of the five low excitation objects studied in this work correspond to the leading edge of the stellar outflow (i.e., all of them are produced some time after the onset of the jet).  It is also interesting to note that we have assumed the molecular clouds in which these HH objects reside have approximately solar Fe abundance (in gas and dust phases).  Though this is a reasonable assumption, it may be possible that the abundances in present-day molecular clouds could differ from solar by a factor of a few.  The entire range of Fe abundances of the objects we have studied is from $\\sim 0.4$ to 1.9 times solar.  In light of this, one may argue that the dust grains are completely destroyed (by passing through fast shocks) in all 21 objects, and that the observed spread in Fe abundance is due to different abundances in their parent molecular clouds.  At present, our data cannot resolve this issue."
    },
    "0010/astro-ph0010146_arXiv.txt": {
        "abstract": "We obtained HST/WFPC2 images of the dwarf spheroidal (dSph) galaxies K61, K63, K64, DDO78, BK6N, and kk77 in the M81 group. Our color-magnitude diagrams show red giant branches with tips (TRGB) falling within the range of $I = [23.8 -24.0]$ mag. The derived true distance moduli (DM) of the 6 dSphs ranging from 27.71 to 27.93 mag are consistent with their membership in the group. Given accurate distances of 5 other group members, which have been derived via TRGB or cepheids, the mean DM of the M81 group is $(27.84\\pm0.05)$ mag. We find the difference of the mean distances to the M81 and NGC~2403 groups to be $D_{\\rm M81} - D_{\\rm NGC2403} =(0.5\\pm0.2)$ Mpc, which yields a deprojected separation of 0.9 Mpc. With respect to the Local group, M81 and NGC~2403 have radial velocities  of 106 and 267 km~s$^{-1}$   respectively, while the velocities of the group centroids are 142 and 281 km~s$^{-1}$. The higher velocity of the closer system may indicate that these groups are moving towards each other, similar to the Milky Way and M31 in the Local group. Several globular cluster candidates have been identified in the galaxies. ",
        "introduction": "Until recently most studies of dwarf spheroidal galaxies have been carried out in the Local Group only. The properties of the Local Group sample have been reviewed by  Grebel (1997), Da Costa (1998), and Mateo (1998). But now progress is being made in the study of the dwarf population of other nearby groups, such as the M81 group, thanks to the efforts of various authors. Being among the faintest galaxies and of very low surface brightness, dSphs are rather difficult to recognize optically in the sky. As gas-poor galaxies, they are usually undetectable in the  H{\\sc I} line as well. In the case of the M81 group an additional observational difficulty arises, because of the group location in an area of the sky that is contaminated with  dense Galactic  H{\\sc I} emission (Appleton et al. 1993) and Galactic cirrus (Sandage 1976). Therefore, the only way to establish the true dwarf spheroidal membership in the M81 group is by measurement of their distances via photometry of their stellar populations. Below we briefly outline the history of discovery of dwarf spheroidal galaxies in the M81 group.  Van den Bergh (1959) published the DDO catalog of low surface brightness objects with angular diameters $a> 1\\arcmin$ revealed by visual inspection of the POSS-I plates. In the vicinity of M81 he found a concentration of LSB galaxies, which he classified as iregulars and spheroidals. As a prototype of dSphs, the LG dwarf spheroidal system in Draco was chosen. Out of 13 supposed dwarfs in the vicinity of M81, the objects DDO~44, 71, 78, and 87 were classified as dSphs. Later van den Bergh (1966) ascribed them to luminosity class V, without morphological type indication. Later DDO 87 was detected in  H{\\sc I} by Fisher and Tully (1975) and reclassified as dIrr.  Karachentseva (1968) made an independent visual all-sky survey of the POSS-I prints for probable Sculptor-type \\begin{figure*}[hbt] \\caption{Digital sky survey images of eight dwarf spheroidal members of M81 group. The field size is 6$\\arcmin$, North is up, and East is to the left.} \\end{figure*} \\begin{figure*} \\caption{WFPC2 images of six dSph galaxies: kk077, K61, DDO~71, K64, DDO~78, and BK6N produced by combining the two 600 s exposures through the F606W and F814W filters. Each galaxy is centered in the WF3 chip (WF3-FIX mode). The arrows point to the North and the East. Globular cluster candidates are indicated by circles.} \\end{figure*} \\begin{figure*}[hbt] \\caption{ $R$ images of K61 (a) and K64 (b) obtained with the 6-m telescope with exposures of 1200 s. Both the images have about the same orientation (top is NE).} \\end{figure*}  \\begin{figure*}[hbt] \\caption{Color-magnitude diagrams from the WFPC2 data for six dSph galaxies. The three panels for each galaxy show diagrams based on stars within the central (WF3) field, the \"medium\" field (the neighbouring halves of the WF2 and WF4 chips), and the outer field (remaining halves of the WF2 and WF4 chips). Each of these three fields covers an equal area of 800$\\times$800 pixels. The solid lines in the left panel show the mean loci of the red giant branches of globular clusters with different metallicities, M15 ($-$2.17 dex). M2 ($-$1.58 dex), and NGC~1851 ($-$1.29 dex.) from left to right, based on Da Costa \\& Armandroff (1990).} \\end{figure*} \\setcounter{figure}{4} \\begin{figure*}[hbt] \\caption{Globular cluster candidates shown in the F606W+F814W filters from HST. In case of K64 the central background galaxy is presented.} \\end{figure*} dwarf galaxies with an angular diameter limit of 0$\\farcm5$. Around M81 she found 14 objects, and some of the largest of them were identified with the DDO objects. New supposed dSphs were K59, K61, and K64 ( = UGC~5442, Nilson 1973). Later Mailyan (1973) rediscovered the objects K61 (=Mai 47), DDO~71 (=Mai 49), and K64 (=Mai 50) on the POSS-I prints.  B\\\"{o}rngen and Karachentseva (1982) carried out a special photographic survey with the 2-m Tautenburg Schmidt telescope to search for new dwarf galaxies in the M81 group. All previously known objects were confirmed, and some previously uncatalogued objects were found, in particular, BK2N (dSph/dE),  BK5N (dSph), and BK6N (dSph).  To clarify the morphology of supposed dwarf members in the M 81 group, about 40 objects were imaged with the 6-m telescope (Karachentseva et al. 1985, hereafter Atlas). Due to their very low surface brightness and lack of a significant surface brightness gradient (in contrast to dE galaxies), as well as to  the absence of bright knots (in contrast to dIrr galaxies), the galaxies DDO~44, BK2N, K59, K61, BK5N, DDO~71 = K63, K64, DDO~78, and BK6N were classified as spheroidal dwarfs. Later Karachentsev~(1994) found on the POSS-II films a new object, An 0946+6745, also classified as dSph.  Using CCD data obtained with the Burrell Schmidt telescope at Kitt Peak Caldwell et al. (1998) found several probable  dSph members of the M81 group. One of them, named F8D1, was confirmed as a new spheroidal dwarf companion of M81 via HST observations. Recently Karachentseva \\& Karachentsev (1998, hereafter kk-list) have performed all-sky visual inspection of the POSS-II films for nearby $(V_0 < 500$ km~s$^{-1}$) dwarf  galaxy candidates. As probable dSph members of the M81 group, they included into the kk-list the following objects: DDO~ 44 (=kk61), An0946+6745 (=kk77), K61 (=kk81), DDO~71 (=kk83), K64 (=kk85), DDO~78 (=kk89), and BK6N (=kk91). The galaxies BK2N, and K59 were omitted as background objects because of their high H{\\sc I} velocities, and BK5N bacause of its having a small (0$\\farcm$5) angular diameter. However, Caldwell et al. (1998) imaged BK5N with HST and showed that it is a true dwarf spheroidal galaxy belonging to the M 81 group. The same conclusion has been drawn about DDO~44 based on HST observations (Karachentsev et al 1999).  In this paper we consider the six dSph candidates that have not been imaged so far with HST . We briefly review previous data and present new results based on our HST observations, which allow membership in the M81 group to be established. Digital Sky Survey images of these galaxies are shown in Fig. 1, where the size of each square is 6$\\arcmin$. At the bottom of Fig. 1 we attached DSS images of the two dSphs, F8D1 and BK5N, studied by Caldwell et al. (1998). ",
        "conclusions": ""
    },
    "0010/astro-ph0010236_arXiv.txt": {
        "abstract": "We present first results from a submillimetre continuum survey of the Galactic Centre `Central Molecular Zone' (CMZ), made with SCUBA on the James Clerk Maxwell Telescope. SCUBA's scan-map mode has allowed us to make extremely wide-field maps of thermal dust emission with unprecedented speed and sensitivity. We also discuss some issues related to the elimination of artefacts in scan-map data. Our simultaneous 850/450$\\,\\micron$ maps have a total size of approximately $2.8\\times0.5\\arcdeg$ ($400\\times75\\,\\pc$) elongated along the galactic plane. It covers the Sgr~A region---including Sgr~A*, the circumnuclear disc, and the $+20\\,{\\rm km}/{\\rm s}$ and $+50\\,{\\rm km}/{\\rm s}$ clouds; the area around the Pistol; Sgr~B2---the brightest feature on the map; and at its Galactic Western and Eastern edges the Sgr~C and Sgr~D regions. There are many striking features such as filaments and shell-like structures, as well as point sources such as Sgr~A* itself. The total mass in the Central Molecular Zone is greater than that revealed in previous optically-thin molecular line maps by a factor of $\\sim 3$, and new details are revealed on scales down to $0.33\\,\\pc$ across this $400\\,\\pc$-wide region. ",
        "introduction": "The Galactic Centre provides our best opportunity to study the astrophysical processes in a galactic nucleus with high spatial resolution. These include star-formation and the molecular cloud structures in a region subjected to strong shear forces, magnetic fields, and gravitational potentials. Thermal dust continuum emission traces the temperature-weighted column density of material in a less biased way than molecular line maps, which are affected by excitation, optical depth, and abundance variations. Thus, we can use the continuum emission to map the mass distribution in molecular clouds with good accuracy and sensitivity.  We used SCUBA \\markcite{1999MNRAS.303..659H}({Holland} {et~al.} 1999) on the 15-m James Clerk Maxwell Telescope (JCMT), between April 1998 and April 2000, to map the Galactic Centre region in the submillimetre continuum at $450\\,\\micron$ and $850\\,\\micron$. The new images cover a region 4 times larger and $\\sim 100$ times deeper than earlier surveys \\markcite{1994ApJ...424..189L}({Lis} \\& {Carlstrom} 1994). The map size is approximately $2.8\\times0.5^\\circ$, or $400\\times75\\,\\pc$ at an assumed distance of $8.5\\,{\\rm kpc}$. The map covers the entire `Central Molecular Zone' (CMZ), which extends to radii of $200\\,\\pc$ and contains up to $10\\%$ of the Galaxy's molecular interstellar medium (ISM) \\markcite{1996ARA&A..34..645M}({Morris} \\& {Serabyn} 1996). ",
        "conclusions": "\\label{sec:further-discussion}  The improved depth, resolution, and extent of the SCUBA survey reveal many new features in the Galactic Centre. Not only are there the known large star-forming cloud complexes but also a wide-ranging network of dusty filaments is revealed. For example, at negative Galactic longitudes close to $b=0$ there is filamentary structure which may be continuous and stretches along the Galactic plane for up to $0.8\\arcdeg$ ($120\\,\\pc$). There are partial shells, possibly associated with the known SNRs Sgr~A East (G0.0+0.0), G0.3+0.0, G0.9+0.1, and G1.4-0.1 \\markcite{1998cgsr.book.....G}({Green} 1998). There is also a circular shell of dust clouds centred on $l,b = (0.8,-0.18)$ which does not appear to be documented, and which we have labelled PPR~G0.80-0.18.  This may trace a wind-blown region or perhaps a supernova remnant. In a future paper, we will contrast these structures with star-forming regions near the Solar circle, comparing masses, densities, dust grain properties and star-forming efficiency.  The total mass detected by our survey is $(53\\pm10)\\times10^6 M_\\sun$, in the inner $400\\,\\pc$. \\markcite{1998A&A...331..959D}{Dahmen} {et~al.} (1998) derived a weighted best estimate of the total mass in the inner $600\\,\\pc$ of the Galactic centre using a combination of dust emission, gamma rays, and C$^{18}$O molecular line emission. This estimate was $20$--$50\\times10^6 M_\\sun$. The best optically-thin tracer in this set is the C$^{18}$O survey, from which was derived a mass of $17\\times10^6 M_\\sun$ in the central $600\\,\\pc$. Although our survey covers a smaller area, the total mass detected is significantly greater. Therefore, the SCUBA data comprise the first optically-thin map to trace essentially all the mass in the CMZ at high resolution."
    },
    "0010/astro-ph0010409_arXiv.txt": {
        "abstract": "s{ } ",
        "introduction": "The history of the scientific investigation of galaxy clusters starts with the XVIII century, when Charles Messier and F. Wilhelm Herschel independently produced the first catalogues of nebul\\ae, and noticed remarkable concentrations of nebul\\ae~ on the sky.  Many astronomers of the XIX and early XX century investigated the distribution of nebul\\ae~ in order to understand their relation to the local {\\em ``sidereal system''}, the Milky Way. The question they were trying to answer was whether or not the nebul\\ae~ are external to our own galaxy. The answer came at the beginning of the XX century, mainly through the works of V.M.~Slipher and E.~Hubble (see, e.g., Smith\\cite{smi79a}).  The extragalactic nature of nebul\\ae~ being established, astronomers started to consider clusters of galaxies as physical systems. The issue of how clusters form attracted the attention of K.~Lundmark\\cite{lun27} as early as in 1927. Six years later, F.~Zwicky\\cite{zwi33} first estimated the mass of a galaxy cluster, thus establishing the need for dark matter. The role of clusters as laboratories for studying the evolution of galaxies was also soon realized (notably with the collisional stripping theory of Spitzer \\& Baade\\cite{spi51}).  In the 50's the investigation of galaxy clusters started to cover all aspects, from the distribution and properties of galaxies in clusters, to the existence of sub- and super-clustering, from the origin and evolution of clusters, to their dynamical status, and the nature of dark matter (or {\\em ``positive energy''}, see e.g., Ambartsumian\\cite{amb61}).  As a matter of fact, the topic expanded so much that in 1959 a new separate section specifically devoted to galaxy clusters -- {\\sl Galaxienhaufen} -- appeared in the {\\sl Astronomischer Jahresbericht.}  Galaxy clusters had become one of the main research topics in extragalactic astrophysics.  In this historical review I have tried to cover all aspects of astrophysics research on galaxy clusters, spanning a temporal range of exactly 200 years, from 1784 to 1983.  In 1784, Charles Messier\\cite{mes84} was the first to write about a cluster of galaxies, Virgo, in his {\\sl Catalogue des n\\'ebuleuses et des amas d'\\'etoiles que l'on d\\'ecouvre parmi les \\'etoiles fixes, sur l'horizon de Paris.} In 1983, on October 7$^{th}$, George O. Abell, the eponymous of nearby rich clusters of galaxies, prematurely died at the age of 56. A practical reason for stopping this review with 1983, is that the exponential increase of publications makes it increasingly difficult for the historian to keep pace with the new scientific results.  This review is divided into four main topics: \\begin{enumerate} \\item {\\sc The distribution of clusters}, including: \\begin{itemize} \\item the discovery of clusters \\item cluster catalogues \\item the large scale structure (superclusters) \\item distribution functions of cluster properties \\end{itemize} \\item {\\sc The cluster components}, including: \\begin{itemize} \\item the properties and distribution of cluster galaxies \\item the properties and distribution of intracluster (IC hereafter) hot gas \\item cluster radio-sources \\end{itemize} \\item {\\sc The cluster structure}, including: \\begin{itemize} \\item the dynamical status of clusters (stability and subclustering) \\item cluster masses \\item cluster luminosities (the luminosity function) \\item the nature of the {\\sl missing mass} \\end{itemize} \\item {\\sc The evolution of clusters}, including: \\begin{itemize} \\item the evolution of clustering \\item the evolution of galaxies in clusters \\item the evolution of the IC gas \\item cooling flows and the evolution of cD galaxies \\end{itemize} \\end{enumerate}  I consider here both theoretical and observational aspects.  However, I rarely mention technical aspects, such as the development of new telescopes and instruments, which were certainly very relevant to our understanding of galaxy clusters. In this respect, this review traces the history of the scientific thought, rather than the history of science.  For the sake of homogeneity, all quantities that are $H_0$-dependent, have been re-scaled to the same value the Hubble constant, $H_0 = 75$~km~s$^{-1}$~Mpc$^{-1}$. ",
        "conclusions": "In the course of time, the concept of cluster of galaxy has significantly evolved. A concentration of nebul\\ae, maybe galactic objects, like a star cluster, in the early days of the XX century. A remarkable (but relatively rare) concentration of external galaxies, which nevertheless were much smaller than our own, in Hubble's times. Or rather the extreme of a continuous clustered distribution of galaxies, according to Carpenter.  A stable, bound dynamical system, with an incredible mass, according to Zwicky. Or instead, a light, rapidly disrupting system, whose explosion was powered by unknown mechanisms operating in the centres of its galaxies, according to Ambartsumian. A galaxy incubator, according to Zel'dovich' top-down scenario, or rather an association of free galaxies, according to Peebles' bottom-up scenario. A dangerous place to live, for spiral galaxies, according to {\\sl nurture} scenarios for galaxy evolution. Or maybe a very quiet place, where old ellipticals can passively evolve for billions of years, according to {\\sl nature} scenarios for galaxy formation. A knot in the filamentary structure of the Universe, when the Large Scale Structure was finally unveiled by observations in the 80's.  A cluster of gas, rather than a cluster of galaxies, in the 90's, when X-ray surveys became more effective in finding high redshift clusters than the traditional optical methods. And now, finally, a cluster of dark matter, a dark cluster, which will be found through the weak lensing surveys (see {\\sc Mellier}, these proceedings).  If the evolution of clusters is slow (see, e.g., {\\sc Dickinson}, these proceedings), not so slow is the evolution of science. Moreover, this evolution is often discontinuous and non-monotonic. Zwicky's missing mass was re-discovered in galaxy halos after 40 years; the existence of significant subclustering in clusters was demonstrated in the 60's by van den Bergh, but the irregular X-ray morphologies of clusters came as a surprise to many astronomers. The Local Supercluster was hinted at by J.~Herschel in the XIX century, and rediscovered several times before \\dV~ re-affirmed its existence, in the 50's.  And many other examples can be found by reading this review.  We certainly need to keep track of the evolution of science, or we risk to forget about fundamental results that might take years to be re-discovered. I hope this modest review can be helpful in this respect.  \\begin{quote} {\\em ``Quello che lei non sa \\`e il vero scopo del nostro lavoro} [\\ldots] {\\em \\`E perch\\`e tutto non sia stato inutile, per trasmettere tutto quello che sappiamo ad altri che non sappiamo chi sono n\\`e cosa sanno.''}  \\hspace{5cm} Italo Calvino, {\\sl La memoria del mondo} \\end{quote}"
    },
    "0010/astro-ph0010123_arXiv.txt": {
        "abstract": "In this paper I put forward a model in which GRB980425 is both associated with SN1998bw and is also a standard canonical (long; $\\sim$ seconds) gamma-ray burst.  Herein it is argued that if gamma-ray bursts are relativistic jets with the fastest moving material at the core, then the range of observed jet inclinations to the line-of-sight produces a range in the observed properties of GRBs, i.e.~the lag-luminosity relationship.  In particular, if the jet inclination is high enough, the observed emitter will move slowly enough to render relativistic beaming ineffective, thus distinguishing the jet from apparent isotropic emission.  Thus we expect a break in the lag-luminosity relationship.  I propose that GRB980425 defines that break.  The position of this break gives important physical parameters such as the Lorentz factor ($\\gamma_{max} \\sim 1000$), the jet opening angle ($\\sim 1$ degree), and thus the beaming fraction ($\\sim 10^{-4}$).  Estimates of burst rates are consistent with observation.  If correct, this model is evidence in favor of the collapsar mode as the progenitor of cosmological, long gamma-ray bursts. ",
        "introduction": "GRB980425 and its apparent association with SN 1998bw \\citep{gvpk+98} has drawn much attention among gamma-ray burst researchers.  While the connection of the gamma-ray burst (GRB) to the supernova (SN) remains uncertain, it is striking that GRB980425 and SN 1998bw were each, taken individually, unusual events.  To start, SN 1998bw was a relatively rare and unusually luminous Type I b/c supernova \\citep{gvpk+98}.  It was the brightest radio supernova ever observed \\citep{wl99} which may have been due to relativistic outflow with Lorentz factor $\\gamma \\sim 2$ \\citep{kfwe+98}.  Furthermore, GRB980425 was an unusual GRB.  It was comprised of a single, unusually rounded peak \\citep{bkhp+98}.  A cool burst, it was not seen in BATSE's highest energy channel ($> 300$ keV) \\citep{nmb00}.  If this burst is indeed associated with SN1998bw ($z=0.008$), then the burst is apparently vastly weaker than all other known bursts, with an inferred isotropic gamma-ray energy of $8 \\times 10^{47}$ ergs \\citep{gvpk+98}. Finally, of particular interest to this paper, \\citet{nmb00} found that the lag of the peak of this burst between BATSE channels 1 and 3 was exceptionally large: $\\Delta t_{980425} \\approx 4.5$ seconds.  In light of the respective idiosyncrasies of these two events, we may either conclude that we have seen a chance coincidence of two unusual events, with a probability of $10^{-4}$ or less \\citep{gvpk+98}, or perhaps the discovery of an entirely new type of GRB \\citep{gvpk+98,bkhp+98}.  In this paper I propose a third alternative; that GRB980425/SN1998bw is a canonical gamma-ray burst, deriving from a relativistic jet driven by a collapsar \\citep{mfw99}, observed at high angle of inclination.  The idea that this burst was a jet viewed off-axis has been proposed by several other authors \\citep{ww98,wes99,naka99,hww99}.  However in this paper I show how GRB980425 may be a canonical gamma-ray burst by its relation to other bursts on the lag-luminosity relationship discovered by \\citet{nmb00}. Thus this burst need not be a distinct class of GRB or a special case of a failed collapsar, polluted with excessive baryon entrainment within the jet \\citep{wmf99}.  Physical GRB parameters can be gleaned from this identification.  In brief, if we assume that the core of the jet has the highest velocity material and that the velocity monotonically decreases with increasing angle from the core axis, then there will be an angle at which the $1/\\gamma$ aperture imposed by relativistic beaming becomes comparable to the angular size $\\theta_0$ of the emitting region.  At this angle there will be a change in the observed properties of the GRB.  In particular, one would expect a break in the lag-luminosity relationship discovered by \\citet{nmb00} and futher interpreted by \\citet{jay00}.  Herein I show that the peak number luminosity's inverse dependence on spectral lag, $N_{pk} \\propto \\Delta t^{-1}$, will steepen to $N_{pk} \\propto \\Delta t^{-3}$.  By fitting the $\\Delta t^{-3}$ curve to intersect with GRB980425 one obtains a complete lag-luminosity curve for GRBs which is consistent with observed data (Figure \\ref{lag-lum-980425}).  Knowledge of the shape of this curve allows determinations of some key quantities of interest for GRBs; particularly jet opening angle, maximum lorentz factor, and total energy.  \\begin{figure}[tb] \\centering \\epsfig{figure=nLum425.eps, width = 11 cm} % \\caption{Peak photon number luminosity $N_{pk}$ versus spectral pulse lag for six bursts with known redshifts plus GRB980425.  A break is inferred by fitting a break slope $\\propto \\Delta t^{-3}$ to intersect GRB980425.  Spectral cross-correlation function lags between BATSE channels 3 and 1 (CCF31) for regions down to 0.5 and 0.1 of peak intensity were obtained from \\citet{nmb00}.  The line of best fit for 0.1 (squares) is $\\propto \\Delta t^{-0.98}$.  } \\label{lag-lum-980425} \\end{figure} ",
        "conclusions": "\\label{sectionbeaming}  The obvious prediction of the model present herein is that all (long) GRBs will fall along the curve in Figure \\ref{lag-lum-980425}.  In addition, since this model supports the collapsar model for GRB progenitors, all (long) GRBs should have a SN buried within as predicted by \\citet{wmf99}.  Many questions remain.  For instance GRB980425 had a quickly decaying X-ray afterglow \\citep{bkhp+98}.  Perhaps numerical simulations of afterglows from relativistic jets \\citep[e.g.][]{gmps00} could yield insights into this behavior.  If this is a common characteristic of off-axis jets, this might have an effect on the predicted rates of observable so called ``orphan afterglows'' \\citep{rhoads97}.  This work was performed under the auspices of the U.S. Department of Energy by University of California Lawrence Livermore National Laboratory under contract W-7405-ENG-48."
    },
    "0010/astro-ph0010315_arXiv.txt": {
        "abstract": "One of the main scientific drivers of the Space InfraRed Telescope Facility (SIRTF) is the search for brown dwarfs and extrasolar superplanets. We discuss observational strategies for identification of these objects, and conclude that an optimal strategy is a wide IRAC survey (18~deg$^2$) with a 5~$\\sigma$ sensitivity of 3.9~$\\mu$Jy in channel 2 (M$\\sim$19.1$^m$). For this sensitivity, we provide estimates of the number of low mass brown dwarfs and isolated planets detected per square degree for power-law mass functions with $\\alpha$=1.5, and 1.0. Shallower surveys covering a larger area are inefficient because of large overheads and detector noise. Deeper surveys covering a smaller area become more and more affected by crowding with galaxies. A survey like the one that we propose would determine the field mass function down to a few Jupiter masses through the identification of a large sample of brown dwarfs and isolated planets. The proposed SIRTF survey would also allow the first detection of ultracool substellar objects with temperatures between 700~K and 200~K. The cooling curves of substellar objects with masses less than 20 Jupiters imply that they should spend most of their lifetimes at temperatures below 700~K. Preliminary models indicate that their atmospheres could be dominated by water clouds, which would diminish their optical and near-infrared fluxes. The properties of those objects are still completely unexplored. ",
        "introduction": "The past five years have witnessed the remarkable birth of a new research field of astronomy --- the physical study of substellar objects outside the solar system. Large scale optical and near-infrared surveys of the sky (DENIS, 2MASS and SDSS) have revealed very cool objects that are close to us but nevertheless were unknown. The overwhelming majority of substellar objects in the field, however, cannot be detected by ground-based surveys; such objects cool very rapidly with time and become extremely faint at wavelengths shorter than about 4~$\\mu$m.  SIRTF (Fanson et al. 1998) will allow to probe deeper in the substellar regime, extending the census of substellar objects in the field to lower masses and temperatures (see Figure~1).  The separation between brown dwarfs and planets is a matter of convention. We concur with other authors (e.g., Basri \\& Marcy 1997; Oppenheimer et al. 2000) in arguing that a definition based on the well known laws of gravitation and nuclear physics is more reliable than a definition based on the origin of the objects. Thus, we adopt the following definitions: A brown dwarf is an object unable to stabilize on the hydrogen burning main-sequence, but able to fuse deuterium into helium. For solar metallicity the mass range of brown dwarfs is between 0.075~M$_\\odot$ and 0.013~M$_\\odot$ (Baraffe et al. 1998, Saumon et al. 1996). Planets are objects unable to fuse deuterium, i.e., they have masses lower than 0.013~M$_\\odot$. ``Isolated  planets'' are planets that are not bound to any particular star. Recent results suggest that this kind of planets may be numerous in the disk of the Galaxy (B\\'ejar et al. 2000; Zapatero Osorio et al. 2000). The smooth transition of the mass function from the brown dwarfs to the isolated planets indicates that they have a common origin, which is much like that of star formation from fragmentation and collapse of molecular clouds, but not like the planets in our solar system (accretion in a protoplanetary disk). Despite of their different origins, isolated planets may have the same basic properties (overall chemical composition, luminosity, temperature) as their counterparts orbiting stars.  Studies show that the number of substellar objects increases towards lower masses. The functional form of this relationship, the Initial Mass Function (IMF hereafter), can be approximated by a power law dN/dM $\\sim$ M$^{-\\alpha}$ (Salpeter 1955). For masses between 0.1~M$_\\odot$ and 0.01~M$_\\odot$ (i.e., $\\sim$ 100 and 10~M$_{Jup}$, respectively), several surveys in young open clusters have found $\\alpha=1.0\\pm0.5$. (B\\'ejar et al. 2000, Hillenbrand \\& Carr 2000; Mart\\'\\i n et al. 2000; Najita et al. 2000). The density of field brown dwarfs discovered by DENIS and 2MASS is also consistent with this mass function (Reid et al. 1999). The number of substellar objects in each cluster, or in the field, is still small. Hence the statistical uncertainties are large and it is not possible to conclude if there are significant variations in the IMF from one region to another. The cooling curves of brown dwarfs and planets imply that they spend a large part of their lives at extremely low luminosities and cool temperatures. The launch of the SIRTF satellite provides an unprecedented opportunity for detecting ultracool substellar objects.  This paper is organized as follows: Section 2 gives a brief description of the expected spectra of ultracool substellar objects and justifies the need for SIRTF to detect them; Section 3 discusses observational strategies to detect substellar objects with SIRTF; Section 4 provides estimates of the counts of substellar objects in the disk that can be detected with SIRTF; and Section 5 summarizes the main conclusions. ",
        "conclusions": "We have shown that IRAC channel 2 is very efficient for detecting ultracool substellar objects. We present an optimized observational strategy for detection of brown dwarfs and isolated planets with temperatures below 700~K. Preliminary models indicate that the atmospheres of these objects are dominated by water clouds, which make the objects fainter at optical and near-infrared wavelengths than previously thought. The 5~$\\mu$m window is an excellent region for detecting them because their atmospheres are more transparent. Our strategy consists of a SIRTF survey with a 5~$\\sigma$ sensitivity of 3.88~$\\mu$Jy in IRAC channel 2. We show that substellar objects can easily be identified in an IRAC color-color diagram. Any object with log F(4.5$\\mu$m)/F(3.6$\\mu$m) $>$ 0.3 can be considered as a good brown dwarf or planet candidate. Neither stars nor galaxies contaminate the substellar locus. With our detection limit, we predict a density of 30 isolated planets per square degree for an IMF with a slope $\\alpha$ = 1.5. In 600 hours of SIRTF time, an area of 18 square degrees would be covered. We predict that such a survey would find about 540 isolated planets, and about 160 substellar objects with temperatures cooler than 600~K. Such objects cannot be detected with the current generation of near-infrared all-sky surveys (DENIS and 2MASS).  For shallower IMF slopes, the number of substellar objects diminishes. Thus, the proposed survey can determine the field substellar IMF down to very low masses."
    },
    "0010/astro-ph0010586_arXiv.txt": {
        "abstract": "We present improved visual and near-infrared empirical mass-luminosity relations for very low mass stars (M$<$0.6~\\Msol). These relations make use of all stellar masses in this range known with better than 10\\% accuracy, most of which are new determinations with 0.2 to 5\\% accuracy from our own programme, presented in a companion paper.  As predicted by stellar structure models, the metallicity dispersion of the field populations induces a large scatter around the mean V band relation, while the infrared relations are much tighter. The agreement of the observed infrared mass-luminosity relations with the theoretical relations of Baraffe et al. (\\cite{baraffe98}) and Siess et al. (\\cite{siess00}) is impressive, while we find an increasingly significant discrepancy in the V band for decreasing masses. The theoretical mass-luminosity relation which is insufficiently steep, and has introduced significant errors in the local stellar mass functions derived from V band luminosity functions. ",
        "introduction": "The mass of a star is arguably its most basic characteristic, since most stellar properties have a very steep mass dependency. Yet, it  can only be directly determined for some stars in multiple systems, and for most stars it has to be inferred from more directly observable parameters, the luminosity, the chemical composition, and the evolutionary status. An accurately calibrated Mass-Luminosity (hereafter M/L) relation, or in full generality a Mass-Composition-Age/Luminosity relation, is therefore an essential astrophysical tool (e.g. Andersen \\cite{andersen91}; \\cite{andersen98}). It is needed at many places to convert observable stellar light to the underlying mass, and perhaps most importantly, to derive stellar mass functions from more readily obtained luminosity functions. The latter in turn provide an essential diagnostic of star formation theories. They also represent a basic building block for galactic and stellar cluster dynamical models.   The M/L relation is fairly well constrained for solar-type and intermediate mass stars: a sizeable number of such stars in eclipsing systems have had their mass determined with better than 1\\% relative accuracy (Andersen \\cite{andersen91}), and the theory of these stars approximately matches this excellent precision, when both evolutionary effects and metallicity are taken into account (Andersen \\cite{andersen98}). For both smaller and more massive stars however, the M/L relation is significantly more uncertain, as theory and observations meet there with new difficulties.   Here we address the low-mass end of the HR diagram, below 0.6{\\Msol}, where stellar models face two major hurdles (Chabrier \\& Baraffe \\cite{chabrier00}, for a recent review): \\begin{itemize} \\item{} the onset of low temperature electron degeneracy in the stellar core (Chabrier \\& Baraffe \\cite{chabrier97}; \\cite{chabrier00}); \\item{} a complex cold and high gravity stellar atmosphere, dominated by molecular and dust opacity (Allard et al. \\cite{allard97}; Jones \\& Tsuji \\cite{jones97}; Allard \\cite{allard98}). \\end{itemize} Much progress has been made in the past few years, with the realization that an accurate atmosphere description {\\it must} be used as an outer boundary condition to the stellar interior equations (Baraffe et al. \\cite{baraffe95}), and with an increased sophistication of the atmospheric models (Allard et al., in prep). State of the art models (Baraffe et al. \\cite{baraffe98}; Chabrier et al. \\cite{chabrier00b}) now produce good to fair agreement with most observational colour-magnitude diagrams (Goldman et al. \\cite{goldman99}, for an example). This lends considerable credence to their general reliability. Their description of some of the input physics however remains incomplete or approximate: some molecular opacity sources are still described by relatively crude approximations, or by line lists that remain incomplete (though vastly improved), the validity of the mixing-length approximation in the convective atmosphere is questionable, and atmospheric dust condensation and settling introduces new uncertainties at the lowest effective temperatures. The actual severity of these known shortcomings of the models is unclear, making an independent check of their M/L prediction most desirable.   \\begin{table*} \\tabcolsep 2.23mm \\begin{tabular}{|l|lll|l|rrrrrrrrr|} \\hline Name & $\\pi$ & ${\\sigma}({\\pi})$ & Ref & Spectral & $B$ & $V$ & $R_c$ & $I_c$ & $J$ & $H$ & $K$ & $L$ & $L'$\\\\ & (mas) & (mas) & & type & &     &       &       &     &     &     &   & \\\\ \\hline \\hline Gl~65AB % &  373.7 & 2.7 & Yale95 & M5.5V & 13.87 & 12.00 & 10.37 & 8.31 & 6.24 & 5.67 & 5.33 & 5.00 & \\\\ Gl~234AB % & 243.7{\\ } & 2.0{\\ } & S\\'eg00 & M4.5V & 12.80 & 11.08 & 9.77 & 8.06 & 6.40 & 5.78 & 5.49 & 5.33 &  \\\\ YY~Gem   % &  {\\ \\,}74.7  & 2.5  & Yale95    & M0V & 10.50 & 9.07 & 8.10 & 7.09  % & 6.01 & 5.35 & 5.18 & 5.16 & \\\\ % GJ~2069A % & {\\ \\,}78.05 & 5.69 & HIP & M3.5V &       & 11.89 & 10.68 & 9.09 & & & & & \\\\ Gl~473AB  % & 227.9{\\ }  & 4.6{\\ } & Yale95 & M5V & 14.30 & 12.46 & 10.90 & 8.92 &  6.96 & 6.39 & 6.06 &  & 5.63 \\\\ Gl~570B   % & 169.8{\\ }  & 0.9{\\ }  & For99    & M1V & 9.57 & 8.09 & 7.09 & 5.97 &  4.75 & 4.14 & 3.93 & 3.77 & 3.67 \\\\ Gl~623AB    % & 124.34 & 1.16 & HIP & M2.5V & 11.76 & 10.26 & 9.27 & 7.96 &  6.67 & 6.14 & 5.91 & & \\\\ CM~Dra    % & {\\ \\,}69.2{\\ }   & 2.5{\\ } & Yale95 & M4.5V &  14.49 & 12.91 & 11.67 & 9.99 &  8.54 & 8.07 & 7.79 &  & \\\\ Gl~644AB  % & 154.8{\\ }  & 0.6{\\ }  & S\\'eg00 & M3V & 10.60 & 9.02 & 7.92 & 6.55 &  5.28 & 4.64 & 4.39 & 4.14 & \\\\ Gl~661AB         % & 156.66 & 1.37 & Sod99 & M3.5V & 10.89 & 9.40 & 8.30 & 6.89 & 5.56 & 5.04 & 4.82 & 4.60 & \\\\ Gl~702AB         % & 195.7{\\ \\,}  & 0.9{\\ \\,} & Sod99 & K0V % & 4.88 & 4.02 & 3.51 & 3.05 & 2.42 & 1.96 & 1.89 &  & \\\\ Gl~747AB         % & 120.2{\\ } & 0.2{\\ } & S\\'eg00 & M3V & 12.93 & 11.25 & 10.11 & 8.65 &  7.25 & 6.69 & 6.43 & & \\\\ Gl~791.2AB  % &  112.9  &  0.3 & Ben00 & M4.5V & 14.72 & 13.06 & 11.72 & 9.96 &  8.20 & 7.63 & 7.33 &  &  \\\\ Gl~831AB % & 117.5{\\ } & 2.0{\\ } & S\\'eg00 & M4.5V & 13.68 & 12.01 & 10.71 & 9.02 &  7.29 & 6.70 & 6.42 &  & \\\\ Gl~860AB % &  247.5{\\ } & 1.5{\\ } & HIP & M3V+M4V & 11.25 & 9.59 & 8.40 & 6.91 & 5.56 & 4.97 & 4.71 & 4.48 & \\\\ Gl~866ABC % & 293.6{\\ } & 0.9{\\ } & S\\'eg00 & M5.5V & 14.29 & 12.33 & 10.66 & 8.62 &  6.50 & 5.91 & 5.57 & 5.22 & 5.01 \\\\ \\hline \\end{tabular} \\caption{ Basic parameters for the M-dwarf systems with accurate masses. The parallaxes mostly originate from the Hipparcos catalog (ESA \\cite{esa97}), the Yale General Catalog of trigonometric Parallaxes (Van Altena et al. \\cite{vanaltena95}), and from S\\'egransan et al. (\\cite{segransan00a}). In that paper we derived optimal combinations of orbital and astrometric parallaxes, which often have a strong contribution from either an Hipparcos or a Yale catalog astrometric parallax. Some individual entries are additionally taken from Soderhjelm (\\cite{soderhjelm99}), Forveille et al. (\\cite{forveille99}), and Benedict et al. (\\cite{benedict00}). All spectral types are from either Reid et al. (\\cite{reid95}) or Hawley et al. (\\cite{hawley97}) and refer to the integrated light of the system. The photometry is taken from the extensive homogenized compilation of Leggett (\\cite{leggett92}), except for YY~Gem, Gl~702AB and GJ2069A. The photometry of GJ2069A is from Weiss (\\cite{weiss91}). The optical photometry of YY~Gem is from Kron et al. (\\cite{kron57}) (RI), Eggen (\\cite{eggen68}) (UBV), Barnes et al. (\\cite{barnes78}) (UBVRI), and the infrared photometry from Johnson (\\cite{johnson65}), Glass (\\cite{glass75}), and Veeder (\\cite{veeder74}). The optical photometry for Gl~702AB is from Bessel (\\cite{bessel90}) and the infrared photometry from Alonso et al. (\\cite{alonso94}). All photometry was converted to the Johnson-Cousins-CIT sytem adopted by Leggett (\\cite{leggett92}) using the colour transformations listed in that paper. Multiple measurements for the same band were averaged with equal weights. } \\label{table_basic} \\end{table*}  Detached eclipsing M-dwarf binaries are rare, with only three known to date. Most  mass determinations for Very Low Mass Stars (VLMS) are therefore instead obtained from visual and interferometric pairs, which until recently have not yielded comparable precisions. The current state of the art empirical M/L relation for M dwarfs (Henry \\& McCarthy \\cite{henry93}; Henry et al. \\cite{henry99}) as a result mostly rests on masses determined with 5-20\\% accuracy. The last two years have seen a dramatic evolution in this respect, with two groups breaking through the former $\\sim$5\\% accuracy barrier. The first team used the 1~mas per measurement astrometric  accuracy of the Fine Guidance Sensors (Benedict et al. \\cite{benedict99}) on {\\it HST} to determine system masses for three angulary resolved binaries with 2 to 10\\% accuracy (Franz et al. \\cite{franz98}; Torres et al. \\cite{torres99}; Benedict et al. \\cite{benedict00}). Shortly thereafter our team demonstrated even better accuracies for masses of VLMS, of only 1-3\\% (Forveille et al. \\cite{forveille99}; Delfosse et al. \\cite{delfosse99b}), by combining very accurate radial velocities with precise angular separations from adaptive optics. In a companion paper (S\\'egransan et al. \\cite{segransan00a}) we determine a dozen new or improved masses, with the same method and with accuracies that now range between 0.5 and 5\\%. The same paper also presents improved masses for two of the three known eclipsing M-dwarf systems, with 0.2\\% accuracy. Here we take advantage of this wealth of accurate new data, and reassess the VLMS M/L relation on a much firmer ground.  In Sect.~2 we briefly describe the sample of accurately determined very low stellar masses. We then discuss the resulting empirical M/L relation in Sect.~3, and compare it with theoretical models in Sect.~4.    \\begin{table*} \\tabcolsep 0.4mm \\begin{tabular}{|l|rl|rl|rl|rl|rl|rl|} \\hline Name & ${\\Delta}V${\\ \\ } & Ref & ${\\Delta}R${\\ \\ } & Ref & ${\\Delta}I${\\ \\ } & Ref & ${\\Delta}J${\\ \\ } & Ref & ${\\Delta}H${\\ \\ } & Ref & ${\\Delta}K${\\ \\ } & Ref \\\\ \\hline \\hline Gl~65AB    & 0.45 $\\pm$ 0.08 & Hen99 % &                 &     &            & & 0.38 $\\pm$ 0.03 & Hen93 & 0.30 $\\pm$ 0.02 & Hen93 & 0.40 $\\pm$ 0.07 & Hen93 \\\\ Gl~234AB   & 3.08 $\\pm$ 0.05 & Hen99 % &                 &     &            & & 1.79 $\\pm$ 0.30 & C94 & 1.63 $\\pm$ 0.11 & C94 & 1.62 $\\pm$ 0.02 & D00 % \\\\ YY~Gem     & 0.35 $\\pm$ 0.20 & L78 % & 0.34 $\\pm$ 0.19 & L78 & 0.32 $\\pm$ 0.27 & L78 &                 & &                 & &                 & \\\\ GJ~2069A   & 0.79 $\\pm$ 0.10 & D99a &                 &     &            & &                 &      &         & &             & \\\\ Gl~473AB   & $-$0.01  $\\pm$  0.05 & Hen99 &                 &     &            & & 0.13 $\\pm$ 0.04            &   Tor99 & 0.20 $\\pm$ 0.07            &   Tor99 & 0.44 $\\pm$ 0.09            &   Tor99 \\\\ Gl~570BC    & 1.66  $\\pm$ 0.16 & S % &                 &     &            & & 1.19  $\\pm$ 0.05 & For99 & 1.15  $\\pm$ 0.05 & For99 & 1.18 $\\pm$ 0.03  & For99 \\\\ Gl~623AB     & 5.28 $\\pm$ 0.10 & B96 &                 &     &            & & 3.28 $\\pm$ 0.3{\\ \\,}  & Hen93 & 2.65 $\\pm$ 0.03 & Hen93 & 2.87 $\\pm$ 0.14 &   Hen93 \\\\ CM~Dra & 0.14 $\\pm$ 0.03 & S % &                 &   % & 0.15 $\\pm$ 0.01 & L77 % &                 & &                 & &             & \\\\ Gl~644A-Bab &  $-$0.08 $\\pm$ 0.05 & S &                 &     &            & &                 & & $-$0.48 $\\pm$  0.06  & D00 % & $-$0.46 $\\pm$  0.03 & D00  % \\\\ Gl~644Ba-Bb  &  0.49 $\\pm$ 0.10 & S &                 &     &            & &             & &             & &                 & \\\\ Gl~661AB & 0.05 $\\pm$ 0.05 & TYC &                 &     &            & & 0.41 $\\pm$ 0.01   & Hen93 & 0.46 $\\pm$ 0.02   & Hen93 & 0.42 $\\pm$ 0.07   & Hen93   \\\\ Gl~702AB &  1.86 $\\pm$ 0.02 &  TYC &                 &     &            & & 1.51 $\\pm$ 0.04 & Hen93 &             & & 0.74 $\\pm$ 0.03 & Hen93 \\\\ Gl~747AB   & 0.22 $\\pm$ 0.05 & S &                 &     &            & &                 &     &            & & 0.10 $\\pm$ 0.03 & D00 % \\\\ Gl~791.2AB & 3.27 $\\pm$ 0.10 & Ben00 &                 &     &            & &                 &     &            & &                 & \\\\ Gl~831AB & 2.10 $\\pm$ 0.06 & Hen99 &                 &     &            & &                 & & 1.26 $\\pm$ 0.03  & D00 % & 1.28 $\\pm$ 0.02  & D00 % \\\\ Gl~860AB & 1.70 $\\pm$ 0.09 & HIP % &                 & &                 & & 1.19 $\\pm$ 0.10 & Hen93 & 1.14 $\\pm$ 0.05 & Hen93 & 1.37 $\\pm$ 0.08 & Hen93 \\\\ Gl~866AC-B & 0.40 $\\pm$ 0.10 & Hen99  % &                 &     &            & &            & & 0.52 $\\pm$ 0.03 & D00 % & 0.54 $\\pm$ 0.03 & D00 % \\\\ Gl~866AC  & 2.04 $\\pm$ 0.40 & S &                 &     &            & &                 &     &            & &                 & \\\\ \\hline \\end{tabular} \\caption{Magnitude differences for M-dwarf systems with accurate masses. Reference codes are: TYC for the Tycho catalogue HIP for the Hipparcos catalogue L77 for Lacy (\\cite{lacy77}), L78 for Leung \\& Schneider (\\cite{leung78}), Hen93 for Henry \\& McCarthy (\\cite{henry93}), C94 for Coppenbarger et al. (\\cite{coppenbarger94}), B96 for Barbieri et al. (\\cite{barbieri96}), Ben00 for Benedict et al. (\\cite{benedict00}), Tor99 for Torres et al. (\\cite{torres99}) Hen99 for Henry et al. (\\cite{henry99}), D99a for Delfosse et al. (\\cite{delfosse99a}), For99 for Forveille et al. (\\cite{forveille99}), and D00 for the present paper. % S stands for spectroscopic magnitude differences, infered from the relative line depths for double-lined spectroscopic binaries.} \\label{table_dmag} \\end{table*} ",
        "conclusions": "Empirical masses of 0.2 to 10\\% accuracy validate the near-IR Mass/Luminosity relations predicted  by the recent stellar models of Baraffe et al. (\\cite{baraffe98}) and Siess et al. (\\cite{siess00}), down to $\\sim$0.1~\\Msol. They also point out some low level ($\\sim$0.5~mag) deficiencies of these models in the V band. Perhaps more importantly however, the V band M/L diagram represents direct evidence for an intrinsic dispersion around the mean M/L relation. This had previously remained hidden in the measurement noise, but there is, as theoreticians have kept telling us, no such thing as one single M/L relation for all M dwarfs. This is particularly true for the visible bands, while the dispersion in the near-IR JHK bands is much lower. Comparisons between measured masses and theoretical models will therefore increasingly depend on metallicity measurements for individual systems, which are not easily obtained.     The $\\sim$0.5~magnitude discrepancy between observational and model masses derived from visible photometry has some consequences for mass functions determination. As mass cannot be determined for volume-limited samples, the mass function is always obtained from a luminosity function, by writing that \\begin{displaymath} {\\frac{{\\rm d}N}{{\\rm d}M}}~={\\frac{{\\rm d}N}{{\\rm d}L}}{\\times}{\\frac{{\\rm d}L}{{\\rm d}M}} \\end{displaymath} and the slope of the M/L relation therefore plays a central role in its derivation. Below 0.5{\\Msol} the d$L/$d$M$ slope of the empirical M/L relation is steeper than that of the BCAH models and shallower than for the SDF ones, by 10 to 20\\%.  Their use will therefore respectively underestimate and overestimate the number of lower mass stars by this amount. Probably more seriously, the large dispersion around the V band M/L relation will introduce large Malmquist-like biases in the derived mass function, which would need an excellent characterization of this dispersion to be corrected. The infrared relations have both better agreement with the observations and much lower dispersion. We strongly recommend that they be used rather than the V band relations, whenever possible."
    },
    "0010/astro-ph0010065_arXiv.txt": {
        "abstract": "We have undertaken a study of the globular cluster (GC) systems of a large sample of elliptical and spiral galaxies in order to test specific predictions of galaxy formation models.  Here we present results for the first elliptical in the study, the giant Virgo cluster galaxy NGC~4472 (M49).  The galaxy was observed in three filters (\\bvr) using the wide-field Mosaic Imager on the Kitt Peak 4-meter telescope.  The Mosaic images roughly double the radial coverage of previous CCD observations of NGC~4472.  We have combined the Mosaic data with published spectroscopic data and archival HST observations in order to study NGC~4472's GC system in detail, and to fully characterize the amount of contamination in our sample of GC candidates.  We find that the radial profile of the GC system is fit fairly well by a de~Vaucouleurs law of the form log~$\\sigma_{GC}$ = (3.38$\\pm$0.08) $-$ (1.56$\\pm$0.05) $r^{1/4}$ out to 17\\arcmin\\ ($\\sim$80 kpc), but that the observed profile falls slightly below the de~Vaucouleurs law between 17\\arcmin\\ and 23\\arcmin, the limit of the data.  The bimodal color distribution observed in previous studies is apparent in our data.  We find a small metallicity gradient in the inner 8\\arcmin\\ of the total GC system due to the increasing ratio of blue to red clusters, consistent with results from past studies.  The gradient vanishes, however, when the entire radial extent of the data is taken into account.  We estimate a total of $\\sim$5,900 GCs in NGC~4472 out to 23\\arcmin, yielding a specific frequency of 3.6$\\pm$0.6 within this radius. This specific frequency value is smaller than that found by previous studies of NGC~4472.  We examine the implications of these results with regard to predictions made by four different galaxy formation models, and find that all four models have at least one inconsistency with the data. ",
        "introduction": "Globular clusters (GCs) are among the oldest readily-observable objects in the Universe, and have been detected in all types of massive galaxies and many dwarfs.  The fact that GCs have such large ages and are seen in virtually all types of galaxies indicates that they are closely tied to galaxy formation.  Studying the ensemble properties of the entire system of GCs around a given galaxy can provide important clues regarding the formation and subsequent evolution of that galaxy, and such studies can be particularly useful for testing theories of elliptical galaxy formation.  GC systems of ellipticals are typically more populous than those of spirals (cf.\\ \\cite{az98}).  Moreover, GC systems are generally at least as extended as the light distributions of their host galaxies (cf.\\ \\cite{harris86}, \\cite{az98}), so GCs can be used to probe the formation and enrichment history of the outer regions of elliptical galaxies.  Recent studies have found that the GC systems of many ellipticals have bimodal rather than single-peaked color distributions (e.g., \\cite{geisleretal96}, \\cite{kunwhit99}, \\cite{gebhardt99}).  The presence of two peaks has been interpreted as due to two GC populations with different metallicities, possibly formed in different episodes.  Evidence also suggests that at least in some ellipticals, the bluer (metal-poor) GC population is more extended than the red (metal-rich) population (e.g., \\cite{geisleretal96}, \\cite{neilsen00}).  These observations agree in general with the predictions of a galaxy formation scenario proposed by Ashman \\& Zepf (1992), in which an elliptical galaxy is formed via the merger of two or more disk galaxies.  In the Ashman \\& Zepf picture, a younger, spatially-concentrated, metal-rich system of GCs is formed during the merger itself, and the spatially-extended, metal-poor GC population originates in the halos of the progenitor spirals.  Other models may also explain the observed differences between the GC populations in ellipticals.  Forbes \\et (1997) suggest that the GCs in ellipticals are formed in distinct episodes, with the metal-poor population formed early, from less enriched gas, and the metal-rich population forming later, along with the galaxy field stars.  In the case of giant ellipticals and cD galaxies, Forbes \\et suggest that tidal stripping of GCs from nearby galaxies may contribute to building up the outer GC system.  Another model, proposed by C\\^ot\\'e \\et (1998), suggests that giant ellipticals form via dissipational collapse, producing a metal-rich GC population.  Metal-poor GCs are then captured from other galaxies by accretion or tidal stripping.  These scenarios make different predictions for the detailed properties of the GC systems of ellipticals.  For example, the merger scenario of Ashman \\& Zepf predicts that the mass-normalized number of metal-poor GCs around ellipticals and spirals should be the same, since in this model all of the metal-poor GCs come from the progenitor spirals. The C\\^ot\\'e \\et model predicts that a significant number of metal-poor, tidally-stripped GCs should reside at large distances from giant cluster ellipticals.  We have undertaken a study of a large sample of elliptical and spiral galaxies in order to test the predictions of the various formation models.  Using a new generation of wide-field CCD imagers, we have observed a sample of early-type galaxies in the Virgo cluster and the field, as well as a large sample of fairly isolated, edge-on spirals.  Our aim is to accurately quantify the total numbers, spatial distributions and color distributions of the GC systems in these galaxies out to large radii. By observing the galaxies in three filters and employing image classification analysis we can greatly reduce the amount of potential contamination (from background and foreground objects) in our GC samples, while simultaneously detecting GCs further out than previous CCD studies have done.  Furthermore, we make use of available space-based and spectroscopic data to help quantify the amount of contamination that exists in our GC lists.  Here we present the results of our study of the GC system of the galaxy NGC~4472 (M49). NGC~4472 is located four degrees from the center of the Virgo cluster and is the brightest of the Virgo galaxies.  We chose to analyze NGC~4472's GC system first because the abundance of Hubble Space Telescope (HST) and spectroscopic data currently available made it possible for us to fully quantify the amount of contamination in our data.  Studying this galaxy has allowed us to test our overall methods and ensure that they are sufficient for accomplishing the goals of the project.  In the following section, the observations and initial data reduction are described. Section~\\ref{section:analysis} describes subsequent analysis steps performed on the data, and Section~\\ref{section:results} gives the results.  Section~\\ref{section:models} consists of a discussion of the implications of the results with regard to various galaxy formation models, and the final section contains a summary of the paper. ",
        "conclusions": "\\label{section:summary}  We have recently begun a study of the GC systems of a large sample of elliptical and spiral galaxies, in order to test the predictions of models for elliptical galaxy formation.  Below we summarize our findings for the first elliptical observed for this study, the giant Virgo galaxy NGC~4472:  1.  A de~Vaucouleurs law of the form log~$\\sigma_{GC}$ = (3.38$\\pm$0.08) $-$ (1.56$\\pm$0.05) $r^{1/4}$ provides a fairly good fit to the radial profile of NGC~4472's GC system, out to $\\sim$16$-$17\\arcmin.  At larger radii, the observed profile falls slightly below the de~Vaucouleurs law, and the number density of GCs is consistent with zero in the region from 21\\arcmin\\ to 23\\arcmin.  2. The bimodal color distribution in NGC~4472's GC system observed by previous authors is apparent in our \\bmr\\ distribution.  The overall ratio of red to blue GCs is $\\sim$0.7.  3. For GCs located within 8\\arcmin\\ of the galaxy center, we find a small negative color gradient: the GC system becomes bluer with increasing radius, due to the increasing ratio of blue to red clusters.  The gradient disappears, however, when the entire GC population out to 22\\arcmin\\ is taken into account; {\\it we find no color gradient in the total GC population.}  4.  The local specific frequency rises approximately linearly with radius out to $\\sim$10\\arcmin\\ ($\\sim$50 kpc).  At larger radii, the local $S_N$ values scatter substantially above and below the values predicted by the de~Vaucouleurs law fit to the radial profile.  In the region from 17$-$23\\arcmin, the observed local $S_N$ appears to turn over and begin to decrease.  Our values of local $S_N$ are significantly lower than those found by Lee \\et (1998) using ground-based data, but agree fairly well with values calculated from an HST study by Puzia \\et (1999). We do {\\it not} find, as suggested by McLaughlin (1999), that local $S_N$ is constant at all radii beyond $\\sim$15 kpc.  5.  There are $\\sim$5,900 clusters in NGC~4472's GC system out to 23\\arcmin, yielding a global specific frequency within that radius of 3.6$\\pm$0.6.  6. We have compared our results with the predictions of four different galaxy formation models, and find that the results are inconsistent with at least one aspect of each model.  Our observed radial profile indicates it is unlikely that a large population of GCs resides at radii beyond $\\sim$23\\arcmin, challenging formation models that require a substantial reservoir of GCs in the far outer regions of galaxies.  The ratio of red to blue clusters for NGC~4472 is at odds with a prediction of the simple merger scenario that the numbers of metal-rich and metal-poor GCs should be comparable.  The lack of a color gradient in the total GC population is not explained nor predicted by any of the models considered.  Finally, although our calculated total number of GCs in NGC~4472 is smaller than previous estimates, it appears that there are still slightly more metal-poor GCs per unit mass in NGC~4472 than can be accounted for by merging the halo GC populations of spiral galaxies like the Milky Way and M31."
    },
    "0010/astro-ph0010253_arXiv.txt": {
        "abstract": "We carry out an extensive study of the cosmic Mach number ($\\mach$) on scales of $R=5, 10$ and $20\\himpc$ using a LCDM hydrodynamical simulation. We particularly put emphasis on the environmental dependence of $\\mach$ on overdensity, galaxy mass, and galaxy age. We start by discussing the difference in the resulting $\\mach$ according to different definitions of $\\mach$ and different methods of calculation. The simulated Mach numbers are slightly lower than the linear theory predictions even when a non-linear power spectrum was used in the calculation, reflecting the non-linear evolution in the simulation. We find that the observed $\\mach$ is higher than the simulated mean $\\expec{\\mach}$ by more than 2-standard deviations, which suggests either that the Local Group is in a relatively low-density region or that the true value of $\\Omega_m$ is $\\sim 0.2$, significantly lower than the simulated value of 0.37. We show from our simulation that the Mach number is a weakly decreasing function of overdensity. We also investigate the correlations between galaxy age, overdensity and $\\mach$ for two different samples of galaxies --- DWARFs and GIANTs. Older systems cluster in higher density regions with lower $\\mach$, while younger ones tend to reside in lower density regions with larger $\\mach$, as expected from the hierarchical structure formation scenario. However, for DWARFs, the correlation is weakened by the fact that some of the oldest DWARFs are left over in low-density regions during the structure formation history. For giant systems, one expects blue-selected samples to have higher $\\mach$ than red-selected ones. We briefly comment on the effect of the warm dark matter on the expected Mach number. ",
        "introduction": "\\label{introduction_section} The cosmic Mach number ``$\\mach$'' is the ratio of the bulk flow ``$V$'' of the velocity field on some scale $R$ to the velocity dispersion ``$\\sigma$'' within the region. It was introduced by \\citet[hereafter OS90] {Ostriker90}, who stressed that it is independent of the normalization of the power spectrum, and is insensitive to the bias between galaxies and dark matter (DM). Basically, it characterizes the warmth or coldness of the velocity field by measuring the relative strength of the velocities at scales larger and smaller than the patch size $R$, so that it effectively measures the {\\em slope} of the power spectrum at the scale corresponding to the patch size. OS90 made rough estimates of $\\mach$ using available observational data on three different scales, and found that the observed $\\mach$ was higher than the expected values of the standard cold dark matter model ($\\Omega_m=1$ where $\\Omega_m$ is the cosmological matter-density divided by the critical density of the universe; hereafter SCDM) in the linear regime by more than a factor of 2 (${\\cal M}_{obs}\\simeq 1-4$ and ${\\cal M}_{SCDM}\\ltsim 1$). Subsequently, \\citet{Suto90}, using N-body simulations, argued that the constraint on $\\mach$ derived by OS90 holds at the 90$\\%$ confidence level, and that the distribution of $\\mach$ is close to Maxwellian in linear and mildly non-linear regimes. \\citet{Park90} has also argued that the biased open CDM models are preferred to the SCDM models using an N-body simulation.   The first serious calculation of $\\mach$ using the first generation of large-scale hydrodynamical simulations which include star formation was carried out by \\citet[hereafter SCO92]{Suto92}. Using this type of simulation enables one to examine the velocity field of galaxies and DM independently without an ad hoc assumption of bias between galaxies and DM. They used the patch size of $R=18$ and $40\\himpc$, and argued that there was no significant difference in $\\mach$ between galaxies and DM, although the galaxies had somewhat larger $\\sigma, V$ and $\\mach$ than did DM. Their best estimate of the mean Mach number derived from SCDM simulations is $\\langle \\mach \\rangle=0.6$, lower than the observational estimate of $\\mach_{obs} \\gtsim 1$.  \\citet[hereafter S93]{Strauss93} made more realistic and direct comparison of observations and models. Accepting the fact that the existing peculiar velocity data do not allow us to compute the ideally defined $\\mach$ as in OS90, they defined a modified Mach number which incorporates the observational errors in measured distances due to the scatter in the Tully-Fisher relation. They constructed a mock catalog of the observations using SCDM hydrodynamical simulations similar to those that were used by SCO92, and calculated their modified $\\mach$ from them. S93 obtained smaller $\\mach$ than did OS90 because they included all velocity components on scales less than the bulk flow into the velocity dispersion, whereas OS90 erased the small-scale dispersion by smoothing. As a consequence, the estimates of S93 on $\\sigma$ is larger, resulting in a smaller $\\mach$. They found that 95\\% of the mock catalogs had smaller $\\mach$ than observed, and that the Mach number test rejects the SCDM scenario at 94\\% confidence level.  Since $\\mach$ is defined as $V/\\sigma$ on a certain scale $R$, a larger $\\mach$ implies a smaller $\\sigma$ if the variation of $V$ is weaker than that of $\\sigma$. Observationally, it has been recognized for at least a decade that the velocity field is very cold outside of clusters \\citep{Brown87, Sandage86, Groth89, Burstein90, Willick97, Willick98}. We ask ourselves in this paper how typical such a cold region of space would be in the entire distribution of the velocity field. We note that \\citet{Weygaert99} also address the same question using N-body simulations which simulate the Local Group via constrained initial conditions.   A closely related quantity is the pairwise velocity dispersion $\\sigma_{12}$, which has been much studied due to its cosmological importance in relation to the ``Cosmic Virial Theorem''. This theorem relates $\\sigma_{12}$ to the two- and three-point correlation functions and $\\Omega_m$. Unfortunately, the determination of $\\sigma_{12}$ is quite unstable and its value differs significantly from author to author \\citep{Mo93, Zurek94, Somerville97, Guzzo97, Strauss98}. This is because $\\sigma_{12}$ is a pair-weighted statistic and is heavily weighted by the objects in the densest regions. Inclusion or exclusion of even $\\sim 10$ galaxies from the Virgo Cluster can change $\\sigma_{12}$ by $\\sim 100 - 200\\kms$, and the correction for the cluster-infall also affects the result significantly.   To overcome this problem, alternative statistics have been suggested. \\citet{Kepner97} proposed the redshift dispersion as a new statistic, and suggested a calculation of the dispersion as a function of local overdensity $\\delta$. They analytically showed that $\\sigma_{12}$ is heavily weighted by the densest regions of the sample. In the same spirit, \\citet{Strauss98} defined and calculated a new measure of $\\sigma_{12}$ as a function of $\\delta$ in redshift space using the Optical Redshift Survey \\citep{Santiago95}, and showed that $\\sigma_{12}$ is indeed an increasing function of $\\delta$. As we will show in this paper, the above statement for $\\sigma_{12}$ holds for the velocity dispersion $\\sigma$ as well; it is heavily weighted by the densest regions. \\citet{Davis97} proposed a single-particle-weighted statistic which measures the one-dimensional rms peculiar velocity dispersion of galaxies, and applied the statistic to the sub-sample of the Optical Redshift Survey and the 1.2 Jy $IRAS$ catalog \\citep{Fisher95}. \\citet{Baker00} applied the same statistic to the Las Campanas Redshift Survey \\citep{Shectman96}, and find that the low-$\\Omega_m$ is favored when the results are compared with N-body simulations. We also note the work by \\citet{Juszkiewicz99} who derived a simple closed-form expression relating the mean pairwise relative velocity $v_{12}$ to the two-point correlation function of mass. Their results can also be used to estimate $\\Omega_m$.  Since the original work of OS90, many things have changed. The resolution and the accuracy of simulations have increased significantly due to the increased computer power and more realistic modelling of galaxy formation. The favored cosmology shifted from SCDM to LCDM ($\\Lam$-dominated flat cold dark matter model), as more and more modern observational data suggest a flat low-$\\Omega_m$ universe with a cosmological constant $\\Lam$ \\citep[\\eg,][]{Efstathiou90, Ostriker95, Turner97, Perlmutter98, Garnavich98, Bahcall99, Balbi00, Lange00, Hu00}. Thus, we are motivated to study this statistic again using a state-of-the-art LCDM hydrodynamical simulation, which allows us to treat baryons and dark matter separately without invoking an ad hoc bias parameter.  In this paper, we calculate $\\sigma, V$, and $\\mach$ with patches of size $R=5, 10$ and $20\\himpc$ using an LCDM hydrodynamical simulation which includes star formation. The details of the simulation are explained in \\S~\\ref{simulation_section} and in Appendix A. We correct for the underestimation of the bulk flow due to the limited size of the simulation box using the linear theory of gravitational instability in \\S~\\ref{theory_section}. In \\S~\\ref{method_section}, we describe the method of the calculation of $V$, $\\sigma$, and $\\mach$. The results of the calculation is presented in \\S~\\ref{result_section}, where we discuss the difference in $\\mach$ resulting from different methods of calculation. The distribution of $\\mach$ is discussed in \\S~\\ref{distribution_section}. In \\S~\\ref{overdensity_section} and \\ref{age_section}, we divide the sample of simulated galaxies by the local overdensity and by their age, and study the correlation with $\\mach$ under the hierarchical structure formation scenario.  All velocities in this paper are presented in the CMB-frame, and the velocity dispersion is 3-dimensional (\\ie, not just the 1-dimensional line-of-sight component). ",
        "conclusions": "\\label{conclusion} We have studied the bulk flow, the velocity dispersion and the cosmic Mach number on scales of $R=5, 10$ and $20\\himpc$ using a  LCDM hydrodynamical simulation, putting emphasis on the environmental effects by the local overdensity, and the correlation with galaxy age and size. Different methods of calculation and the different definitions of $\\mach$ were tried out to see the differences in the result. We found $(\\expec{V^2}/\\expec{\\sigma^2})^{1/2} < \\expec{\\mach} < \\expec{V^2/\\sigma^2}^{1/2}$ (Table~\\ref{table1}), and that the different methods of calculation result in different values of bulk flow and velocity dispersion, hence different Mach number as well. We found that the grouping procedure affects the resulting Mach number significantly. However, the difference in $\\mach$ due to different methods of calculation is smaller than the discrepancy between the simulated and the observed $\\mach$, therefore it is not significant enough to change our following conclusions.  We showed the distribution of the bulk flow, the velocity dispersion, and the Mach number in the simulation (Figure~\\ref{bulk_dist.ps}, \\ref{disp_dist.ps}, and \\ref{mach_dist.ps}). The bulk flows are fitted by a Maxwellian distribution well except that the uncorrected $R=5\\himpc$ case has a longer tail. The velocity dispersion is not well fitted by a Maxwellian; it has a longer tail for $R=5\\himpc$ case, and a sharper cutoff at low values of $\\sigma$ for $R=10$ and 20$\\himpc$ cases. As a result, Mach number is relatively well fitted by a Maxwellian, but with a longer tail for $R=5$ and 10$\\himpc$ cases.   We discussed the theoretical predictions of $\\mach$ in \\S~\\ref{theory_section}, including the scale- and $\\Omega_m$-dependence of $\\mach$. The range of the simulated Mach numbers fall just below the theoretical prediction (\\Fig{mach_scale_plain.ps}), reflecting the non-linear evolution in the simulation which cannot be fully taken into account by simply plugging the non-linear power spectrum into the linear equations. We also discussed in \\S~\\ref{theory_section} how we corrected the simulated bulk flow for the lack of long wavelength perturbations beyond the simulation box size.   The first main conclusion of this paper is that the observed velocity configuration of the Local Group is not the most typical one if the adopted LCDM cosmology is correct. Our calculation shows that the observed Mach numbers are higher than the simulated mean by more than 2-standard deviations at high confidence levels (\\Fig{mach_dist.ps}), and that the observed velocity configuration is off the peak of the number density distribution in the $\\mach-\\sigma$ plane (\\S~\\ref{distribution_section}, \\Fig{machall.ps}). This discrepancy is mainly due to the low observed velocity dispersion, while the observed bulk flow is not that uncommon.   Second, we showed that {\\it the cosmic Mach number is a weakly decreasing function of overdensity} (\\Fig{mach_overdall.ps}). The correlation originates from that between overdensity and the velocity dispersion (\\Fig{vel_overdall.ps}). This is a similar situation to that of the pairwise velocity dispersion. Roughly speaking, high Mach number suggests a low-density environment. It is important to take this overdensity dependence of $\\mach$ into account in any analysis of cosmic Mach number or velocity dispersion, as is the case for the pairwise velocity dispersion.   Third, a few new observational estimates of $\\mach$ were made in this paper on scales of $R=20$ and $30\\himpc$ (\\Fig{mach_scale_overd.ps}). They are much higher than the SCDM prediction, confirming the conclusions of earlier studies by other authors. Combined with our second point, the observed local high-$\\mach$ is simply a reflection of the fact that the Local Group is in a relatively low overdensity region as we know from the $IRAS$ survey. Another possibility which resolves the discrepancy between the simulated and the observed $\\mach$ is that our Universe has a much lower mass density than the simulated value of $\\Omega_m=0.37$. If so, the observed low-$\\sigma$ and high-$\\mach$ would be typical in such universes. As we showed in \\Fig{mach_scale_overd.ps}, the observed Mach numbers are in good agreement with the linear theory prediction with $\\Omega_m=0.2$. This may be interpreted that the local value of $\\Omega_m$ is closer to 0.2 instead of the simulated value of 0.37. We also explored the possibility of the warm dark matter proposal, and found that for a particle mass in the permitted range ($\\gtsim 1{\\rm keV}$, cf. \\citealt{Narayanan00, Bode00}) the effect on the expected Mach number is negligible on scales greater than $1\\himpc$.   Fourth, we studied the correlation between galaxy mass, galaxy age, local overdensity and the Mach number. Our major points are summarized in \\Fig{dwarfgiant.ps}. Using the simulation, we showed that the older (redder) systems are strongly clustered in higher density regions with smaller $\\mach$, while younger (bluer) systems tend to reside in lower density regions with larger $\\mach$ (\\Fig{age_overd.ps}, \\ref{slice.ps}, \\ref{corr.ps} and Table~\\ref{table5}), as expected from the hierarchical structure formation scenario. We divided the galaxy sample into DWARFs and GIANTs in the simulation, and found that the GIANTs follow this expected trend, while DWARFs deviate from this trend on large scales ($R>5\\himpc$) due to the presence of the old DWARFs in low-density regions which did not merge into larger systems. The two point correlation functions and the cumulative number fraction distributions of different populations were presented."
    },
    "0010/astro-ph0010535_arXiv.txt": {
        "abstract": "We have carried out 21 cm radio continuum observations of 16 \\uchii\\ regions using the VLA (D-array) in a search for associated extended emission. We have also observed \\hrrl\\ recombination line towards all the sources and \\herrl\\ line at the positions with strong \\hrrl\\ line emission. The \\uchii\\ regions have simple morphologies and large ($\\gtrsim$10) ratios of single-dish to VLA fluxes. We detected extended emission towards all the sources. The extended emission consists of one to several compact ($\\sim$1$'$ or 0.5$-$5~pc) components and a diffuse extended (2$'$$-$12$'$ or 4$-$19~pc) envelope. All the \\uchii\\ regions but two are located in the compact components, where the \\uchii\\ regions always correspond to their peaks. The compact components with \\uchii\\ regions are usually smaller and denser than those without \\uchii\\ regions. For individual sources, we derive the spectral types (O7$-$O4) of the ionizing stars and the fractions of UV photons absorbed by dust within the nebulae, which are significantly different from previous estimates based on the \\uchii\\ regions alone.  Our recombination line observations show that the ultracompact, compact, and extended components have approximately the same velocity in the individual sources with one exception (G25.72+0.05), implying that they are physically associated. The compact components in each object appear to ionized by separate sources, while the \\uchii\\ regions and their associated compact components are likely to be ionized by the same sources on the basis of the morphological relations mentioned above. This suggests that almost all of the observed \\uchii\\ regions are not `real' \\uchii\\ regions and that their actual ages are much greater than their dynamical age ($\\lesssim$10$^4$~yr). We find that most of simple \\uchii\\ regions previously known have large ratios of single-dish to VLA fluxes, similar to our sources. Therefore, the `age problem' of \\uchii\\ regions does not seem to be as serious as earlier studies argued. We present a simple model in which the coexistence of the ultracompact, compact, and extended components for a long ($>$10$^5$~yr) time is easily explained by combining the Champagne flow model with the hierarchical structure of massive star-forming regions. The well-known relation between the density and diameter of \\hii\\ regions, $n_{\\rm e} \\propto$~D$^{-1}$, is a natural consequence of the hierarchical structure according to our model. We discuss some individual sources. ",
        "introduction": "Ultracompact (UC) \\hii\\ regions are very small ($\\lesssim$0.1~pc) and dense ($\\gtrsim$$10^4$~cm$^{-3}$) ionized regions (Wood $\\&$ Churchwell 1989a, hereafter WC89; Kurtz, Churchwell, $\\&$ Wood 1994, hereafter KCW). They are bright radio and infrared sources and seem to be still embedded in the parental molecular clouds. Thus these objects have been thought to represent an early evolutionary stage of massive stars. One of the most interesting problems related to them is that too many \\uchii\\ regions are observed in the Galaxy, i.e., the ``age problem'': The number of \\uchii\\ regions is greater by about an order of magnitude than expected from other indicators of massive star formation rate based on their dynamical age (WC89; Wood $\\&$ Churchwell 1989b). Hence, it has been considered that \\hii\\ regions should remain in the ultracompact phase for a much longer ($\\gtrsim$10$^5$~yr) period than their sound crossing time ($\\lesssim$10$^4$~yr) in a uniform medium. Several different models were proposed to resolve this problem: the infall model, photo-evaporating disk model, pressure-confined model, and stellar-wind supporting bow shock model (see Churchwell 2000 and references therein). These models present various mechanisms which constrain \\hii\\ regions to remain longer in the ultracompact phase.  On the other hand, there is some observational evidence to suggest that \\uchii\\ regions are {\\it not} as young as they appear. The ultracompactness of \\uchii\\ regions might arise because interferometric observations are insensitive to structures larger than a certain extent depending on the baseline range. Using the Very Large Array (VLA) D-array, which is sensitive to large ($\\sim$15$'$) structures, Koo et al. (1996) found that G5.48$-$0.24, known as an \\uchii\\ region, is actually a large \\hii\\ region complex that contains structures of diverse length scales ranging from 0.04 to 40~pc: It has an ultracompact core, a compact core, an extended halo, and a diffuse large plateau. The presence of the extended envelope could be direct evidence supporting that the \\uchii\\ region is much older than 10$^4$~yr (see \\S~5 for more details). It also implies that there are much more ionizing photons than what would have been estimated from the ultracompact core alone. A few other groups have also noted the importance of extended emission around \\uchii\\ regions in understanding the nature of \\hii\\ regions (Garay et al. 1993; Kurtz et al. 1999). For example, Kurtz et al. (1999) detected extended emission towards 12 out of 15 randomly selected sources from the KCW survey, and suggested on the basis of its morphology that the extended emission may be directly connected with \\uchii\\ regions in about half (8/15) of the sources. However, the physical relationship between \\uchii\\ regions and their extended envelopes is still poorly understood.  In this study, we undertook VLA radio continuum observations of 16 \\uchii\\ regions in search of associated extended emission and detected diffuse extended emission in all the sources. A preliminary version of these results was presented by Kim $\\&$ Koo (1996). We also made \\hrrl\\ and \\herrl\\ radio recombination line (RRL) observations in order to study the physical association between the \\uchii\\ regions and their extended envelopes. The observations are described in \\S~2 and the results are presented in \\S~3. We investigate the relationship between the parameters of the \\uchii\\ regions and their envelopes in \\S~4 and discuss the physical association between the two, the origin of extended envelopes, and their implication in resolving the age problem in \\S~5. Some individual sources are discussed in \\S~6 and our main conclusions are summarized in \\S~7. ",
        "conclusions": "We made VLA radio continuum observations of 16 simple \\uchii\\ regions with large ratios of $S_{\\rm SD}$/$S_{\\rm VLA}$ and detected extended emission towards all the sources. The extended emission consists of one to several compact components and a diffuse envelope. All the \\uchii\\ regions except for two spherical ones (G23.46$-$0.20 and G25.72+0.05) are located in the compact components. There is no significant velocity difference between the ultracompact, compact, and extended components in each source with one exception (G25.72+0.05), which strongly suggests that they are physically associated in almost all of our sources. The \\uchii\\ regions always correspond to the peaks of their associated compact components and the compact components with \\uchii\\ regions are usually more compact than those without \\uchii\\ regions. It is therefore likely that the associated compact components and their immediate envelopes have been produced by UV photons escaping from the \\uchii\\ regions, although we can not completely exclude the possibility that they could have been created by separate ionizing sources in a cluster or by sequentially formed massive stars. Consequently, almost all of the \\uchii\\ regions in our sample are unlikely to be `real' \\uchii\\ regions and their actual ages may be much greater than 10$^4$~yr. The two \\uchii\\ regions with no associated compact components may be `real' \\uchii\\ regions or externally ionized dense clumps.  We calculated the ratios of $S_{\\rm SD}$/$S_{\\rm VLA}$ for the previously known \\uchii\\ regions with simple morphology and found that most of them have large ratios like our sources. Therefore, the age problem of \\uchii\\ regions would be significantly alleviated. We presented a simple model in which the presence of extended emission around \\uchii\\ regions can be understood using both the Champagne flow model and the hierarchical structure of massive star-forming regions. In our model the \\uchii\\ regions and their associated compact and extended components may not represent an evolutionary sequence. The three components could coexist for a long ($>$10$^5$~yr) time for an \\hii\\ region which evolves within a hierarchically clumpy molecular cloud. In this context, the compact components without \\uchii\\ regions seem to be in a later evolutionary phase than those with \\uchii\\ regions. The relation $n_{\\rm e} \\propto$~D$^{-1}$ may also be a corollary of the hierarchical structure of molecular clouds."
    },
    "0010/astro-ph0010229_arXiv.txt": {
        "abstract": "The effects of Hall electromotive forces (HEMFs) on the linear stability of protostellar disks are examined.  Earlier work on this topic focused on axial field and perturbation wavenumbers.  Here we treat the problem more generally.  Both axisymmetric and nonaxisymmetric cases are investigated.  Though seldom explicitly included in calculations, HEMFs appear to be important whenever Ohmic dissipation is.  They allow for the appearance of electron whistler waves, and since these have right-handed polarization, a helicity factor is introduced into the stability problem.  This factor is the product of the components of the angular velocity and magnetic field along the perturbation wavenumber, and it is destabilizing when negative.  Unless the field and angular velocity are exactly aligned, it is always possible to find destabilizing wavenumbers.  HEMFs can destabilize any differential rotation law, even those with angular velocity increasing outward.  Regardless of the sign of the angular velocity gradient, the maximum growth rate is always given in magnitude by the local Oort A value of the disk, as in the standard magnetorotational instability.   The role of Hall EMFs may prove crucial to understanding how turbulence is maintained in the ``low state'' of eruptive disk systems. ",
        "introduction": "The stability of differentially rotating gas disks depends sensitively upon whether or not a magnetic field is present.  A weak (subthermal) magnetic field undermines the stabilizing influence of Coriolis forces (Balbus \\& Hawley 1991), and the resulting turbulence greatly enhances internal angular momentum transport (Hawley, Gammie, \\& Balbus 1995; Balbus \\& Hawley 1998).  This important behavior, which we refer to here as the standard magnetorotational instability (MRI), probably is the underlying cause of ``anomalous viscosity'' in accretion disks.  While other modes of transport (e.g.~waves) are certainly possible, the classical enhanced {\\em turbulent} transport associated with $\\alpha$ disks is almost surely MHD turbulence (Balbus \\& Papaloizou 1999).  Protostellar disks remain problematic.  They seem too dense and too cool to be magnetically well-coupled over their full radial extent. Understanding how protostellar and protoplanetary disks behave is an ongoing theoretical challenge.  It has also been recently noted (Gammie \\& Menou 1998, Menou 2000) that the outer regions of dwarf novae disks may have a very low ionization fractions.  That protostellar disks and dwarf novae both show episodic eruptions may be no coincidence; in both systems the question of magnetic coupling (and therefore the presence of MHD turbulence) is a delicate one.  This paper addresses a very small but fundamental part of this problem: the linear effects of Hall electromotive forces (HEMFs) on disk dynamics.  While the results are more broadly applicable, protostellar disks will be the focus of this paper.  The importance of the Hall effect in modifying the MRI was pointed out in a recent paper by Wardle (1999; hereafter W99), who analyzed the linear stability of a protostellar disk in the presence of a vertical magnetic field.  Such disks are generally in the Hall regime.  That is, their combination of densities, temperatures and ionization fractions places them in a realm of parameter space in which departures from ideal MHD are important, which are strongly influenced by HEMFs.  The Hall effect tends to be important simultaneously with finite conductivity.  Numerical simulations of low ionization disks (e.g., Fleming, Stone, \\& Hawley 2000) have not yet included Hall terms, but have focused rather exclusively on finite conductivity.  W99 found that standard, locally unstable MRI modes may either be stabilized or destabilized depending upon whether $\\bb{\\Omega\\cdot B}$ is positive or negative.  (Here, as usual, $\\bb{\\Omega}$ is the local angular velocity vector of the disk, and $\\bb{B}$ is the magnetic field.)  In other words, HEMFs appear to introduce a helicity effect into the problem.  This immediately raises an interesting question: what happens if $\\bb{\\Omega\\cdot B} = 0$?  Would the Hall effect disappear?  Since the W99 analysis was restricted to axial fields and axial wavenumbers, it could not address this problem.  Clearly HEMFs do not simply vanish when the magnetic field lies in the disk plane, and the identification of $\\bb { \\Omega\\cdot B}$ as the fundamental Hall stability parameter therefore seems suspicious.  Identifying this key parameter more precisely is important, for it is at the heart of MHD stability in low ionization disks.  We are motivated, therefore, to examine the stability of Hall regime accretion under very general circumstances.  We restrict neither the local field geometry, nor the local wavenumber direction.  Furthermore, we have chosen to couch the analysis in a dynamical context, instead of ascribing the inductive Hall terms to effective conductivities (W99). The effective conductivity approach offers terseness and mathematical economy; the dynamical approach makes the physics of the coupling between the magnetic field and the disk coupling more transparent. Both methods must of course ultimately be equivalent.  It is enlightening, however, to view this important problem from a different physical perspective, and we have accordingly chosen this route.  The plan of the paper is as follows.  In \\S 2, we discuss the basic formulation of the problem.  This includes the various physical and compositional parameters, the fundamental equations, and an order of magnitude estimate for the relative sizes of the Ohmic, Hall, and ambipolar diffusion terms in the induction equation.  Section 3 presents the linear dispersion relation for a uniformly rotating disk, because this is the simplest setting to see the interplay between rotation and the Hall EMFs.  Section 4 is a detailed and extensive analysis of differentially rotating disks for arbitrary field geometries and perturbation wavenumbers.  Section 5 summarizes the findings of this paper.   \\section {Basics}  \\subsection {Physical and Compositional Parameters}  A typical protostellar disk consists of molecular gas with cosmic abundances, and spans a temperature range from 10 K in its outer regions to a few $10^3$ K near the central star.  The characteristic disk size may be up to of order 100 AU, and its total mass will typically be less than roughly 0.1 of the central star's mass.  If 0.01 $M_\\odot$ is spread over a cylinder of radius 1 AU and height 0.1 AU, this would yield an average mass density $\\overline{\\rho} \\simeq 2\\times 10^{-10}$ g cm$^{-3}$, and an integrated column density of $\\Sigma \\simeq 3\\times 10^2$ g cm$^{-2}$, fiducial numbers one should bear in mind.  The dynamics of the disk is likely to be dominated by the Keplerian orbital velocity $v_K$: \\beq v_K^2 = {GM\\over R}\\equiv R^2 \\Omega^2 \\eeq where $M$ is the central mass, $R$ the cylindrical radius, and $\\Omega$ the angular velocity.  The disk is assumed to be thin; the characteristic temperature $T$ is much less than any virial temperature.  The scale height $h$ (defined below) satisfies $h/R \\ll 1$.  The number density of species $X$ is denoted $n_X$.  The H$_2$ and He number densities are assumed to be related by \\beq n_{He} = 0.2 n_{H_2}  , \\eeq which implies a total neutral number density \\beq n = n_{H_2} + n_{He} = 1.2 n_{H_2}, \\eeq and a neutral mass density \\beq \\rho = 2.8 n_{H_2} m_p, \\eeq where $m_p$ is the proton mass.  This gives a mean mass per particle of \\beq \\mu = \\rho/n = 2.33 m_p, \\eeq and isothermal and adiabatic sound speeds respectively of \\beq c_S^2 = 0.429 k T/m_p, \\qquad a^2 = 0.6 kT/m_p, \\eeq where $k$ is the Boltzmann constant.  For simplicity we will define the disk scale height $h$ by $h = c_S/\\Omega $.  Under conditions of interest, the dominant ion will generally be once-ionized potassium K${^+}$, and accordingly we take a mean ion mass of \\beq m_i = 39m_p \\eeq In molecular cloud studies, $m_i=30m_p$ is more typically used (e.g., Draine, Roberge, \\& Dalgarno 1983).  The coupling between ions and neutrals depends upon the rate coefficient, denoted $\\langle \\sigma v\\rangle_{iN}$.  In essence, this is the the product of a cross section with the component of the relative velocity along the drift axis, averaged over the electron Maxwellian distribution (Spizter 1978).  Similarly, there is an electron-neutral coupling rate, $\\langle \\sigma v\\rangle_{eN}$. We follow Draine et al.~(1983) and Blaes \\& Balbus (1994), adopting the values \\beq \\langle \\sigma v\\rangle_{iN} = 1.9 \\times 10^{-9} {\\rm cm}^3\\, {\\rm s}^{-1}  \\qquad \\langle \\sigma v\\rangle_{eN} = 10^{-15} \\left( 128kT\\over 9\\pi m_e \\right)^{1/2}= 8.28\\times 10^{-10} T^{1/2} {\\rm cm}^3\\, {\\rm s}^{-1} \\eeq The force per unit volume on the neutral fluid due to ion drag is given by an expression of the form (Shu 1992) \\beq \\bb{f_{ni}} = - \\gamma \\rho \\rho_i (  \\vv - \\bb{v_i}), \\eeq where $\\rho_i$ is the ion mass density, $\\vv$ the neutral velocity, $\\bb{v_i}$ the ion velocity, and $\\gamma$ is the so-called drag coefficient, \\beq \\gamma = {\\langle \\sigma v\\rangle_{iN}\\over m_i + \\mu}= 2.75 \\times 10^{13} {\\rm cm}^3\\, {\\rm s}^{-1} {\\rm g}^{-1}. \\eeq Our numerical value differs somewhat from Draine et al.~(1983) because we use the potassium value for the ion mass $m_i$.  Electron-neutral coupling results in a finite electrical conductivity for the disk gas (Krall \\& Trivelpiece 1973): \\beq \\sigma_e = {n_e e^2\\over m_e \\nu_{eN}} ={n_e e^2\\over m_e n \\langle \\sigma v\\rangle_{eN}}  , \\eeq where $e>0$ is the magnitude of the electron charge, $n_e$ is the electron number density, and $\\nu_{eN}$ is the electron-neutral collision frequency. The associated resistivity is \\beq \\eta = {c^2\\over 4 \\pi \\sigma_e} = 234 \\left(n\\over n_e\\right) T^{1/2} {\\rm cm}^2 {\\rm s}^{-1}, \\eeq where $c$ is the speed of light. The dimensionless measure of the relative importance of resistivity is the magnetic Reynolds number, \\beq Re_M = {v_A h\\over \\eta}, \\eeq where $v_A$ is the magnitude of the Alfv\\'en velocity \\beq \\bb{v_A} = { \\bb{B}\\over \\sqrt{ 4\\pi\\rho} }, \\eeq and $\\bb{B}$ is the magnetic field vector.  Finally, the cyclotron frequencies of the electrons, ions, and neutrals are denoted \\beq \\omega_{c\\alpha} = {eB\\over m_\\alpha c} \\eeq where $\\alpha$ is respectively $e$, $i$, and $\\mu$ ($m_\\mu\\equiv\\mu$). Although the neutrals are of course not directly affected by the magnetic field, the neutral cyclotron frequency is an important characteristic frequency of the analysis.  \\subsection{Equations}  The fundamental dynamical equations are mass conservation, \\begin{equation}\\label{mass} {\\dd\\rho\\over \\dd t} + \\del\\bcdot (\\rho \\bb {v}) =  0 \\end{equation} the dynamical equation of motion, \\beq\\label{mom} \\rho {\\dd\\vv \\over \\dd t} + (\\rho \\vv\\bcdot\\del)\\vv = -\\del\\left( P + {B^2\\over 8 \\pi} \\right)-\\rho \\del \\Phi + \\left( {\\B\\over 4\\pi}\\bcdot \\del\\right)\\B \\eeq and the induction equation, \\beq {\\dd\\B\\over \\dd t} = \\del\\btimes\\left( \\ve \\btimes {\\B} - \\eta \\del\\btimes{\\B} \\right). \\eeq Here, $\\vv$ is the velocity of the neutrals, and $\\ve$ is the electron velocity, and $\\Phi$ is the central gravitational potential.  In equation (\\ref{mom}), which is dominated by the neutral component, we have made implicit use of the low inertia limit for the ions.  Ion momentum balance is set by balancing the ion-neutral drag force with the Lorentz force, \\beq\\label{i_mom} {1\\over c} \\bb{J\\times B} = \\bb{f_{ni}}, \\eeq since all other terms in the ion momentum equation are small. It is actually $\\bb{f_{ni}}$ that appears on the right hand side of equation (\\ref{mom}); the low ion inertia approximation consists of instantaneously balancing this force against the Lorentz force.  \\subsection {The Induction Equation}  Note the appearance of the electron velocity $\\ve$ in the induction equation.  In ideal MHD, the distinction between the fluid velocity and the electron velocity is unimportant.  But in fact, in the absence of resistance, it is the electron fluid in which the magnetic field is ``frozen,'' since the electrons are the mobile charge carriers.  The difference between ion and neutral velocities is responsible for ambipolar diffusion, and the difference between ion and electron velocities gives rise to the Hall effect (e.g., Wardle \\& K\\\"onigl 1993).  In general, \\beq\\label{ve} \\ve = \\vv + (\\ve - \\bb{v_i})+ (\\bb{v_i} - \\vv) = \\vv  - {\\bb{J}\\over n_e e}+{ \\bb{J\\times B}\\over \\gamma\\rho\\rho_i c}, \\eeq We are assuming that any ions present are singly ionized (which is appropriate for a very low ionization gas), so that the electron and ion number densities are equal.  In the ideal MHD limit, both $\\ve - \\bb{v_i}$ and $\\bb{v_i} - \\vv$ are assumed to be negligibly small. Retaining them in the full induction equation leads to  \\beq\\label{ind} {\\dd\\B\\over \\dd t} = \\del\\btimes\\left( \\vv \\btimes {\\B} - {4\\pi\\eta\\bb{J}\\over c} - { \\bb {J\\times B}\\over n_e e} + { \\bb{ (J\\times B)\\times B}\\over c \\gamma\\rho_i\\rho} \\right),  \\eeq where we have used the standard MHD approximation \\beq \\bb{J} = {c\\over4\\pi} \\bb{\\del\\times B}. \\eeq  The four terms on the right side of the induction equation can be arranged by order of magnitude.  Reading from left to right, we denote them $I$ (Inductive), $O$ (Ohmic), $H$ (Hall), and $A$ (Ambipolar).  Assume that typical fluid velocities are of order $v_A$, and that typical gradients are of order an inverse scale-height $1/h$. Then, the relative sizes of these terms are \\beq {O\\over I} \\sim {1\\over  Re_M}, \\quad {H\\over O} \\sim {\\omega_{ce} \\over \\nu_{en}} , \\quad {A\\over H} \\sim {\\omega_{ci}\\over\\gamma\\rho} \\eeq The first ratio just reflects the familiar fact that when the magnetic Reynolds number approaches unity, ohmic losses are important.  The results of \\S 2.1 give \\beq {O\\over I} \\sim {1\\over Re_M} = 4.2\\times 10^{-14} \\left( n\\over n_e\\right) \\left( c_S\\over v_A\\right) \\left(1 \\>{\\rm AU}\\over R\\right)^{3/2} \\left( 10^3\\over T \\right)^{1/2} \\left(M\\over M_\\odot\\right)^{1/2}, \\eeq and \\beq {H\\over O} \\sim {\\omega_{ce} \\over \\nu_{en}} \\simeq \\left( 8\\times 10^{17}\\over n\\right)^{1/2} \\left(v_A\\over c_S\\right) , \\quad {A\\over H} \\sim {\\omega_{ci}\\over\\gamma\\rho} \\simeq \\left( 9\\times 10^{12}\\over n \\right)^{1/2} \\left( T\\over 10^3\\right)^{1/2} \\left(v_A\\over c_S\\right).  \\eeq Under interstellar conditions $1/Re_M$ is tiny, even though $n/n_e$ may be large.  (Note that in equation [13], $h$ would be a characteristic interstellar length, not a disk thickness.)  In this case, neither the $O$ nor the $H$ term is important.  But the ambipolar diffusion $A$ term may be non-negligible.  This leads to a regime of ideal MHD in the ions, which then couple the magnetic field to the neutrals via ion-neutral collisions.  In protostellar disks, our interest here, the far smaller ionization fraction can reduce $Re_M$ to the point where resistivity can affect the dynamics.  This Reynolds number need not be of order unity in an MHD-turbulent fluid before it is significant; some field configurations are affected when $Re_M$ is larger than $10^4$ (Fleming, Stone, \\& Hawley 2000).  The key point here, however, is not the well-known result that resistivity is important in protostellar disks, it is the less appreciated fact that HEMFs will generally be important in a magnetized disk whenever resistivity is.  The two should be treated on the same footing (W99).  The importance of the Hall term relative to Ohmic losses is noteworthy because it is surprisingly general.  As formulated above, for a given ratio of magnetic to thermal energy, the relative importance of the Hall term is {\\it independent\\/} of the ionization level, and depends only upon the number density of the dominant neutrals. Furthermore, the relative importance of the Hall term is not a feature unique to gases of low ionization fraction.  Using a nominal resistivity value of $5\\times 10^{12}T^{-3/2}$ cm$^2$ s$^{-1}$ for a fully ionized plasma (Spitzer 1962), we find \\beq {H\\over O} \\sim {\\omega_{ce}\\over \\nu_{ei}} \\simeq 4 \\left(v_A\\over c_S\\right) \\left(T\\over 10^5\\right)^2 \\left(10^{16}\\over n\\right)^{1/2}, \\eeq where $\\nu_{ei}$ is the electron-ion collision rate. We see that if Ohmic dissipation is of interest (as it would be for magnetic reconnection), the temperature and density regimes of ionized accretion disks imply that the Hall effect cannot be ignored.  At typical disk protostellar disk densities, ambipolar diffusion can be neglected, though it may become important in the outermost regions of the disk.  Since \\beq \\bb{ (J\\times B)\\times B} = \\bb{(J\\cdot B)B} - B^2\\bb{J}, \\eeq if the currents and fields are perpendicular, ambipolar diffusion acts like a field-dependent resistivity.  This is the case for the special vertical field geometry treated by W99 and in \\S 4.2 below: all of ambipolar diffusion may be incorporated into an effective resistivity.  But for the more general field geometries we consider, no such simplification is possible.  We shall accordingly assume a neutral number density much in excess of $10^{13}$ cm$^{-3}$, and ignore ambipolar diffusion.  The form of the induction equation we shall use henceforth is \\beq\\label{finind} {\\dd\\B\\over \\dd t} = \\del\\btimes\\left( \\vv \\btimes {\\B} - \\eta\\bb{\\del\\times B} - { c\\bb {(\\del\\times B)\\times B}\\over 4\\pi  n_e e} \\right).  \\eeq ",
        "conclusions": ""
    },
    "0010/astro-ph0010190_arXiv.txt": {
        "abstract": "We present results on the total mass and temperature determination using two samples of clusters of galaxies. One sample is constructed with emphasis on the completeness of the sample, while the advantage of the other is the use of the temperature profiles, derived with ASCA. We obtain remarkably similar fits to the $M-T$ relation for both samples, with the normalization and the slope significantly different from both prediction of self-similar collapse and hydrodynamical simulations. We discuss the origin of these discrepancies and also combine the X-ray mass with velocity dispersion measurements to provide a comparison with high-resolution dark matter simulations. Finally, we discuss the importance of a cluster formation epoch in the observed $M-T$ relation. ",
        "introduction": "There is increasing interest in using galaxy clusters characterized by X-ray observations as probes for cosmic structure and its evolution. In one of the applications, Press-Schechter models (Press \\& Schechter 1974) are used to constrain the shape and amplitude of the primordial density fluctuation spectrum at present day scales of $5 - 10 h^{-1}$ Mpc (\\eg Henry \\& Arnaud 1991, Oukbir \\& Blanchard 1992, Henry \\etal 1992, Eke \\etal 1996, 1998, Borgani \\etal 1999). Another important measurement is that of the spatial correlations of X-ray clusters or, alternatively, the measurement of the power spectrum of the density fluctuations in the cluster density distribution (\\eg Nichol \\etal 1992, Romer \\etal 1994, Retzlaff \\etal 1998, Miller \\& Batuski 2000, Schuecker \\etal 2000, Collins \\etal 2000). In both cases, the masses of the clusters have to be known for a large test sample in order to apply it to the structure analysis. The mass determination is still only feasible for a limited number of well-observed galaxy clusters, however. Therefore, in the above approaches the empirically-found correlation of the mass with the more easily observable parameters, X-ray luminosity or intracluster gas temperature, are used to make the connection between theoretical modeling and observations. The mass-temperature relation plays a crucial role in this analysis, since it was found to be particularly tight. Since the time of the first X-ray temperature measurements it was observed that the X-ray temperature is closely connected to the velocity dispersion of galaxies in clusters, indicating that both cluster components trace the cluster potential in a similar way (\\eg Mushotzky \\etal 1978, Mushotzky 1984).  The validity of this approach was investigated in numerical calculations based on $N$-body dynamic and hydrodynamic simulations making predictions on the mass-temperature relation and its dispersion, as measured in X-ray studies (Evrard, Metzler, Navarro 1996, Evrard 1997). A close correlation with a dependence of $M \\propto T^{3/2}$ was predicted. It was found in this work and also by \\eg Schindler (1996) that the X-ray mass determination should be reliable (with an uncertainty in the range $14-29$\\%). Evrard \\etal (1996 and Evrard 1997) argue, however, that the predicted mass-temperature relation has such a small dispersion that a mass estimate based on the temperature measurement only is more accurate than that determined on the basis of the additional knowledge of the gas density profile as obtained from $\\beta$-model surface brightness profile fits. These predictions from simulations have been tested recently by means of observational results by several authors (\\eg B\\\"ohringer 1995, Hjorth \\etal 1998, Horner \\etal 1999, Nevalainen \\etal 2000). Disagreements have been found concerning the predicted slope (see \\eg Horner \\etal 1999, Nevalainen \\etal 2000) and the normalization of the mass-temperature relation.  At the time of writing, this discrepancy has not been solved, and various aspects of the data analysis and simulations have been discussed: the correct measurement of temperature gradients in the intracluster gas (\\eg Markevitch \\etal 1998; Irwin, Bregman, Evrard 1999, White 2000), influence of cooling flows (Allen 1998), non-thermal pressure support (\\eg Schindler 1996; Fukazawa \\etal 2000), influence of the heating of the intracluster medium by supernovae (\\eg Balogh, Babul, Patton 1999; Loewenstein 2000; Finoguenov, Arnaud \\& David 2000, hereafter FAD), density and velocity bias in the galactic component of the cluster (Col\\'{\\i}n \\etal 1999) and effects of numerical resolution in the simulations (Nevalainen \\etal 2000).  In this paper we reinvestigate the observational mass-temperature relation using an extended sample of clusters with measured temperature profiles covering a wide range of systems with luminosity-averaged temperatures from below 1 up to 10 keV. In particular, we include 22 low temperature systems, thus improving on sampling the low-mass part of the $M-T$ relation, compared to previous studies (\\eg\\ Horner \\etal 1999; Nevalainen \\etal 2000).  To study in detail the parameters that influence the normalization and the slope of the $M-T$ relation, we rederive the $M-T$ relation assuming an isothermal temperature distribution and also compare with the $M-T$ relation for \\gcs, a statistically complete sample of clusters, described in more detail below, where the isothermality assumption was the only choice. This sample is included in our study primarily to demonstrate that the result of the $M-T$ relation is not affected by any selection bias. In addition, we discuss implications from the results of the $M-T$ relation study on the epoch (redshift) of cluster formation.  The paper is organized as follows. In section \\ref{sec:data} we describe the sample compilation, in \\ref{sec:poly} we compare the results obtained for the two different samples, in \\ref{sec:opt} we combine the X-ray mass with velocity dispersion measurements to provide a comparison with high-resolution dark matter simulations and in \\ref{sec:zf} we discuss a correction for the redshift of cluster formation. Unless noted otherwise, we assume $\\Omega_{\\rm M}=1$, $\\Lambda=0$ and \\h0, $\\rho_{\\rm crit, o}=4.6975\\times10^{-30}$ g cm$^{-3}$ throughout the paper. ",
        "conclusions": ""
    },
    "0010/astro-ph0010473_arXiv.txt": {
        "abstract": "In this paper we use a generalized version of a method originally proposed by Holz and Wald \\cite{art:HolzWald1998} to investigate the effects from gravitational lensing on Type Ia supernovae measurements. We find that results for different mass distributions in smooth dark matter halos are very similar, making lensing effects predictable for a broad range of density profiles. Also, a sample of 100 supernovae at $z\\sim 1$, should be sufficient to discriminate between the case of all dark matter in smooth halos and the extreme case of all dark matter in point-like objects. ",
        "introduction": "Gravitational lensing has become an increasingly important tool in astrophysics and cosmology. In particular, the effects of lensing has to be taken into account when studying sources at high redshifts. In an inhomogeneous universe, sources may be magnified or demagnified with respect to the case of a homogeneous universe with the same average energy density.  Holz and Wald (HW; \\cite{art:HolzWald1998}) have presented a method for determining gravitational lensing effects in inhomogeneous universes. Their use of realistic galaxy models has been limited to the singular, truncated isothermal sphere (SIS) with a fixed mass. We have generalized their method to allow for matter distributions more accurately describing the actual properties of galaxies. The list of matter distributions have been extended to include the density profile proposed by Navarro, Frenk and White (NFW; \\cite{art:NFW}) and we use a distribution of galaxy masses. Also, other matter distribution parameters such as the scale radius of the NFW halo and the cut-off radius of the SIS halo are determined from distributions reflecting real galaxy properties. For further details we refer to Bergstr\\\"om {\\it et al.} \\cite{MGEM-lens} where also the method of HW has been generalized to allow for general perfect fluids with non-vanishing pressure.  As an application of the method, we investigate gravitational lensing effects on observations of distant supernovae. Specifically, we consider the effects on supernova luminosity distributions. ",
        "conclusions": "We have generalized the method of Holz and Wald \\cite{art:HolzWald1998} to allow for matter distributions reflecting the actual properties of galaxies. One of the virtues of this method is that it can be continuously refined as one gains more information about the matter distribution in the universe from observations. The motivation for these generalizations is to use this method as part of a model for simulation of high-redshift supernova observations, {\\em the SuperNova Observation Calculator (SNOC)} .  In this paper, we have considered lensing effects on supernova luminosity distributions. Results for different mass distributions in smooth dark matter halos were found to be very similar, making lensing effects predictable for a broad range of density profiles. Furthermore, given a sample of 100 supernovae at $z\\sim 1$, one should be able to discriminate between the case with smooth dark matter halos and the (unlikely) case of having a dominant component of dark matter in point-like objects."
    },
    "0010/astro-ph0010159_arXiv.txt": {
        "abstract": "We present preliminary results of our work on blazar unification. We assume that all blazars have a broad line region (BLR) and that the size of the BLR scales with the power of the source in a manner similar to that derived through reverberation mapping in radio quiet active galactic nuclei (AGNs). Using a self-consistent emission model that includes particle acceleration we show that according to this scaling, in weak sources  like MKN 421, the inverse Compton (IC) scattering losses are dominated by synchrotron-self Compton scattering (SSC), while in powerful sources, like 3C 279,  they are  dominated by external Compton (EC) scattering of BLR photons. In agreement with other workers, we show that even in the powerful sources that are dominated by EC scattering, the hard X-ray emission is due to SSC. Finally, we show that this scaling reproduces well the observed sequence of blazar properties with luminosity. ",
        "introduction": "Blazars have been shown to exhibit a  sequence of  properties as a function of source power (Fossati et al. 1998). As the source power  increases, the emission line luminosity and the ratio of Compton to synchrotron luminosity increase, while the synchrotron peak frequency $\\nu_{s}$ and the IC peak frequency decrease. Recent multiwavelength studies support this scheme (e.g. Kubo et al. 1998), although the result that the  ratio of the Compton to synchrotron luminosity increases with source luminosity suffers from limited statistics, and should be only considered tentative.  The initial division of blazars into flat spectrum radio quasars (FSRQs) and BL Lacertae objects (BLs) was based on the equivalent width (EW) of the broad emission lines. Sources with EW $>$  5 \\AA $ $ were classified as FSRQs and sources with EW $<$  5 \\AA $ $ as BLs. This difference in the EW of the emission lines has been interpreted as absence of a substantial BLR in BLs, and has been used to advance the idea that in BLs  the GeV-TeV emission is due to  SSC, while in FSRQs the GeV emission is due to EC. However, the EW criterion does not correspond to a dichotomy (Scarpa \\& Falomo 1997), because a weak BLR is present in BLs as well. Additionally, it is not the BLR luminosity $L_B$ that is relevant for the EC luminosity, but the BLR photon energy density $U_{E}$ that is measured in the comoving frame of the non-thermal source. One needs to know not only  $L_B$, but also the BLR radius $R_B$ and the location of the non-thermal emitter relative to the BLR in order to quantify the relative importance of SSC versus EC emission.  The size of the BLR has  been measured only for radio quiet AGNs through reverberation mapping (Kaspi et al. 2000), and it has been found to scale with the ionizing luminosity $L_{acc}$ of the acrretion disk as $R_B \\propto L_{acc}^{0.7}$. It was additionally found that, assuming Keplerian motions for the BLR clouds, the mass of the central object scales with the ionizing luminosity as $M\\propto L_{acc}^{0.5}$.  Here we assume the scalings apply also to radio loud AGN and examine the implications for blazars. Following the formalism of internal shocks propagating in a conical jet (Rees 1978), we scale  the location $D_{blob}$  and the radius $R_{blob}$ of the non-thermal spherical emitter  with the mass $M$ of the central object. We assume that  $L_B$ is a fraction of $L_{acc}$ and scale the kinetic luminosity and the Poynting flux of the blob with  $L_{acc}$. ",
        "conclusions": ""
    },
    "0010/astro-ph0010645_arXiv.txt": {
        "abstract": "We present initial results from a {\\it Hubble Space Telescope\\/} ultraviolet imaging survey of stars known to have hot white-dwarf companions which are unresolved from the ground.  The hot companions, discovered through their EUV or UV emission, are hidden by the overwhelming brightnesses of the primary stars at visible wavelengths. Out of 17 targets observed, we have resolved eight of them with the Wide Field Planetary Camera~2, using various ultraviolet filters. Most of the implied orbital periods for the resolved systems are hundreds to thousands of years, but in at least three cases (56 Persei, $\\zeta$~Cygni, and RE~J1925$-$566) it should be possible to detect the orbital motions within the next few years, and they may eventually yield new dynamically determined masses for the white-dwarf components. The 56~Persei and 14~Aurigae systems are found to be quadruple and quintuple, respectively, including the known optical components as well as the newly resolved white-dwarf companions. The mild barium star $\\zeta $~Cygni, known to have an 18 year spectroscopic period, is marginally resolved. All of these new resolved Sirius-type binaries will be useful in determining gravitational redshifts and masses of the white-dwarf components. ",
        "introduction": "White dwarfs (WDs) represent the evolutionary end point for most stars, since it is believed that all stars with initial masses below $\\approx$$8\\,M_\\odot$ become WDs. An understanding of the evolution of WDs is important in several contexts. For example, the loss of the stellar envelope in the form of a planetary nebula, during the transition from red giant to hot WD, recycles stellar material back into the interstellar medium. Imprinted on the WD luminosity function is the history of star formation in our Galaxy, and the cut-off at $\\approx$3500~K, below which no WDs are known, places a limit of 8--10 Gyr on the age of the disk (e.g. Oswalt \\etal\\ 1996; Leggett, Ruiz, \\& Bergeron 1998).  Theoretical studies have led to the formulation of the mass-radius relation and Chandrasekhar mass limit for zero-temperature WDs (e.g. Chandrasekhar 1931, 1935; Hamada \\& Salpeter 1961), and later to inclusion of the effects of the actual WD temperature due to the presence of a non-degenerate envelope (Wood 1992) and of variations in the core composition (Wood 1995).  Direct tests of the theory are still rare, however, due to the difficulty of obtaining accurate measurements of the radii and masses of known WDs. $T_{\\rm eff}$ and $\\log g$ can be measured from the Balmer-line profiles in optical spectra, using theoretical model atmospheres (Holberg \\etal\\ 1985; Kidder 1991). Bergeron, Saffer, \\& Liebert (1992) used this technique to determine the mass distribution of the population of single H-rich DA WDs, by combining the observational results with the theoretical mass-radius relation of Wood (1992).  A combination of Balmer-line measurements and parallax data can provide an independent means of obtaining WD masses and radii and, therefore, a new route for examining the evolutionary models. Until recently, most parallax measurements of WDs have not been of sufficient accuracy to provide meaningful constraints, but the improvement in ground-based observations (Monet \\etal \\ 1992) and the availability of space-based parallaxes from \\hip\\ (ESA 1997) has improved the situation. Recent results suggest that the masses and radii are generally consistent with expectations (Vauclair \\etal\\ 1997; Barstow \\etal\\ 1997; Provencal \\etal\\ 1998; Holberg \\etal \\ 1998). However, most of this work is restricted to the narrow range of masses near $0.6\\,M_\\odot$, encompassed in the population of isolated WDs. Furthermore, even with the new parallaxes, the overall uncertainties remain too large to provide a really thorough examination of the differences between theoretical models which make differing assumptions about the core and envelope compositions.  Observations of WDs in binaries potentially represent a more direct test of the evolutionary models, since a dynamical WD mass can be determined purely from the orbital elements of the system. In practice, however, very few such systems have been available to be studied in sufficient detail to make such comparisons; only Sirius, Procyon (Girard \\etal \\ 2000), and a handful of other binary systems have provided useful directly measured WD masses to date.  This situation is likely to improve dramatically in the next few years, because the \\ro\\ Wide Field Camera (WFC) and \\eu\\ sky surveys have revealed a number of new sources in which strong EUV emission is detected from a previously unknown hot, but optically faint, companion of a visually bright star. Several similar systems have also been discovered serendipitously in the TD-1 survey data (Landsman \\etal\\ 1996) and in \\iu\\ observations of individual stars (e.g. B\\\"ohm-Vitense 1980, 1993).  This interesting, but rather disparate, group of objects has been the subject of a series of studies (e.g. Barstow \\etal\\ 1994; Vennes \\etal\\ 1995; Genova \\etal\\ 1995; Burleigh, Barstow, \\& Fleming 1997; Burleigh, Barstow, \\& Holberg 1998; Vennes, Christian, \\& Thorstensen 1998). Some of them should be new members of the class of wide binaries represented by Sirius, comprising a normal dwarf or subgiant and a WD companion, and they present an opportunity to extend the sample of well-studied binary systems. Others may belong to much closer binaries, which have passed through a common-envelope interaction; however, with the exception of HR~8210 (a spectroscopic binary with an orbital period of 21.7~days), radial-velocity measurements have not revealed any of the objects to have a sufficiently short binary period (e.g. Vennes \\etal\\ 1998).  None of these systems have been resolved using ground-based imaging, down to typical resolution limits of $0\\secpoint5$ to $1\\secpoint0$, but a major complication is the overwhelming brightness of the primary in the visible band. Since the difference in visual brightness between primary and WD can be 5~mag or more, a WD could easily be hidden in the wings of the image of the primary. This makes the effective resolution several times worse than what can be achieved with smaller magnitude differences.  The {\\it Hubble Space Telescope\\/} (\\HST\\/) and its Wide Field Planetary Camera~2 (WFPC2) provide the answer to these problems. WFPC2 allows imaging in the ultraviolet, where the brightnesses of both components become comparable. Moreover, WFPC2 delivers diffraction-limited images at UV wavelengths of $\\approx0\\secpoint08$ FWHM, making it possible to resolve binary components with separations down to about this amount.  In this paper we report the initial results of an \\HST\\/ ``snapshot'' survey for resolved Sirius-like binary systems.  Out of 17 targets observed to date, we have resolved 8 of them.  In the following sections we discuss our analysis of the WFPC2 images, the nominal orbital periods for the resolved systems, and the limits on the orbital periods for those binaries that we were unable to resolve. In addition, we discuss the results for each individual object in the context of their known properties from earlier observations. ",
        "conclusions": "Using \\HST's WFPC2 camera at UV wavelengths, we are carrying out a systematic ``snapshot'' imaging survey of binary systems, each comprising a non-degenerate primary with a WD secondary. The majority of these systems were discovered through the EUV or UV emission from the WD, which is swamped by the overwhelming brightness of the primary star at visible wavelengths. All of our targets are unresolved in ground-based observations.  Of 17 systems observed to date, WFPC2 has resolved eight of them. These are new analogs of the well-known Sirius system.  Several of them, notably 56~Per, $\\zeta$~Cyg, and perhaps RE~J1925+566 may have orbital periods short enough for the visual orbits to be determined in the foreseeable future, providing a significant augmentation to the small number of WDs with dynamically determined masses.  Two of the targets were found to be interesting quadruple (56~Per) or quintuple (14~Aur) multiple systems, providing a sample of ``mini-clusters'' which should provide new constraints on WD evolution.  All of the resolved systems provide new targets for determinations of gravitational redshifts."
    },
    "0010/astro-ph0010651_arXiv.txt": {
        "abstract": "In this talk we briefly review the inverse Compton (IC) scattering of nuclear photons in the lobes of radio galaxies and quasars. We concentrate on the possibility to test this model with the {\\it Chandra} observatory and to constrain the electron energy distribution by measuring the X--ray fluxes produced by this effect. We also discuss the evidence for IC scattering of nuclear photons in powerful radio galaxies. ",
        "introduction": "A full understanding of the energetics and of the energy distribution of the relativistic particles in the radio jets and lobes of radio galaxies and quasars is of basic importance to a complete description of the physics and time evolution of these sources. Synchrotron emission properties result from a complicated convolution of magnetic fields and relativistic electron distributions, so that they cannot be unambiguously derived by radio observations alone (see Rudnick, this meeting). In the lobes of radio galaxies and quasars this degeneracy could be broken by measurements of the X-rays produced by the inverse Compton (IC) scattering of CMB photons (e.g. Harris \\& Grindlay 1979), and/or nuclear photons (Brunetti, Setti \\& Comastri 1997).  While the X--rays produced by the IC scattering of the CMB photons typically involve the less energetic part of the radio emitting electrons (Lorentz factor $\\gamma \\sim 1000$), those from the IC scattering of the far IR -- optic nuclear photons are mainly powered by $\\gamma=100-300$ electrons which are not detected, since their synchrotron emission falls in the hundred kHz to MHz frequency range. By combining measurements of the 0.1--10 keV extended fluxes it is then possible to obtain broad band information on the energy distribution of the relativistic electrons ($\\gamma \\sim 50-3000$) in the lobes of radio galaxies and quasars.  The knowledge of the low energy part of the electron spectrum is particularly important. Indeed, while at higher energies radiative losses are dominant, at lower energies the electron spectrum is particularly sensitive to the initial injection conditions and to the acceleration mechanisms (Eilek \\& Huges 1990). Furthermore, the detection of X--ray fluxes produced by IC scattering of nuclear photons, implying the lack of a prominent low energy cut off ($\\gamma < 300$) in the electron energy distribution, may significantly increase the energy budget in the radio lobes compared with present estimates based on standard equipartition formulae (Brunetti et al. 1997). ",
        "conclusions": ""
    },
    "0010/astro-ph0010467_arXiv.txt": {
        "abstract": "The formation of the first stars and quasars marks the transformation of the universe from its smooth initial state to its clumpy current state. In popular cosmological models, the first sources of light began to form at a redshift $z=30$ and reionized most of the hydrogen in the universe by $z=7$. Current observations are at the threshold of probing the hydrogen reionization epoch. The study of high-redshift sources is likely to attract major attention in observational and theoretical cosmology over the next decade. \\vskip 0.5truein {\\it Key Words: Cosmology, First Galaxies, Intergalactic Medium} \\vspace{.4in} ",
        "introduction": "\\label{sec1}  The detection of cosmic microwave background (CMB) anisotropies (Bennett et al.\\  1996; de Bernardis et al.\\  2000; Hanany et al.\\  2000) confirmed the notion that the present large-scale structure in the universe originated from small-amplitude density fluctuations at early times. Due to the natural instability of gravity, regions that were denser than average collapsed and formed bound objects, first on small spatial scales and later on larger and larger scales. The present-day abundance of bound objects, such as galaxies and X-ray clusters, can be explained based on an appropriate extrapolation of the detected anisotropies to smaller scales. Existing observations with the {\\it Hubble Space Telescope}\\, (e.g., Steidel et al.\\ 1996; Madau et al.\\ 1996; Chen et al.\\ 1999; Clements et al.\\ 1999) and ground-based telescopes (Lowenthal et al.\\ 1997; Dey et al.\\ 1999; Hu et al.\\ 1998, 1999; Spinrad et al.\\ 1999; Steidel et al.\\ 1999), have constrained the evolution of galaxies and their stellar content at $z\\la 6$.  However, in the bottom-up hierarchy of the popular Cold Dark Matter (CDM) cosmologies, galaxies were assembled out of building blocks of smaller mass. The elementary building blocks, i.e., the first gaseous objects to form, acquired a total mass of order the Jeans mass ($\\sim 10^4 M_\\odot$), below which gas pressure opposed gravity and prevented collapse (Couchman \\& Rees 1986; Haiman \\& Loeb 1997; Ostriker \\& Gnedin 1997). In variants of the standard CDM model, these basic building blocks first formed at $z\\sim 15$--$30$.  An important qualitative outcome of the microwave anisotropy data is the confirmation that the universe started out simple. It was by and large homogeneous and isotropic with small fluctuations that can be described by linear perturbation analysis. The current universe is clumpy and complicated.  Hence, the arrow of time in cosmic history also describes the progression from simplicity to complexity (see Figure~\\ref{fig1a}). While the conditions in the early universe can be summarized on a single sheet of paper, the mere description of the physical and biological structures found in the present-day universe cannot be captured by thousands of books in our libraries.  The formation of the first bound objects marks the central milestone in the transition from simplicity to complexity.  Pedagogically, it would seem only natural to attempt to understand this epoch before we try to explain the present-day universe. Historically, however, most of the astronomical literature focused on the local universe and has only been shifting recently to the early universe. This violation of the pedagogical rule was forced upon us by the limited state of our technology; observation of earlier cosmic times requires detection of distant sources, which is feasible only with large telescopes and highly-sensitive instrumentation.  \\begin{figure} \\psfig{file=figI1.ps,width=4.8in} \\caption{Milestones in the evolution of the universe from simplicity to complexity. The ``end of the dark ages'' bridges between the recombination epoch probed by microwave anisotropy experiments ($z\\sim 10^3$) and the horizon of current observations ($z\\sim 5$--6).} \\label{fig1a} \\end{figure}  For these reasons, advances in technology are likely to make the high redshift universe an important frontier of cosmology over the coming decade. This effort will involve large (30 meter) ground-based telescopes and will culminate in the launch of the successor to the {\\it Hubble Space Telescope},\\, called {\\it Next Generation Space Telescope}\\footnote{More details about the telescope can be found at http://ngst.gsfc.nasa.gov/} (\\NGST\\,). \\NGST,\\, planned for launch in 2009, will image the first sources of light that formed in the universe. With its exceptional sub-nJy ($1~{\\rm nJy}=10^{-32}{\\rm erg~cm^{-2}~s^{-1}~Hz^{-1}}$) sensitivity in the 1--3.5$\\mu$m infrared regime, \\NGST\\, is ideally suited for probing optical-UV emission from sources at redshifts $\\ga 10$, just when popular Cold Dark Matter models for structure formation predict the first baryonic objects to have collapsed.  The study of the the formation of the first generation of sources at early cosmic times (high redshifts) holds the key to constraining the power-spectrum of density fluctuations on small scales. Previous research in cosmology has been dominated by studies of {\\it Large Scale Structure} (LSS); future studies are likely to focus on {\\it Small Scale Structure} (SSS).  The first sources are a direct consequence of the growth of linear density fluctuations. As such, they emerge from a well-defined set of initial conditions and the physics of their formation can be followed precisely by computer simulation. The cosmic initial conditions for the formation of the first generation of stars are much simpler than those responsible for star formation in the Galactic interstellar medium at present. The cosmic conditions are fully specified by the primordial power spectrum of Gaussian density fluctuations, the mean density of dark matter, the initial temperature and density of the cosmic gas, the primordial composition according to Big-Bang nucleosynthesis, and the lack of dynamically-significant magnetic fields. The chemistry is much simpler in the absence of metals and the gas dynamics is much simpler in the absence of dynamically-important magnetic fields.  The initial mass function of the first stars and black holes is therefore determined by a simple set of initial conditions (although subsequent generations of stars are affected by feedback from photoionization heating and metal enrichment).  While the early evolution of the seed density fluctuations can be fully described analytically, the collapse and fragmentation of nonlinear structure must be simulated numerically. The first baryonic objects connect the simple initial state of the universe to its complex current state, and their study with hydrodynamic simulations (e.g., Abel, Bryan, \\& Norman 2000; Bromm, Coppi, \\& Larson 1999) and with future telescopes such as \\NGST\\, offers the key to advancing our knowledge on the formation physics of stars and massive black holes.  The {\\it first light} from stars and quasars ended the ``dark ages'' \\footnote{The use of this term in the cosmological context was coined by Sir Martin Rees.} of the universe and initiated a ``renaissance of enlightenment'' in the otherwise fading glow of the microwave background (see Figure~\\ref{fig1a}).  It is easy to see why the mere conversion of trace amounts of gas into stars or black holes at this early epoch could have had a dramatic effect on the ionization state and temperature of the rest of the gas in the universe. Nuclear fusion releases $\\sim 7\\times 10^6$ eV per hydrogen atom, and thin-disk accretion onto a Schwarzschild black hole releases ten times more energy; however, the ionization of hydrogen requires only 13.6 eV. It is therefore sufficient to convert a small fraction, $\\sim 10^{-5}$ of the total baryonic mass into stars or black holes in order to ionize the rest of the universe. (The actual required fraction is higher by at least an order of magnitude [Bromm, Kudritzky, \\& Loeb 2000] because only some of the emitted photons are above the ionization threshold of 13.6 eV and because each hydrogen atom recombines more than once at redshifts $z\\ga 7$). Recent calculations of structure formation in popular CDM cosmologies imply that the universe was ionized at $z\\sim 7$--12 (Haiman \\& Loeb 1998, 1999b,c; Gnedin \\& Ostriker 1998; Chiu \\& Ostriker 2000; Gnedin 2000a). Current observations are at the threshold of probing this epoch of reionization, given the fact that galaxies and quasars at redshifts $\\sim 6$ are being discovered (Fan et al.\\ 2000; Stern et al.\\ 2000). One of these sources is a bright quasar at $z=5.8$ whose spectrum is shown in Figure~\\ref{fig1c}. The plot indicates that there is transmitted flux shortward of the Ly$\\alpha$ wavelength at the quasar redshift. However, Gunn \\& Peterson (1965) showed that even a tiny neutral hydrogen fraction in the intergalactic medium would produce a large optical depth at these wavelengths. Indeed, modeling of the transmitted flux (Fan et al.\\ 2000) implies an optical depth $\\tau<0.5$ or a neutral fraction $x_{\\HI}\\la 10^{-6}$, i.e., the universe is fully ionized at $z=5.8$!  One of the important challenges of future observations will be to identify {\\it when and how the intergalactic medium was ionized}. Theoretical calculations (see \\S \\ref{sec6.3.1}) imply that such observations are just around the corner.  \\begin{figure} \\psfig{file=figI3.ps,width=5in} \\caption{Optical spectrum of the highest-redshift known quasar at $z=5.8$, discovered by the Sloan Digital Sky Survey (Fan et al.\\  2000). } \\label{fig1c} \\end{figure}  Figure~\\ref{fig1d} shows schematically the various stages in a theoretical scenario for the history of hydrogen reionization in the intergalactic medium. The first gaseous clouds collapse at redshifts $\\sim 20$--$30$ and fragment into stars due to molecular hydrogen (H$_2$) cooling. However, H$_2$ is fragile and can be easily dissociated by a small flux of UV radiation (Stecher \\& Williams 1967; Haiman, Rees, \\& Loeb 1997).  Hence the bulk of the radiation that ionized the universe is emitted from galaxies with a virial temperature $\\ga 10^4$ K, where atomic cooling is effective and allows the gas to fragment (see Haiman, Abel, \\& Rees 1999 for an alternative scenario).  \\begin{figure} \\psfig{file=figI4new.ps,width=6in} \\caption{Stages in the reionization of hydrogen in the intergalactic medium. } \\label{fig1d} \\end{figure}  Since recent observations confine the standard set of cosmological parameters to a relatively narrow range, we assume a $\\Lambda$CDM cosmology with a particular standard set of parameters in the quantitative results in this review. For the contributions to the energy density, we assume ratios relative to the critical density of $\\Omm=0.3$, $\\Oml=0.7$, and $\\Omega_b=0.045$, for matter, vacuum (cosmological constant), and baryons, respectively. We also assume a Hubble constant $H_0=70\\mbox{ km s}^{-1}\\mbox{Mpc}^{-1}$, and a primordial scale invariant ($n=1$) power spectrum with $\\sigma_8=0.9$, where $\\sigma_8$ is the root-mean-square amplitude of mass fluctuations in spheres of radius $8\\ h^{-1}$ Mpc. These parameter values are based primarily on the following observational results: CMB temperature anisotropy measurements on large scales (Bennett et al.\\ 1996) and on the scale of $\\sim 1^\\circ$ (Lange et al.\\ 2000; Balbi et al.\\ 2000); the abundance of galaxy clusters locally (Viana \\& Liddle 1999; Pen 1998; Eke, Cole, \\& Frenk 1996) and as a function of redshift (Bahcall \\& Fan 1998; Eke, Cole, Frenk, \\& Henry 1998); the baryon density inferred from big bang nucleosynthesis (see the review by Tytler et al.\\ 2000); distance measurements used to derive the Hubble constant (Mould et al.\\ 2000; Jha et al.\\ 1999; Tonry et al.\\ 1997); and indications of cosmic acceleration from distances based on type Ia supernovae (Perlmutter et al.\\ 1999; Riess et al.\\ 1998).  This review summarizes recent theoretical advances in understanding the physics of the first generation of cosmic structures. Although the literature on this subject extends all the way back to the sixtees (Saslaw \\& Zipoy 1967, Peebles \\& Dicke 1968, Hirasawa 1969, Matsuda et al.\\ 1969, Hutchins 1976, Silk 1983, Palla et al.\\ 1983, Lepp \\& Shull 1984, Couchman 1985, Couchman \\& Rees 1986, Lahav 1986), this review focuses on the progress made over the past decade in the modern context of CDM cosmologies. ",
        "conclusions": ""
    },
    "0010/astro-ph0010184_arXiv.txt": {
        "abstract": "Variations of $\\sim0.4$~mag are expected in the $I$-band absolute magnitude of red clump giants, $M_I^{\\rm RC}$, as a function of both stellar age and metallicity. This regardless of some potential theoretical uncertainties. Due to the quite large differences in mean ages and metallicities of clump stars among galaxies, systematic changes (also amounting up to $\\sim0.4$~mag) come out in their $M_I^{\\rm RC}$. These numbers also indicate a distance to the LMC that is not necessarily ``short''. ",
        "introduction": " ",
        "conclusions": ""
    },
    "0010/astro-ph0010521_arXiv.txt": {
        "abstract": "In the standard picture of structure formation, initially random-phase fluctuations are amplified by non-linear gravitational instability to produce a final distribution of mass which is highly non-Gaussian and has highly coupled Fourier phases. Second-order statistics, such as the power spectrum, are blind to this kind of phase association. We discuss the information contained in the phases of cosmological density fluctuations and their possible use in statistical analysis tools. In particular, we show how the bispectrum measures a particular form of phase association called quadratic phase coupling, show how to visualise phase association using colour models. These techniques offer the prospect of more complete tests of initial non-Gaussianity than those available at present. ",
        "introduction": "The local Universe displays a rich hierarchical pattern of galaxy clusters and superclusters~\\cite{lasc}. The early Universe, however, was almost smooth, with only slight ripples seen in the cosmic microwave background radiation~\\cite{Cobe}. Models of the evolution of structure link these observations through the effect of gravity, because the small initially overdense fluctuations attract additional mass as the Universe expands~\\cite{p80}. During the early stages, the ripples evolve independently, like linear waves on the surface of deep water. As the structures grow in mass, they interact with other in non-linear ways, more like waves breaking in shallow water. Cosmic structure can be characterized by phase correlations associated with these non-linear interactions, but this information is missed by standard analysis techniques such as the power spectrum. In order to do justice to the large data sets about to become available, it is important to design techniques sensitive to the fine details of cosmic structure they will reveal. Here we report a method of quantifying phase information \\cite{cc} and suggest how this information may be exploited to build novel statistical descriptors that can be used to mine the sky more effectively than with standard methods. ",
        "conclusions": "In fairly recent history,  cosmological data sets were sparse and incomplete, and the statistical methods deployed to analyse them were crude. Second-order statistics, such as $P(k)$ and $\\xi(r)$, are blunt instruments that throw away the fine details of the delicate pattern of cosmic structure. These details lie in the distribution of Fourier phases to which second-order statistics are blind. It would not do justice to massively improved data if effort were directed only to better estimates of these quantities. Moreover, as we have shown, phase information provides a unique fingerprint of gravitational instability developed from Gaussian initial conditions (which have maximal phase entropy). Methods such as those we have described above can therefore be used to test this standard paradigm for structure formation. They can also furnish direct tests of the presence of initial non-Gaussianity ~\\cite{fmg,pvf,bt}. As the raw material is increasing in both quality and quantity, it is time to refine our statistical technology so that the subtle and precious artifacts previously ignored can be both detected and extracted."
    },
    "0010/astro-ph0010247_arXiv.txt": {
        "abstract": "We investigate the impact of recent observational advances in research on the gaseous content of the universe on our knowledge of star formation histories of disk galaxies.  Several contemporary surveys indicate that large quantities of gas which has not been astrated or has been astrated only weakly are bound to galaxies at later epochs ($z\\le 1$). ",
        "introduction": " ",
        "conclusions": ""
    },
    "0010/astro-ph0010071_arXiv.txt": {
        "abstract": "We detect the weak gravitational lensing distortion of 450,000 background galaxies ($20<R<23$) by 790 foreground galaxies ($R<18$) selected from the Las Campanas Redshift Survey(LCRS).  This is the first detection of weak lensing by field galaxies of known redshift, and as such permits us to reconstruct the shear profile of the typical field galaxy halo in absolute physical units (modulo $H_0$), and to investigate the dependence of halo mass upon galaxy luminosity.  This is also the first galaxy-galaxy lensing study for which the calibration errors due to uncertainty in the background galaxy redshift distribution and the seeing correction are negligible. Within a projected radius of 200 \\hkpc, the shear profile is consistent with an isothermal profile with circular velocity $v_c=164\\pm20$ \\kms\\ for an $L_\\ast$ galaxy, consistent with the typical circular velocity for the disks of spirals at this luminosity. This halo mass normalization, combined with the halo profile derived by \\citet{F00} from a galaxy-galaxy lensing analysis of the Sloan Digital Sky Survey, places a lower limit of $(2.7\\pm0.6)\\times10^{12}h^{-1}M_\\odot$ on the mass of an $L_\\ast$ galaxy halo, in good agreement with the satellite galaxy studies of \\citet{Za97} Given the known luminosity function of LCRS galaxies, and assumption that $M\\propto L^\\beta$ for galaxies, we determine that the mass within 260\\hkpc\\ of normal galaxies contributes $\\Omega=0.16\\pm0.03$ to the density of the Universe (for $\\beta=1$) or $\\Omega=0.24\\pm0.06$ for $\\beta=0.5$.  These lensing data suggest that $0.6<\\beta<2.4$ (95\\% CL), only marginally in agreement with the usual $\\beta\\approx0.5$ Faber-Jackson or Tully-Fisher scaling. This is the most complete direct inventory of the matter content of the Universe to date. ",
        "introduction": "Gravitational lensing provides the only non-dynamical measure of the mass distribution of a galaxy halo, and it thus enables a crucial means of quantifying the dark matter halos that are now an essential component of all theories of cosmology and galaxy formation.  A reliable measure of the masses of galaxies and their halos also provides a significant lower limit to the density parameter $\\Omega$. Spiral galaxy halos have traditionally been probed via rotation of stars and gas, but these tracers are not available outside radii of $\\sim30$~kpc, though their velocities are consistent with with an isothermal halo extending beyond this distance.  At larger radii, dynamical test particles are scarce: \\citet{Za97} use satellite galaxies to trace the halo mass profile to $\\approx200$~kpc. Because there are only $\\approx1.2$ satellites per spiral, the masses of individual galaxies are poorly constrained, but by ``stacking'' the signals from an ensemble of galaxies, an accurate measure of the mean halo is obtained.  A further complication of such dynamical studies of the outer halo is that the test particles are not virialized, and determination of the halo mass requires some model of infall.  With such modelling, \\citet{Za97} estimate a mass of $\\sim2\\times10^{12}\\,M_\\odot$ within 200~kpc of an $L_\\ast$ spiral galaxy.  The gravitational lensing approach uses photons from background galaxies (bggs) as the test particles, enabling halo measurements to large radii.  The greatest problem with lensing is the extremely weak shear ($\\lesssim3\\%$) associated with the mass of a single galaxy.  This is more than a factor of 10 below the noise due to intrinsic shape variations in a single background galaxy.  As with the satellite galaxy study, the weak lensing approach must therefore overcome the poor $S/N$ per galaxy by using large numbers of lens-source pairs, measuring the average galaxy halo. The weak lensing measurement, however, requires no modeling of the dynamical state and can give a non-parametric estimate of the halo profile.  The first attempted detection was \\citet{Ty84}, and over the last five years several groups have verified the existence of galaxy-galaxy lensing based upon analysis of a small number of deep fields \\citep{Br96, DA96, Gr96, Hu98}. Interpretation of these results is problematic not only because of poor $S/N$, but also because the redshifts of the lens galaxies (and hence their distances and luminosities) are not known individually.  Interpretation relies upon positing a model for the joint distance/luminosity distribution of the lenses and sources, and a model for dependence of halo mass and size upon luminosity, then projecting these models down to a function of angular variables to be compared with the observations.  In such deep fields, even the statistical distributions of lens and source distances are currently ill-determined.  More recently \\citet[F00]{F00} have measured a galaxy-galaxy lensing signal at high significance from preliminary Sloan Digital Sky Survey data.  These data are relatively shallow (foreground galaxies $16<r'<18$, background galaxies $18<r'<22$), which weakens the shear signal, but this is more than compensated by the extremely large sample size of $\\approx16$ million fgg/bgg pairs. These data indicate that halos of typical galaxies continue an isothermal profile to a radius of at least 260~\\hkpc.  The conversion from shear amplitude to mass density in the F00 data is still reliant upon photometric redshift distributions for the fggs and bggs.  We present here a detection of galaxy-galaxy lensing around fggs with redshifts determined by the Las Campanas Redshift Survey \\citep{Sc96} We thus have knowledge of the fgg luminosity and distance, and the impact parameter of the lensed photons, in physical units ({\\it i.e.\\/} kpc rather than arcsec), information that is normally lost in projection.  The bgg's ($R<23$) are at magnitudes within the reach of current pencil-beam redshift surveys, and hence their distance distribution is well enough determined to introduce negligible uncertainty in the calibration.  We are therefore in theory able to measure the galaxy luminosity $L$, halo mass $M$, and surface mass density $\\Sigma$ in physical units without ambiguity (save the scale factor $H_0$).  In practice our $S/N$ is too low to constrain the extent of the galaxy halo meaningfully, so we combine the F00 profile {\\it shape} with our {\\it normalization} to obtain a mass profile in absolute units. Combining the resultant $M/L$ with the well-determined luminosity function of the LCRS galaxies \\citep{Li96} we obtain the galactic contribution to $\\Omega$. We assume an Einstein-deSitter Universe for the values presented here, and use $H_0=100h {\\rm \\,km\\, s}^{-1}\\,{\\rm Mpc}^{-1}$.  Because the fggs are relatively nearby, and the bggs are far behind them, the dependence of the results upon the cosmological parameters $\\Omega$ and $\\Lambda$ is negligible ($<2\\%$). ",
        "conclusions": "The relations between galaxy luminosity and halo mass determined herein provide basic constraints to models of galaxy formation and evolution.  We find, perhaps surprisingly, that the halo circular velocity at 60--100\\hkpc\\ radii are similar to the disk circular velocities measured at $\\lesssim10\\hkpc$, extending the ``disk-halo conspiracy'' to larger radii than previously known. There are hints in the data, however, that this conspiracy breaks down in the sense that halo masses vary more sharply with galaxy luminosity then would be indicated by the Tully-Fisher or Faber-Jackson relations.  The matter density $\\Omega_m$ has been measured by many methods in the past, and there is a currently fashionable consensus that $\\Omega_m\\approx0.3$.  Our results (Equation~[\\ref{omega}]) are in agreement with this consensus given that our method produces only a lower limit.  Our limits are, however, close to the consensus value, especially if $\\beta\\approx0.5$. Our lower limit on $\\Omega$ is determined by a direct inventory of the matter in the Universe; the only untested assumption is that the formula for gravitational deflection of light can be extrapolated from Solar-System scale (where it has been verified) to galactic scales. Other determinations of $\\Omega$ rely upon assumptions about the nature of our Universe. Agreement between our direct inventory and these other measures serves as valuable and unique verification that these assumptions---often fundamental to our current view of cosmology---are correct.  The more precise estimates of $\\Omega_m$ include the following: Dynamical studies of galaxy clusters, combined with the assumption that the $M/L$ ratio for clusters is universal, yield $\\Omega_m=0.24^{+0.05}_{-0.09}$ \\citep{Ca96}. Another cluster-based means to $\\Omega$ is to measure the fraction $f_b$ of the cluster matter that is baryonic, assume that the cluster $f_b$ is universal, and then use the baryon density $\\Omega_b$ from Big Bang nucleosynthesis to estimate $\\Omega_m=\\Omega_b/f_b$.  \\citet{Mo99} conclude that $\\Omega_m<0.30\\pm0.04$ (for $h=0.7\\pm0.1$).  Other estimates of $\\Omega$ measure its effects upon global geometry or the growth rate of fluctuations in the Universe.  In the former category, the combination of high-redshift supernovae measurements and the location of the cosmic background Doppler peaks measured by MAXIMA-1 constrain $0.25<\\Omega_m<0.50$ at 95\\% CL \\citep{Ba00}. Peculiar velocity studies determine the parameter $\\beta_I\\approx\\Omega^{0.6}/b$ to be $\\approx0.5\\pm0.05$ \\citep{St98}, with $b$ a galaxy bias parameter.  Taking $b\\approx1$ gives a value of $\\Omega\\approx0.3$, again slightly above our halo density. Conversely our lower limit on $\\Omega$ implies a bias parameter for IRAS galaxies $b\\gtrsim0.7$.  Thus our results may be taken as independent evidence that these underlying assumptions are correct:  cluster $M/L$ and baryon fractions can be taken as representative; general relativity does properly describe the geometry of the Universe to $z\\approx1000$ given our directly measured matter content; and large-scale motions are consistent with the predictions of gravitational instability.  Conversely, we may take the $\\Omega_m\\approx0.3$ value as truth and conclude from these lensing results that most of the matter in the Universe is indeed contained within 260\\hkpc\\ of normal galaxies. This implies that galaxies missed by the LCRS survey---{\\it e.g.} low-surface-brightness galaxies or exotic dark galaxies---cannot be the dominant reservoirs of mass in the Universe.  Likewise there cannot be a dominant matter component which is not clustered around galaxies.  Most of the mass in the Universe appears to have been located, though of course its material nature remains unknown.  The accuracy of our results is currently limited by the number of foreground/background pairs in our survey.  This will be remedied in dramatic fashion by the full Sloan Digital Sky Survey, which will obtain imaging to $r^\\prime<22$ around roughly {\\em three orders of magnitude} more foreground galaxies than used in this study with a concomitant increase in the number of pairs.  This will allow for non-parametric mapping of the shear profile as a function of radius, galaxy luminosity, and galaxy type, with much lower noise than in our LCRS lensing survey.  This weak lensing information will be unique in unveiling the relation between the various classes of galaxies and the dark halos in which they reside. With the random errors reduced, closer attention will have to be paid to some of the $\\lesssim5\\%$ corrections we have ignored (as in F00), but these are tractable.  Our estimates of $M_\\ast$ and $\\Omega$ are lower bounds since the profile of the halo has not yet been determined beyond 260\\hkpc. Beyond this radius it becomes inappropriate to speak of ``the halo'' of a galaxy, as the mass profile of a given galaxy becomes inseparable from the mass associated with neighboring galaxies. More stringent limits on $\\Omega$ will thus require a more elaborate formalism for describing the effects of galaxy correlations upon the weak lensing measurements."
    },
    "0010/gr-qc0010036_arXiv.txt": {
        "abstract": "We formulate a new method to calculate the gravitational reaction force on a particle of mass $\\mu$ orbiting a massive black hole of mass $M$. In this formalism, the tail part of the retarded Green function, which is responsible for the reaction force, is calculated at the level of the Teukolsky equation. Our method naturally allows a systematic post-Minkowskian (PM) expansion of the tail part at short distances. As a first step, we consider the case of a Schwarzschild black hole and explicitly calculate the first post-Newtonian (1PN) tail part of the Green function. There are, however, a couple of issues to be resolved before explicitly evaluating the reaction force by applying the present method. We discuss possible resolutions of these issues. ",
        "introduction": "In the black hole perturbation approach to gravitational radiation from an inspiralling compact binary, one calculates the gravitational waves emitted by a point particle orbiting a black hole, assuming the mass of the particle ($\\mu$) is much less than the mass of the black hole ($M$). At the lowest order in $\\mu/M$, the particle moves along a geodesic trajectory in the black hole geometry. Already at this lowest order, this approach has been proved to be very powerful for evaluating general relativistic corrections to the gravitational wave forms from a compact binary, even for neutron star-neutron star (NS-NS) binaries\\cite{MSSTT}. However, in such calculations one inevitably has to assume that the radiation reaction time scale is sufficiently long compared to the orbital period, so that the adiabatic approximation to the orbital evolution is valid. That is, the orbital evolution is determined by the energy and angular momentum balance between the orbit and the graviational radiation.  However, there are several situations in which such an approximation cannot be justified. For example, when the particle is nearly in the last stable circular orbit, the radiation reaction time scale becomes comparable to the orbital period. When the particle is in a very eccentric orbit or in a scattering orbit, presumably one has to determine the orbital evolution not by averaging over many periods but by each scattering event.  Another case is when the massive object is in fact a rotating black hole. In this case the geometry is Kerr, and the geodesic orbit is determined not only by the energy and the $z$-component of the angular momentum but also by the Carter constant\\cite{Carter}. Since the Carter constant is not associated with the Killing vector field of the Kerr geometry, its evolution cannot be determined by evaluating the gravitational radiation at infinity, even though the adiabatic approximation may still be valid.  In order to treat these situations, it is therefore necessary to evaluate the correction to the equation of motion of a particle in the background geometry. That is, we must know the reaction force of $O(\\mu/M)$ acting on the particle, \\begin{eqnarray} {D^2z^\\mu(\\tau)\\over d\\tau^2}=F^\\mu(z)\\,. \\label{eq:for} \\end{eqnarray} Formally $F^\\mu$ is given by \\cite{MiSaTa,QuiWal} \\begin{eqnarray} F^\\mu=-{1\\over2}(g^{\\mu\\nu}+u^\\mu u^\\nu) \\left( 2h^{\\rm tail}_{\\nu\\beta;\\alpha}-h^{\\rm tail}_{\\alpha\\beta;\\nu} \\right)u^\\alpha u^\\beta\\,, \\label{eq:force} \\end{eqnarray} where $u^\\mu=dz^\\mu/d\\tau$ and $h^{\\rm tail}_{\\mu\\nu}$ is the tail part of the linear perturbation caused by the particle, i.e., the contribution from a part of the retarded Green function that has support inside the past lightcone. Thus our task is to evaluate the tail part of the Green function\\footnote{In this paper, we refer to the `retarded Green function' simply as the `Green function'.}.  However, since the tail part depends non-locally on the background geometry, it seems practically impossible to evaluate it directly. As a way to avoid this difficulty, Mino proposed a very interesting approach\\cite{Mino}. He noticed that the direct part of the Green function, which has support only on the lightcone, is determined by the local geometry. Hence the corresponding metric perturbation can be evaluated solely with knowledge of the local geometry. On the other hand, in the black hole background, the linear perturbation is known to be described by the Teukolsky formalism\\cite{Teuk}, and there exists a method to obtain the general solution in a systematic manner\\cite{MaSuTa}. Then, using the transformation obtained by Chrzanowski\\cite{Chrz}, one can construct the corresponding metric perturbation. Hence, on the one hand, one may construct the full metric perturbation by solving the Teukolsky equation, and on the other hand, one can calculate the direct part of the metric perturbation by analyzing the local geometry. One can then obtain the tail part by subtracting the direct part from the full metric perturbation.  This approach was further advanced by Mino and Nakano\\cite{MinNak}, and they succeeded in obtaining some interesting results. Yet, their method seems rather cumbersome. It involves many steps of lengthy calculations, mainly resulting from the different choice of gauges in which the full metric perturbation and its direct part are evaluated.  In this paper, we propose a new, possibly more practical method to evaluate the tail part of the perturbation. As noted by Mino \\cite{Mino}, what one needs to do is to extract the direct part of the Green function. But this does not have to be done on the level of the metric perturbation. Instead, if one considers the Teukolsky equation in the spacetime domain, one can evaluate the direct part of the Green function for the Teukolsky equation directly, without transforming it to the metric perturbation. By Fourier-harmonic expanding the direct part obtained in this manner, one can then systematically evaluate the tail part. Then the corresponding metric perturbation and the reaction force $F^\\mu$ can be calculated by using the Chrzanowski transformation\\cite{Chrz}.  A technical problem in this approach is that the Teukolsky equation has a very non-trivial angular spin weight dependence, even in the limit of flat spacetime, $M\\to0$. Thus the evaluation of the direct part becomes very involved. As a way to avoid this problem, we consider the transformation of the Teukolsky equation with spin weight $s$ to a Klein-Gordon-type equation (i.e., with spin weight $s=0$). In the case of the Schwarzschild background, this is essentially the Chandrasekhar transformation that transforms the Teukolsky equation into the Regge-Wheeler equation\\cite{Chantrans}. In the case of the Kerr background, there is no known transformation of this kind. One possibility is to use the transformation given by Sasaki and Nakamura \\cite{SasNak} in a perturbative manner, with respect to the spin parameter $a/M$ of the Kerr geometry. In this paper, however, we focus on the case of the Schwarzschild black hole background and defer the extension to the Kerr case to a future work.  There are also a couple of issues to be resolved before applying the present method to the actual calculation of the reaction force. First, the Teukolsky equation, by its nature, does not give the $\\ell=0$ and 1 spherical harmonic components of the perturbation. Hence the tail part of these components must be calculated separately. Second, and most importantly, since the reaction force $F^\\mu$ is a gauge-dependent notion, its physical effect can be clarified only after we obtain the $O(\\mu)$ correction to the gravitational radiation emitted to infinity, but what we have at hand is only the linear-order metric perturbation. Thus, strictly speaking, there is no point in evaluating the reaction force unless we have a method to evaluate the corresponding correction to the gravitational radiation.  There have been other proposals for the regularization of the self force. Barack and Ori\\cite{BarOri} proposed a mode-sum regularization scheme. The calculation of the reaction force in their approach has two steps. First, each Fourier-harmonic mode of the bare self force is calculated. Then the sum over all modes is made subject to certain regularization procedure, which requires several regularization parameters. The calculation of the bare force on a scalar charge is done in \\cite{Bur1,Bur2,BarBur}. The analytical determination of the regularization parameters is carried out using a local perturbative analysis in \\cite{BarOri,Bar}. A method similar to that in \\cite{BarOri} was proposed independently by Lousto\\cite{Lousto}. Very recently, using the mode-sum regularization prescription, the regularization of the scalar and electromagnetic self force in the spacetime of spherical shells has been discussed by Burko, Liu and Soen\\cite{BuLiSo}.  This paper is organized as follows. In Sec.~\\ref{sec:RWGF}, we consider the Chandrasekhar transformation in the spacetime domain and discuss the nature of the direct part and the tail part of the Regge-Wheeler Green function. Then we give the Fourier-harmonic expanded form of the Teukolsky and Regge-Wheeler Green functions. In Sec.~\\ref{sec:1PMtail}, we formulate a method to calculate the tail part of the Regge-Wheeler Green function. Then, the tail part correct to the first post-Minkowskian (1PM) order is given under the short distance approximation, and an explicit expression of the tail in the first post-Newtonian (1PN) order is given. Then in Sec.~\\ref{sec:RV}, the procedure to obtain the regularized reaction force is discussed. Finally, in Sec.~\\ref{sec:discussion}, we summarize the results and discuss a possible way to resolve the remaining issues mentioned above. ",
        "conclusions": "\\label{sec:discussion}  In this paper we have proposed a new method to derive the gravitational reaction force on a point particle in the Schwarzschild background. In this method, the tail part of the metric perturbation, which is responsible for the reaction force, is obtained by first considering the Fourier-harmonic expansion of the Green function for the Regge-Wheeler equation. After calculating the tail part of the Green function, the result is transformed to the metric perturbation through the Chandrasekhar transformation followed by the Chrzanowski transformation. Since the extraction of the tail part is done at the level of the Green function in the Fourier-harmonic expanded form, we call our method the mode-by-mode regularization of the Green function. An advantage of our method is that it is relatively easy to obtain the tail part of the Green function for the Regge-Wheeler equation if we adopt the post-Minkowskian (or post-Newtonian) expansion.  However, there are still a couple of issues to be resolved. One is that the $\\ell=0$ and 1 components are not included in the metric perturbation obtained with the Chrzanowski transformation. Hence the tail part in them must be derived by some other method. The $\\ell=0$ and 1 components of the full metric perturbation can be solved exactly in the Regge-Wheeler-Zerilli gauge, as shown by Zerilli\\cite{Zer}. Therefore, what we need is a method to identify the tail part in the $\\ell=0$ and 1 components. The other, most crucial issue is that the physical meaning of the resulting reaction force remains unclear, since the reaction force is a gauge-dependent notion. For an ultimate resolution of this essential problem, we need to develop a second-order perturbation theory so as to make it possible to calculate the $O(\\mu)$ correction to the gravitational radiation.  Nevertheless, at low post-Newtonian orders, there is a way to resolve these issues. The equations of motion for a two-body system and the gravitational radiation from such a system are investigated extensively in \\cite{BIWW,WilWis,BFP,BlaFay} in the harmonic gauge. Hence, by performing the gauge transformation to the harmonic gauge, we should be able to clarify the physical effect of our result, as well as to examine the consistency of our result with those of these previous works. As a first step, we are currently working on a consistency check at the 1PN order\\cite{NakSas}.   \\noindent"
    },
    "0010/astro-ph0010301_arXiv.txt": {
        "abstract": "The energy range of IBIS is a promising ground for testing mutual relations of distinct components expected in the spectra of high-energy radiation from rotation powered pulsars. According to some polar-cap models two such components - due to curvature and synchrotron emission - may contribute comparable amounts of power between $15$~keV and $10$~MeV (Rudak \\& Dyks, 1999). Zhang \\& Harding (2000) argued recently for the inclusion of a third possible component, due to inverse Compton scattering (ICS) of soft thermal photons on secondary $\\epm$-pairs. Here we present the results of Monte Carlo calculations of all three spectral components within a polar-cap model which allows for interactions of relativistic particles with the soft photons coming from the pulsar surface. For teragauss pulsars with the surface temperature of a few$\\times 10^5$~K the ICS component dominates the spectrum in the energy range below $10$~MeV, and thus its presence increases the ratio of X-ray to $\\gamma$-ray luminosity (in comparison to the models ignoring the ICS on secondary $\\epm$-pairs) to a level observed in the Vela pulsar. ",
        "introduction": "The standard polar cap model of pulsars refers to the pair-photon cascades induced by photons of curvature radiation (CR) due to ultrarelativistic electrons leaving the neutron star (NS) surface (primary particles). Although successful in reproducing gamma-ray spectra observed above $100$~MeV \\citep{dh82} the model fails to reproduce the relative level of X-ray and gamma-ray emission. As shown by \\citet{zh00} this might be caused by neglecting the resonant inverse Compton scattering (ICS) of thermal photons from the surface by the secondary $\\epm$ pairs. The authors developed an approximate analytical method to model the X-ray luminosities (with the ICS contribution included) of pulsars detected by ROSAT and ASCA. Their results indicate that in terms of power in the X-ray range the ICS component may be competitive to both the curvature and synchrotron components. However, in order to verify the conclusion of \\citet{zh00}, and to obtain detailed spectra the Monte Carlo simulations are necessary.  In this work we present basic features of such Monte Carlo spectra for different pulsar parameters. In particular, the role and properties of the spectral component due to resonant inverse Compton scattering on secondary $\\epm$-pairs are addressed. We find that the shape and the strength of this component are sensitive to the strength of the local magnetic field as well as to the size and temperature of the thermal source. ",
        "conclusions": ""
    },
    "0010/astro-ph0010137_arXiv.txt": {
        "abstract": "A simple model of how objects of different masses stream towards each other as they cluster gravitationally is described. The model shows how the mean streaming velocity of dark matter particles is related to the motions of the parent dark matter haloes. It also provides a reasonably accurate description of how the pairwise velocity dispersion of dark matter particles differs from that of the parent haloes. The analysis is then extended to describe the streaming motions of galaxies.  This shows explicitly that the streaming motions measured in a given galaxy sample depend on how the sample was selected, and shows how to account for this dependence on sample selection.  In addition, we show that the pairwise dispersion should also depend on sample type.  Our model predicts that, on small scales, redshift space distortions should affect red galaxies more strongly than blue. ",
        "introduction": "Gravity makes objects cluster.  Therefore, the motions of objects towards each other may provide information about the background cosmology.  Of course, different subsets of the clustering particles may trace the underlying streaming motions differently. The scale dependence of the mean streaming $v_{12}^{\\rm dm}(r)$ of dark matter particles has been understood for some time now (Hamilton et al. 1991; Nityananda \\& Padmanabhan 1994). But there has been little study of how this statistic depends on trace-particle type.  To do this, we build a model in which gravitational clustering is viewed as the combination of two processes.  The first arises from the fact that gravity causes matter to stream towards local minima of the gravitational potential.  This requires a model of how matter which was initially distributed rather smoothly around the centre of collapse becomes redistributed into a more centrally concentrated density profile as the collapse proceeds.  The second process is that these centres around which local collapses are occurring, these clusters, are themselves moving towards each other:  clusters cluster.  It is the combination of these two types of motions which gives rise to the spatial distribution and streaming motions of objects today.  Section~\\ref{v12old} summarizes useful results which follow from linear theory.  Section~\\ref{2halo} shows how the streaming motions of collapsed dark matter haloes depend on halo mass. Section~\\ref{dm} uses this to model the streaming motions of particles, rather than haloes.  It shows what fraction of a particle's streaming motion arises from the motion of its parent halo, and what fraction must arise from motions within the halo.  These smaller scale motions are essentially a consequence of the collapse around the halo centre we referred to earlier.  It then presents measurements from numerical simulations which show that the model predictions are reasonably accurate. It also shows that the model provides a reasonable description of how the second moment of the pairwise velocity distribution of the dark matter differs from that of haloes.  Section~\\ref{gals} shows how to extend the model to study the mean streaming motions and the pairwise velocity dispersion of galaxies and presents measurements from semianalytic galaxy formation simulations which show that the model predictions are reasonably accurate. Section~\\ref{discuss} discusses what this model implies if one wishes to use measurements of the streaming motions of galaxies to make inferences about cosmology. ",
        "conclusions": "\\label{discuss} We presented a simple model of how the streaming motions of haloes and galaxies depends on separation.  We tested the model using the publically available simulations of Kauffmann et al. (1999). In the semi-analytic galaxy formation model, the mean streaming motions depend rather strongly on how the galaxy sample was selected. For example, blue galaxies have smaller streaming motions than red galaxies.  We showed that our model was able to describe the differences between a wide range of simulated galaxy catalogues rather well (Fig.~\\ref{gifgals}).  Our model predicts a very close relationship between the streaming motions of the galaxies and their spatial distribution. Optical and IRAS galaxies cluster differently (e.g. Marzke et al. 1995; Fisher et al. 1994).  Therefore, they must be biased differently relative to the dark matter. If the streaming motions of optical galaxies are the same as the dark matter, then our model predicts that the streaming motions of IRAS galaxies must be different from that of the dark matter.  In other words, whether or not the streaming motions of a given galaxy sample trace the motions of the underlying dark matter depends on how the sample was selected.  Thus, in the absence of strong arguments for why a given galaxy sample is expected to be unbiased, one should be cautious when interpretting measurements of the streaming motions of galaxies.  If the correlation function and the streaming motions of two different galaxy samples have been measured, then the model described here (equations~\\ref{v12gal} and~\\ref{v12twog}) says that the square root of the ratio of the two correlation functions (at, say, 20 Mpc/$h$) should equal the ratio of the streaming motions on the same scale.  This can be used to test the validity of the model. Again, however, caution is required because this relationship is only true on large scales.  For the semi-analytic galaxy samples we presented, this simple linear biasing was a good approximation on scales larger than about 10 Mpc$/h$ (although our model is able to describe the scale dependence of this ratio even on smaller scales).  In this context, we think it worth noting that the semi-analytic galaxy samples we presented here cannot explain the results of Juszkiewicz et al. (2000) who found that ellipticals and spirals have the same value of $v_{12}$ on separations of about 10 Mpc/$h$, even though they estimate that the spatial distributions have bias factors which differ by a factor of two:  $b_E/b_S\\approx 2$. It will be interesting if future data sets confirm this.  In addition to studying how the mean streaming velocity depends on galaxy type, our Fig.~\\ref{gifgals} also shows that the second moment of the pairwise velocity distribution depends strongly on galaxy type. For example, our model of the pairwise dispersion suggests that the dispersion of blue galaxies should be substantially smaller than that of red ones (although this difference depends on the colour cut), especially on scales of 1 Mpc$/h$ or so.  Thus, it is not surprising that $\\sigma_{12}$ for optical galaxies (Marzke et al. 1995) is almost a factor of two larger than for IRAS galaxies (Fisher et al. 1994). The effects of redshift space distortions are larger if the pairwise dispersion is larger.  This means that, on small scales, the amplitude of the redshift space correlation function of red galaxies should be substantially smaller than the real space correlation function, but the difference between real and redshift space correlation functions of blue galaxies should not be as dramatic. This is a generic prediction of these sorts of galaxy formation models. This suggests that galaxy redshift samples cut by colour should provide a useful and direct test of these models.  Dividing a galaxy sample by colour allows another simple test of these models.  At large separations, where most pairs are in separate haloes, the model described above predicts that the cross-correlation function of the two colour samples should simply be the geometric mean of the two individual samples. If the two galaxy samples both trace their parent dark matter haloes in the same way (our Fig.~\\ref{xigals} shows that this assumption describes the semianalytic model well), then this will be approximately true even on smaller scales.  (It will not be exactly true because the scatter in the $N_{\\rm gal}(m)$ relations are, typically, sub-Poisson.) Data sets currently available should be able to test this prediction."
    },
    "0010/astro-ph0010446_arXiv.txt": {
        "abstract": "New CCD observations of the pre-cataclysmic binary HW~Virginis are presented and discussed. The R-filtered CCD photometry was supplemented with medium-band (Cousins VRI) spectrophotometry based on low-resolution objective-prism spectra. The period variation is reanalysed by means of the standard O--C technique. The new data support the conclusions of Kilkenny et al. (2000) on the strong and continuous period change. The long-term period variation can be described approximately with two linear branches in the O--C diagram corresponding to a sudden period jump between two constant periods around JD 2448500 (1991). Additional smooth small-scale changes in the period distort the linearity. The present data do not support the hypothetical light-time effect of \\c Cakirli \\& Devlen (1999).  The eclipse depths in V, R, and I do not show colour dependence, which suggests negligible continuum variations due to the cool secondary component in the far-red region (up to 8800 \\AA). The magnitude of the reflection effect was used to estimate the mean effective temperature of the illuminated hemisphere of the secondary. The result is 13300$\\pm$200 K or 11000$\\pm$200 K, depending on the primary temperature (35000 K or 26000 K). An albedo near unity is implied for the cool component. ",
        "introduction": "The eclipsing binary nature of HW~Vir (BD$-07^\\circ$3477, V$_{\\rm max}$=10\\fm5, P=0.1167 days) was discovered by Menzies (1986). Before the discovery it was identified as a UV-bright object (Carnochan \\& Wilson 1983). Both primary and secondary minima are observed, and the lightcurve exhibits a striking reflection effect with an amplitude about 0\\fm2. Different lightcurve solutions were published (Menzies \\& Marang 1986, Wood et al. 1993, W\\l odarczyk \\& Olszewski 1994, \\c Cakirli \\& Devlen 1999). Although there are some discrepancies between the models presented, the general appearence is well-defined: the primary is a bright and evolved sdB star being overluminous compared to the low luminosity and cool secondary star, which is most probably an M-type main-sequence star. The temperature estimates range between 26000--36000 K and 3200--3700 K for the primary and secondary, respectively. The large brightness difference has prevented direct spectroscopic detection of the cool component, while Wood \\& Saffer (1999) reported extra H$\\alpha$ absorption features around the maximal reflection effect, which were associated with the illuminated and heated secondary atmosphere. The relatively large uncertainty of the primary temperature is reflected in the published distance estimates ranging from 42-151 pc (Wood et al. 1993), 125 pc (W\\l odarczyk \\& Olszewski 1994), 210 pc (calculated from parameters in \\c Cakirli \\& Devlen 1999), 171$\\pm$19 pc (Wood \\& Saffer 1999) and 179 pc (also derived from parameters in Hilditch et al. 1996). In contrast to these values, direct astrometric measurements by the Hipparcos satellite resulted in a parallax of 1.8$\\pm$1.9 mas (ESA 1997), i.e. the star {\\it may} be at a much larger distance than was thought earlier.  The first note on the strong period decrease in HW~Vir was published by Kilkenny et al. (1994). They discussed the period change over a 9-year baseline and pointed out possible reasons. They suggested angular momentum loss via magnetic braking in a modest stellar wind to be the most likely possibility for the period decrease, though light-time effects due to orbital motion around a third body were not ruled out. This latter approach was revised by \\c Cakirli \\& Devlen (1999), who presented a light-time effect solution of the O$-$C diagram, though the observations covered only 69\\% of the suggested orbital period. This has been the initiator of our photometric observations, because we wanted to check the recent period change. Very recently, Kilkenny et al. (2000) analysed an updated O$-$C diagram concluding that even a 6th order polynomial fit is not perfect.  The main aim of this paper is to present an analysis of new photometric and low-resolution spectroscopic observations of HW~Vir carried out in May, 2000. The paper is organised as follows: Sect.\\ 2 deals with the observations and data reductions. The period analysis is presented in Sect.\\ 3, while several simple considerations based on the multicolour spectrophotometry are given in Sect.\\ 4. A brief summary of the presented results is listed in Sect.\\ 5. ",
        "conclusions": "The new results presented in this paper can be summarized as follows:  \\noindent 1. New photometric observations are presented revealing the recent period change of HW~Vir. The O$-$C diagram covering 17 years does not support earlier suggestions of a possible light-time effect, but can be very well described by two linear branches. There are also additional smooth period changes.  \\noindent 2. Low-resolution objective-prism spectra were obtained and standard VRI photometry was done by convolving the calibrated spectra with the filter transmission functions. Our I-band lightcurve is the first in the literature. We have tried to detect spectral features that can be associated with the cool secondary component (e.g. variations in the infrared continuum), but this attempt failed. The depth of the primary minimum is the same in all three bands suggesting that the secondary is totally invisible even in the I band.  \\noindent 3. We present a simple estimate for the mean effective temperature of the illuminated hemisphere of the secondary based on the V, R, and I magnitude of the reflection effect. Their values (0\\fm18, 0\\fm20 and 0\\fm22) give a 13300$\\pm$200 K or 11000$\\pm$200 K temperature depending on the adopted primary temperature, which implies a near unity albedo for the cool component. This is also supported by an approximate lightcurve model."
    },
    "0010/astro-ph0010393_arXiv.txt": {
        "abstract": "We present {\\it Submillimeter Wave Astronomy Satellite} (\\swas) observations of the $1_{10}\\rightarrow1_{01}$ transition of ortho-\\water\\ at 557 GHz toward 12 molecular cloud cores.  The water emission was detected in NGC~7538, $\\rho$ Oph A, NGC~2024, CRL~2591, W3, W3(OH), Mon R2, and W33, and was not detected in TMC-1, L134N, and B335.  We also present a small map of the \\water\\ emission in S140.  Observations of the \\wateriso\\ line were obtained toward S140 and NGC~7538, but no emission was detected.  The abundance of ortho-\\water\\ relative to \\hh\\ in the giant molecular cloud cores was found to vary between $6\\times10^{-10}$ and $1\\times10^{-8}$.  Five of the cloud cores in our sample have previous \\water\\ detections; however, in all cases the emission is thought to arise from hot cores with small angular extents.  The \\water\\ abundance estimated for the hot core gas is at least 100 times larger than in the gas probed by \\swas.  The most stringent upper limit on the ortho-\\water\\ abundance in dark clouds is provided in TMC-1, where the 3$\\sigma$ upper limit on the ortho-\\water\\ fractional abundance is $7\\times10^{-8}$. ",
        "introduction": "The {\\it Submillimeter Wave Astronomy Satellite} (\\swas) has detected spatially extended \\water\\ emission associated with the quiescent dense molecular gas in Orion and M17 \\citep{sne00a,sne00b}.  Based on these observations the relative abundance of ortho-\\water\\ was found to be surprisingly small, ranging from only $1\\times10^{-9}$ to $8\\times10^{-8}$.  We present here \\swas\\ observations of the ortho-\\water\\ transition at 557 GHz in the three dark cloud cores TMC-1, L134N and B335 and in the nine giant molecular cloud cores W3, W3(OH), Mon R2, $\\rho$ Oph A, W33, CRL~2591, S140, and NGC~7538.  Five of the cloud cores in our survey have had previously reported water detections.  NGC~7538, S140, W3, and CRL~2591 were detected by the {\\it Infrared Space Observatory} in the ro-vibrational lines of \\water\\ from the $\\nu_2$ bending mode at 6 $\\mu$m in absorption against the dust continuum \\citep{van96,hel96,van98}.  Water emission has also been detected in NGC~7538, W3, and W3(OH) from the $3_{13}\\rightarrow2_{20}$ transition of \\water\\ at 183 GHz \\citep{cer90} and the $3_{13}\\rightarrow2_{20}$ transition of \\wateriso\\ at 203 GHz \\citep*{gen96}. The 183 GHz and 203 GHz lines arise from transitions with relatively high energies ($E_u/k \\ga 200$ K) and thus are likely produced in relatively warm gas.  The 6 $\\mu$m absorption features can arise from a range of gas temperatures, however the absorption seen in these sources is thought to be dominated by relatively warm gas. The relative abundance of \\water\\ in the warm gas has been estimated to be $1\\times10^{-6}$ to $1\\times10^{-4}$.  \\swas, on the other hand, can observe the lowest energy rotational transition of ortho-\\water\\ permitting detection of \\water\\ emission from the more extended, cold core gas.  We estimate the abundance of ortho-water in the relatively cold ($T < 50$ K) core gas and compare the abundance results with those for the warm gas and those for Orion and M17. ",
        "conclusions": "\\swas\\ has detected thermal water emission from nine giant molecular cloud cores which permit us to make the first estimate of the water abundance in the cold, dense core gas.  The abundance of ortho-\\water\\ relative to \\hh\\ in these cloud cores varies from $6\\times10^{-10}$ to $1\\times10^{-8}$. The relative ortho-\\water\\ abundance in the majority of these cores is of order $1\\times10^{-9}$, similar to the abundances found for M17SW \\citep{sne00b}.  In S140 the relative abundance of ortho-\\water\\ is about ten times larger, closer to the average ortho-\\water\\ abundance for Orion \\citep{sne00a}.  In general the relative abundance of ortho-\\water\\ found in these cores is inconsistent with chemical equilibrium models of well-shielded regions of clouds \\citep*{ber98}, even considering the most extreme uncertainties that might arise in our modeling procedure. Possible explanations for this discrepancy are discussed in \\citet{ber00}.  Water has been previously detected from the hot gas in W3 IRS~5, W3(OH), CRL~2591, S140, and NGC~7538 IRS~1 via the 183 and 203 GHz emission lines and the ro-vibrational absorption lines at 6 $\\mu$m \\citep{van96,hel96,gen96,van98}.  It it worth noting that the \\swas\\ observation of W3 is centered about midway between IRS~4 and IRS~5, less than 1\\arcmin\\ from IRS~5; and the \\swas\\ observation of NGC~7538 is centered about 1\\arcmin\\ south of IRS~1.  In both cases the regions probed via the 6 $\\mu$m, 183 GHz, and 203 GHz transitions are well within the \\swas\\ beam.  The water emission and absorption in these cores is thought to arise from hot cores surrounding young, massive stars with angular extents of only a few arcseconds and gas temperatures $\\gtrsim$300 K \\citep{van98,van99}.  Abundance estimates for water in the hot cores range from $1\\times10^{-6}$ to $1\\times10^{-4}$, several orders of magnitude greater than the abundance of water in the surrounding core gas.  Even our limit on the \\water\\ abundance based on \\wateriso\\ in S140 rules out the possibility that the large \\water\\ abundance found for the warm gas applies to the general cloud core. At temperatures exceeding $\\backsim$200-400 K, found in shocks and hot cores, a series of neutral-neutral reactions rapidly converts nearly all of the gas phase oxygen into water \\citep{cha97,ber98}.  Thus, the substantial difference predicted by the gas-phase chemistry models for the water abundance in warm and cold gas is confirmed by our observations.  \\swas, with a large beam size and the ability to observe a low excitation line of water, is most sensitive to the water emission from the extended, colder dense gas.  However, if the water abundance in the cold gas is extremely small, then the hot cores may still contribute to the emission seen by \\swas.  If we assume that the brightness temperature of the 557 GHz emission from the hot core gas is 300 K (the same excitation temperature as implied by the absorption lines seen by {\\it Infrared Space Observatory}) and that the core has an angular size of at most 2\\arcsec, then the contribution to the observed \\swas\\ \\water\\ emission is negligible in most sources, except CRL~2591.  Although \\citet{tak99} suggest that the CRL~2591 hot core has an angular diameter much less than 1\\arcsec, the extremely weak emission detected in this source could have a significant hot-core contribution.  Further information on the temperature and angular size of the hot core is needed to evaluate this possibility. If a significant fraction of the water emission detected by \\swas\\ in CRL~2591 is attributed to the hot core, then the abundance of \\water\\ in the cold gas is even smaller than that quoted in Table 1.  Only upper limits to the ortho-\\water\\ abundance could be established for the dark cloud cores.  If the abundance of water in these cores is as small as that found for the giant molecular cloud cores, it will be difficult to detect water emission in quiescent dark cloud cores. The most stringent limit on the ortho-\\water\\ abundance is set in TMC-1, and in this core the 3$\\sigma$ upper limit on the \\water\\ abundance is just marginally consistent with equilibrium gas-phase chemistry predictions \\citep{ber00}.  B335 is the site of a well-studied class 0 protostar, while neither L134N nor TMC-1 are actively forming stars. Broad \\water\\ emission has been detected by \\swas\\ toward two other class 0 protostellar cores, L1157 and NGC~1333 IRAS~4 \\citep{neu00}.  The \\water\\ emission in these sources clearly arises from the molecular outflows associated with class 0 sources.  The relative abundance of ortho-\\water\\ in the outflow gas found by \\citet{neu00} is of order $1\\times10^{-6}$, considerably larger than the limit set in the two non-star forming cores, L134N and TMC-1.  \\citet{neu00} suggests that the enhanced abundance is due to shocks that convert a large fraction of the gas-phase oxygen into water.  B335 also has an associated molecular outflow, with a outflow mass between that of L1157 and NGC~1333 IRAS~4 \\citep{mor89,neu00}. The absence of \\water\\ emission in B335 is difficult to understand, unless either the density of the outflow gas is much lower than in L1157 or NGC~1333 IRAS~4, or the shock velocities are sufficiently small that \\water\\ formation by neutral-neutral reactions is inhibited."
    },
    "0010/astro-ph0010500_arXiv.txt": {
        "abstract": "{ Many models of globular cluster formation assume the presence of cold dense clouds in early universe. Here we re-examine the Fall \\& Rees (1985) model for formation of proto-globular cluster clouds (PGCCs) via thermal instabilities in a protogalactic halo. We first argue, based on the previous study of two-dimensional numerical simulations of thermally unstable clouds in a stratified halo of galaxy clusters by Real \\etal (1991), that under the protogalactic environments only nonlinear ($\\delta \\ga 1$) density inhomogeneities can condense into PGCCs without being disrupted by the buoyancy-driven dynamical instabilities. We then carry out numerical simulations of the collapse of overdense clouds in one-dimensional spherical geometry, including self-gravity and radiative cooling down to $T=10^4 \\K$. Since imprinting of Jeans mass at $10^4\\K$ is essential to this model, here we focus on the cases where external UV background radiation prevents the formation of \\Htwo molecules and so prevent the cloud from cooling below $10^4\\K$. The quantitative results from these simulations can be summarized as follows: 1) Perturbations smaller than $M_{\\rm min} \\sim (10^{5.6}\\Msun) (n_h/0.05\\cm3)^{-2}$ cool {\\it isobarically}, where $n_h$ is the unperturbed halo density, while perturbations larger than $M_{\\rm max}\\sim (10^8\\Msun) (n_h/0.05 \\cm3)^{-2}$ cool {\\it isochorically} and thermal instabilities do not operate. On the other hand, intermediate size perturbations ($M_{\\rm min}<M_{\\rm pgcc}<M_{\\rm max}$) are compressed {\\it supersonically}, accompanied by strong accretion shocks. 2) For supersonically collapsing clouds, the density compression factor after they cool to $T_c=10^4\\K$ range $10^{2.5}-10^6$, while the isobaric compression factor is only $10^{2.5}$. 3) Isobarically collapsed clouds ($M<M_{\\rm min}$) are too small to be gravitationally bound. For supersonically collapsing clouds, however, the Jeans mass can be reduced to as small as $10^{5.5} \\Msun (n_h/0.05 \\cm3)^{-1/2}$ at the maximum compression owing to the increased density compression. 4) The density profile of simulated PGCCs can be approximated by a constant core with a halo of $\\rho \\propto r^{-2}$ rather than a singular isothermal sphere.} ",
        "introduction": "Globular clusters (GCs) are the oldest relics in our galaxy, presenting a cosmological challenge, as their ages often burst the allowed bounds of the expansion age. Since there are {\\it at least} three distinct populations of GCs, the old halo and the middle-aged disk populations of GCs in our Galaxy (Zinn 1985), and the young GCs in interacting galaxies (Ashman \\& Zepf 1992), we may well need more than one model to explain their formation and evolution. In the present study, we focus on the formation of the halo GCs. Half of the total mass of our galaxy appears to be contained within $\\sim 50\\kpc$, whereas roughly half of the luminosity of the disk as well as half of the total number of halo GCs are distributed within the solar galactocentric distance ($\\sim 8\\kpc$). So halo GCs clearly know about the {\\it dissipation} that followed galaxy formation and we must place their formation in the protogalactic context. This encourages models that form halo clusters during the formation of the protogalaxy, such as thermal instabilities in the protogalactic halo (Fall \\& Rees 1985) or pregalactic cloud collisions (Gunn 1980; Lake 1987; Kang \\etal 1990; Kumai, Basu, \\& Fujimoto 1993).  Many models of globular cluster formation assume the presence of {\\it proto-globular cluster clouds} (PGCCs) in pressure equilibrium with hot halo gas (\\eg  Gunn 1980; Dopita \\& Smith 1986; Brown, Burkert \\& Truran 1991; Kumai, Batsu, \\& Fujimoto 1993). The model by Fall \\& Rees (1985, FR85 hereafter) has been most widely adopted to explain the formation of such PGCCs: the thermal instability drives catastrophic cooling of over-dense regions in the hot halos of protogalaxies and condenses them to dense clouds. In the present study, adopting FR85 model, we follow the thermal and dynamical evolution of over-dense clouds by numerical simulations in order to examine more quantitatively the physical properties of resulting PGCCs.  Although the thermal instability was extensively investigated in several areas of astrophysics, in recent years it has received much attention especially for X-ray cluster cooling flows: ({\\it cf.} Balbus 1986; Malagoli, Rosner \\& Bodo 1987; Balbus \\& Soker 1989; Malagoli, Rosner, \\& Fryxell 1990). According to the results from numerical simulations in one-dimensional (1D) geometries (David, Bregman \\& Seab 1988 [DBS88 hereafter]; Brinkmann, Massaglia \\& M\\\"uller 1990 [BMM90]), the over-density undergoes a {\\it quasi-static compression} in near pressure equilibrium when a cloud is small enough to adjust to pressure change faster than it cools, while larger clouds may undergo a ``supersonic compression stage'', leading to a density increase much higher than what is expected from the isobaric compression. The plane-parallel calculations by DBS88 showed that the density contrast increases to the ``isobaric\" ratio of the temperature of the hot background medium ($T_h\\sim 10^7-10^8\\K$) and the minimum temperature ($T_c\\sim 10^4\\K$) where the cooling becomes inefficient. On the other hand, the spherically symmetric calculations of BMM90 showed that the density contrasts can be more than three orders of magnitude higher than those found in the plane-parallel calculations of DBS88. The difference between two geometries is mainly due to the fact that the gravitational acceleration increases as $GM/R^2$ when the spherical cloud collapses, while it remains the same (independent of $R$) when the plane sheet compresses. Also isotropic spherical compression (focusing) leads to much higher density accelerating the cooling and rise in density. However, 2D simulations of the collapse of an elongated blob by BMM90 showed a shape instability to a pancake-shape leading to evolution best represented by the 1D plane-parallel case. Thus geometry is clearly important in the nonlinear evolution. On the other hand, 2D simulations by V\\'azquez-Semadeni, Gazol, \\& Scalo (2000) showed that the condensation of fluctuating density field via thermal instability leads to a network of transient filaments which later accrete onto isolated round blobs. These blobs form at intersections between filaments just like galaxy clusters in large-scale structure formation in cosmological simulations. This implies that the 1D spherical simulation can actually represent a generic collapse of the highest density peaks in the randomly fluctuating density field in 3D.  Numerical simulations in a 2D stratified background (Yoshida, Hattori, \\& Habe 1991; Reale \\etal 1991 and references therein), on the other hand, have shown that the dynamic instabilities, such as Rayleigh-Taylor and Kevin-Helmholtz instabilities due to buoyancy-driven oscillations can disrupt the thermally unstable bubbles if the timescale for buoyancy oscillation (Brunt-V\\\"as\\\"al\\\"a period) is shorter than the local radiative cooling time. Thus it is believed that the classic theory of thermal instability can not explain the effective mass depletion from the cooling flows of galactic clusters. We will examine the role of such dynamic instabilities for the formation of PGCCs in the protogalactic halo environment in the next section. The details of our models are given also in this section.  The main results are in \\S 3. Conclusions are given in \\S 4. ",
        "conclusions": "We re-examined Fall \\& Rees (1985) model for the formation of PGCCs via thermal instabilities by numerical simulations of overdense clouds in a protogalactic halo. The key idea of this model is that characteristic mass scale of GCs, $M_c\\sim 10^6 \\Msun$, can be explained by imprinting of critical (Jeans) mass of gas cloud at $10^4\\K$ which has cooled from a hot halo gas in pressure equilibrium via thermal instability. If there were significant cooling below $10^4\\K$ due to \\Htwo and metals, the cloud will continue to cool and the imprinting is not possible (Kang et al 1990). Then the model fails. So we considered the cases where the radiative cooling is ineffective below $10^4\\K$, because the formation of \\Htwo molecules is delayed due to UV radiation from a central AGN or diffuse radiation from halo gas.  The main conclusions are summarized as follows: \\begin{enumerate}  \\item{} Unlike X-ray cluster cooling flows, a protogalactic halo provides much more favorable conditions for thermal instability to operate. So PGCCs can form from marginally nonlinear (\\ie $\\delta \\gsim 1$) density inhomogeneities in a protogalactic halo via thermal instability. Linear perturbations are likely to be disrupted by the buoyancy-driven instabilities (Balbus \\& Soker 1989; Reale \\etal 1991).  \\item{} The size of thermally unstable perturbations is determined by the cooling distance, $l_{cool}$, over which a sound wave can travel in a cooling time. According to our numerical simulations in one-dimensional spherical geometry, the clouds of $0.2 l_{cool}\\la R_c \\la l_{cool}$ can condense to form PGCCs via thermal instability. Smaller clouds ($R_c \\la 0.2 l_{cool}$) cool only {\\it isobarically} and become gravitationally unbound clouds, while larger cloud ($R_c \\ga l_{cool}$) cool {\\it isochorically} and so do not condense to form any distinct structures.  \\item{} The thermally unstable clouds, which have mass $10^{5.5}\\Msun \\la M_{PGCC}(n_h/0.05\\cm3)^{2} \\la10^8\\Msun$, are compressed supersonically, and the density enhancement of $10^3-10^6$ results. This is much higher than what is expected from an isobaric compression. Considering the larger compression factor, we estimate the Jeans mass for PGCCs as $10^{5.5}\\Msun \\lsim M_J (n_h/0.05\\cm3)^{1/2} \\lsim 10^{7.2}\\Msun$. The density distribution of simulated PGGCs can be approximated by an isothermal distribution with a constant core.  \\end{enumerate}  Although we focus on the formation of old halo GCs in the Milky Way in this work, we note globular clusters exist in some extremely different systems. The largest of the dwarf spheroidals, Fornax, Sagittarius, NGC 147, NGC 185 and NGC 205 all possess GCs. The specific frequency of clusters in these dwarf systems is as large as in bright ellipticals and larger than in spirals (Hodge 1988). If gas were in hydrostatic equilibrium in these parent galaxies, the temperature would be just $10^4-10^5$K. Thus a different model other than thermal instability model for the formation of PGCCs or the formation of GCs themselves is necessary for these systems.  Many models for GC formation have been suggested so far, but none of them seem to be complete or can explain all the observed properties of GCs.  One of the aspects of FR85 model that has not been touched upon in the present work is the effect of the turbulent motions in the halo. Turbulent energy equivalent to the virial temperature should be injected at the largest scale and cascade down to smaller scales. According to 2D simulations by V\\'azquez-Semadeni, Gazol, \\& Scalo (2000) the turbulence cannot suppress the thermal instability, because the time scale for the thermal instability remains shorter than the turbulent crossing time scale ($ R_{\\rm cloud}/ v_{\\rm turb}$) at smaller scales than the virialization-scale. This is because the turbulent velocity decreases faster than the scale, so the turbulent crossing time increases at smaller scales. On the other hand, the turbulence can provide some heating to the halo gas. Since the halo gas cools radiatively in about $10^8 \\yrs$, additional heating is necessary to maintain the halo temperature at $T_h$ for much longer than the cooling time scale in order to explain the observed age spread of GCs, typically a few Gyr.  As discussed in \\S2, it is necessary to consider more realistic multi-dimensional simulations of thermally unstable clouds in a protogalactic halo environment, which is currently under study. The present work, based on one-dimensional considerations, however, provides useful guidance for the up-coming multi-dimensional works."
    },
    "0010/hep-ph0010203_arXiv.txt": {
        "abstract": "We discuss to which extent the present experiments of direct search for WIMPs, when interpreted in terms of relic neutralinos, probe interesting regions of the supersymmetric parameter space, which are also being progressively explored at accelerators.  Our analysis is performed in a number of different supersymmetric schemes. We derive the relevant neutralino cosmological properties, locally and on the average in the universe.  We prove that part of the susy configurations probed by current WIMP experiments entail relic neutralinos of cosmological interest.  The main astrophysical and particle physics uncertainties, relevant for a proper comparison between theory and experimental data, are stressed and taken into account. ",
        "introduction": "As was first noticed in Ref. \\cite{bdmsbi}, in the last few years the experiments of direct search for Weakly Interacting Massive Particles (WIMP) \\cite{morales} have already reached a sensitivity which allows the exploration of regions of the supersymmetric parameter space, which are also progressively investigated at accelerators.  This property is manifest, when the experimental results are interpreted in terms of relic neutralinos \\cite{bdmsbi}.  The probing of the susy parameter space by WIMP direct searches is even more sizeable at present, with the sensitivities of experiments \\cite{damalast,cdms}; a comparative discussion of the experimental features and implications of the DAMA \\cite{damalast} and CDMS \\cite{cdms} experiments may be found in Ref. \\cite{damapr}.  Detailed studies of the possible interpretation of the annual--modulation effect \\cite{damalast} in terms of relic neutralinos have been reported in \\cite{noi,comp,noi5,noiult}. Comparisons of the experimental data of Ref. \\cite{damalast} with susy calculations have also been performed in Refs. \\cite{an,kkk,efo,acc,gabr,el2}.  In the present paper we intend to clarify the actual capability of WIMP direct searches by exploring in a sistematic way different realizations of minimal supersymmetric models and showing their intrinsic differences in the prediction of neutralino rates and relic abundance.  Specifically, we will consider two different implementations of a supergravity scheme with parameters defined at the grand unification scale and an effective supersymmetric model defined at the electroweak scale.  Our analyses are performed in the light of the following relevant points: i) current uncertainties in astrophysical properties, ii) uncertainties in hadronic quantities, iii) new bounds from LEP searches for Higgs and supersymmetric particles, iv) updated determinations of cosmological parameters.   Let us start by recalling that  the determination of the sensitivity range of an experiment of WIMP direct search {\\it in terms of the WIMP mass and of the WIMP--nucleon cross section} rests on a number of crucial assumptions, since it depends both on the distribution function of the WIMPs in the halo and on the nature of the relic particle.  A WIMP direct experiment provides a measurement (or an upper bound)  of the differential event rate  \\begin{equation} \\frac {dR}{dE_R}=N_{T}\\frac{\\rho_{W}}{m_{W}} \\int \\,d \\vec{v}\\,f(\\vec v)\\,v \\frac{d\\sigma}{dE_{R}}(v,E_{R}) \\label{eq:diffrate0} \\end{equation}  \\noindent where $N_T$ is the number of the target nuclei per unit of mass, $m_W$ is the WIMP mass, $\\rho_W$ is the local WIMP matter density, $\\vec v$ and $f(\\vec v)$ denote the WIMP velocity and velocity distribution function in the Earth frame ($v = |\\vec v|$) and $d\\sigma/dE_R$ is the WIMP--nucleus differential cross section.  The nuclear recoil energy is given by $E_R={{m_{\\rm red}^2}}v^2(1-\\cos \\theta^*)/{m_N}$, where $\\theta^*$ is the scattering angle in the WIMP--nucleus center--of--mass frame, $m_N$ is the nuclear mass and $m_{\\rm red}$ is the WIMP--nucleus reduced mass.  Eq.(\\ref{eq:diffrate0}) refers to the case of a monoatomic detector, like the Ge detectors. Its generalization to more general situations, like for instance the case of NaI, is straightforward. In what follows $\\rho_W$ will be factorized in terms of the local value for the total non--baryonic dark matter density $\\rho_l$ and of the fractional amount of density, $\\xi$, contributed by the candidate WIMP, {\\it i.e.} $\\rho_W = \\xi \\cdot \\rho_l$. For $\\rho_l$ we use the range 0.2 GeV cm$^{-3} \\leq \\rho_l \\leq$ 0.7 GeV cm$^{-3}$, where the upper side of the range takes into account the possibility that the matter density distribution is not spherical, but is described by an oblate spheroidal distribution \\cite{bt,t}.  The WIMP--nucleus differential cross section may conveniently be split into a coherent part and a spin--dependent one  \\begin{equation} \\frac {d\\sigma}{dE_R} = \\left(\\frac{d\\sigma}{d E_R}\\right)_C+ \\left(\\frac{d\\sigma}{d E_R}\\right)_{SD}, \\label{eq:diffrate_sum} \\end{equation}  \\noindent whose generic features are discussed in the seminal paper of Ref. \\cite{gw}. To compare theoretical expectations with experimental data, and experimental data of different detectors among themselves, it is useful to convert the WIMP--nucleus cross--section into a WIMP--nucleon cross section. This procedure is feasible independently of the nuclear model and of the specific nature of the WIMP  only under the hypothesis that the coherent cross--section is dominant and the WIMP couples equally  to protons and neutrons (at least approximately) \\cite{bdmsbi}. Under this assumption, the   WIMP--nucleus cross section may be expressed in terms of a WIMP--nucleon scalar cross section $\\sigma^{\\rm (nucleon)}_{\\rm scalar}$  as  \\begin{equation} \\frac {d\\sigma}{dE_R} \\simeq \\left ( \\frac{d\\sigma}{dE_R} \\right )_C \\simeq \\frac{F^2(q)}{E^{max}_R} \\left(\\frac{1+m_W/m_p}{1+m_W/m_N}\\right )^2 A^2 \\sigma^{\\rm (nucleon)}_{\\rm scalar}, \\label{eq:diffrate_approx} \\end{equation}  \\noindent where $m_p$ and $m_N$ are the proton and nucleus mass, $A$ is the nuclear mass number, $E_R^{max}$ is the maximal recoil energy and $F(q)$ is the nuclear form factor for coherent interactions. This form factor is  usually parametrized in the Helm form \\cite{helm}; however, precise evaluations of the event rates may require specific nuclear calculations for each target nucleus. In the rest of this paper we assume that the WIMP interaction with the nuclei of the detector is dominated by coherent effects, so that a WIMP--nucleon scalar cross section may be derived from the WIMP--nucleus cross section by use of Eq. (\\ref{eq:diffrate_approx}).  Now, coming back  to the general expression in Eq. (\\ref{eq:diffrate0}), we stress  that extracting an information about the WIMP--nucleus cross sections from the experimental data requires the use of a specific expression for the velocity  distribution function $f(\\vec{v})$ (notice that in writing Eq. (\\ref{eq:diffrate_approx}) we have already made the assumption that the WIMP phase--space distribution function may be factorized as $\\rho \\cdot f(\\vec{v})$, and this is certainly not the most general case \\cite{bt}). The usual choice for $f(\\vec{v})$ is the isotropic  Maxwell--Boltzmann distribution in the galactic rest frame, as derived from the isothermal-sphere model.  However, recent investigations have shown that deviations from this standard scheme, either due to a bulk rotation of the dark halo \\cite{kk,dfs} or to an asymmetry in the WIMP velocity distribution \\cite{vu,ecz,amg}, influence the determination of the WIMP--nucleus cross sections from the experimental data in a sizeable way.  In Ref. \\cite{ecz} also triaxial matter distributions are considered; in the present paper deviation from sphericity in the WIMP matter distributions are taken into account only through the physical range allowed for the value of $\\rho_l$ (see our previous comment on $\\rho_W$ after Eq. (\\ref{eq:diffrate0})).  In a typical plot, where the WIMP-nucleus cross section is given in terms of the WIMP mass, the effect introduced by the mentioned deviations from the Maxwell--Boltzmann distribution is generically to elongate the contours towards larger values of $m_W$. This is for instance the case for the the annual--modulation region of the DAMA Collaboration \\cite{damalast}. In Fig. 3 of Ref. \\cite{noi5} it is shown that, by implementing the dark halo with a bulk rotation according to the treatment in Ref. \\cite{dfs}, the annual--modulation region moves towards larger values of the WIMP mass, with an elongation which brings the right--hand extreme from the value of $\\, \\sim$ 150 GeV to $\\, \\sim$ 200 GeV.  A similar effect is obtained by introducing an asymmetry in the WIMP velocity distribution $f(\\vec{v})$: Fig. 4 of Ref. \\cite{amg} illustrates this point. Notice that this asymmetry effect also pushes somewhat downwards the annual--modulation region.  We emphasize that all these effects are extremely important, when experimental results of WIMP direct detection are being compared with theoretical models for specific candidates.  This point has been overlooked in most analyses in terms of relic neutralinos \\cite{caustics}.  In the present paper we focus  our analysis to the WIMP mass range which, in the light of the present experimental data \\cite{damalast,cdms} and of the previous considerations on the astrophysical uncertainties, appears particularly appealing:  \\begin{equation} 40 \\; {\\rm GeV} \\leq  m_W \\leq 200 \\;  {\\rm GeV}. \\label{eq:mass} \\end{equation}    Let us notice that the mass range of Eq. (\\ref{eq:mass}) is quite appropriate for neutralinos. Actually, the lower extreme is indicative of the LEP lower bound on the neutralino mass $m_{\\chi}$ (in the calculations performed in the present work the actual lower bound for $m_{\\chi}$, dependent on the other susy parameters, is employed, according to the constraints given in \\cite{LEPb}). As for the upper extreme, we notice that, though a generic  range for $m_{\\chi}$ might extend up to about 1 TeV, requirements of no excessive fine--tuning  \\cite{bere1} would actually  favour an upper bound of order 200 GeV, in accordance with Eq. (\\ref{eq:mass}).  In what follows we will discuss the discovery potential of WIMP direct searches for WIMPs in the mass range of Eq.(\\ref{eq:mass}). Particular attention will be paid to capabilities of the present experiments; their sensitivity range, in case of WIMPs whose coherent interactions with ordinary matter are dominant over the the spin--dependent ones, may be stated, in terms of the quantity $\\xi \\sigma^{\\rm (nucleon)}_{\\rm scalar}$, as \\cite{damalast,cdms}   \\begin{equation} 4 \\cdot 10^{-10} \\; {\\rm nbarn} \\leq \\ \\xi \\sigma^{\\rm (nucleon)}_{\\rm scalar} \\leq 2 \\cdot 10^{-8} \\; {\\rm nbarn}. \\label{eq:section} \\end{equation}  We will hereafter refer to region $R$ as the one in the space $m_W - \\xi \\sigma^{\\rm (nucleon)}_{\\rm scalar}$ which is  defined by Eqs. (\\ref{eq:mass}-\\ref{eq:section}). The region $R$ represents the sensitivity region already under exploration with present detectors.   Our analysis, based on an interpretation of experimental data in terms of relic neutralinos,  will show by how much the WIMP direct searches probe the supersymmetric parameter space. We remark  that, in the case of neutralinos, the assumption about the dominance  of the coherent cross section over the spin--dependent one is, in general, largely satisfied, except for values of $\\sigma^{\\rm (nucleon)}_{\\rm scalar}$ which are far below  the present   experimental reach \\cite{bdmsbi}.    The present analysis will be performed in the framework of various schemes, from those based on universal or non-universal supergravity, with susy parameters defined at the grand unification scale, to an effective supersymmetric model defined at the Electro--Weak (EW) scale. This is discussed in Sect. II, where we also specify the values employed here for the Higgs--quark--quark and the neutralino--quark--squark couplings.  These quantities are subject to sizeable uncertainties, as was stressed in Ref. \\cite{noi6}, which triggered a reconsideration of this important point in a number of subsequent papers \\cite{efo,acc,cn}.  The most important properties to be established for the relic neutralinos, which are entailed in the exploration by WIMP direct searches, concern their cosmological properties. Here, we perform a general analysis which is not limited to a restricted range of the cosmological matter abundance, $\\Omega_m h^2$ ($\\Omega_m$ is the matter cosmological density divided by the critical density and $h$ is the Hubble constant in units of 100 km s$^{-1}$ Mpc$^{-1}$) . Instead, we derive the average and local cosmological properties of the susy configurations from experimental determinations of $\\sigma^{\\rm (nucleon)}_{\\rm scalar}$, without any {\\it a priori} requested range on $\\Omega_m h^2$. On the basis of the results of our evaluations in the various supersymmetry models, we discuss when the relic neutralino does or does not saturate the expected amount of the local and of the average amount of {\\it total} dark matter.  Our results and conclusions are presented in Sects. III and IV, respectively. ",
        "conclusions": "In this work we have shown that the current direct experiments for WIMPs, when interpreted in terms of relic neutralinos, are indeed probing regions of the supersymmetric parameter space compatible with all present bounds from accelerators. We have quantified the extent of the exploration attainable by WIMP direct experiments in terms of various supersymmetric schemes, from a SUGRA scheme with unification assumptions at the grand unification scale to an effective model, effMSSM, at the electroweak scale. It has been stressed that, due the large uncertainties in the unification assumptions in SUGRA schemes, the effMSSM framework turns out to be the most convenient model for neutralino phenomenology.  We have proved that part of the configurations probed by WIMP experiments, and not disallowed by present accelerator bounds, entail relic neutralinos of cosmological interest. As discussed in the previous Section, this result is at variance with the conclusions of some analyses recently appeared in the literature.  Also neutralinos which might contribute only partially to the required amount of dark matter in the universe have been included in our analysis. The cosmological properties have been displayed in terms of a plot of the local density versus the average relic abundance, {\\it i.e.} in a representation which proves particularly useful to summarize the properties of relic neutralinos.  We have noticed that a Higgs with a mass of about 115 GeV, such as the one now under experimental investigation at LEP2, would fit remarkably well in the above scenario.  In our evaluations we have taken into account that the determination of the actual sensitivity region in terms of the WIMP--nucleon cross section and of the WIMP mass from the experimental data depends quite sizeably on uncertainties of various origins, mainly: i) possible effects due to a halo bulk rotation and/or to asymmetries in the WIMP velocities distribution, ii) significant uncertainties in the determination of Higgs--quark--quark and neutralino--quark--squark couplings. We have stressed that all these effects have to be taken properly into account, when conclusions about comparison of theory with experiments are drawn."
    },
    "0010/astro-ph0010050_arXiv.txt": {
        "abstract": "Two of the greatest mysteries of modern physics are the origin of the dark matter in the universe and the nature of the highest energy particles in the cosmic ray spectrum. We discuss here possible direct and indirect connections between these two problems, with particular attention to two cases: in the first we study the local clustering of possible sources of ultra-high energy cosmic rays (UHECRs) driven by the local dark matter overdensity. In the second case we study the possibility that UHECRs are directly generated by the decay of weakly unstable super heavy dark matter. ",
        "introduction": " ",
        "conclusions": ""
    },
    "0010/astro-ph0010099_arXiv.txt": {
        "abstract": "We report detection of an extended mid-infrared emission from \\iras 18576+0341 (AFGL 2298). The object shows a dusty circumstellar shell that has diameter of $\\age 7\\arcsec$ at 10.3 and 18.0 $\\um$. The dust nebula shows two emission peaks concentrically elongated and symmetrically oriented on the opposite sides of the third, central peak, which appears to be the central star of the system. The observed mid-infrared morphology indicates that the circumstellar dust shell has an equatorially-enhanced material distribution, which is a common signature of stellar objects that have experienced mass loss. Radiative transfer model calculations suggest that the central star is an extremely bright ($\\lstar = 10^{6.4} \\lsun$) star at a distance of about 10 kpc: this object is best described as a new luminous blue variable candidate. The circumstellar dust shell seems to have been generated by an equatorially-enhanced mass loss process with ${\\dot M} \\ge 6.8 \\times 10^{-6} \\msun$ yr$^{-1}$ and ${\\dot M}_{\\rm pole}/{\\dot M}_{\\rm eq} \\sim 0.5$. ",
        "introduction": "Mid-infrared (mid-IR; $8 - 25 \\um$) imaging provides powerful means to directly probe the distribution of circumstellar matter, especially circumstellar dust grains. High-resolution ($\\sim 1\\arcsec$) mid-IR images can reveal the structure at the innermost regions of the circumstellar dust shells (CDSs), because the emission is generally optically thin and arises from the warmest ($100 - 200$ K) dust grains in the CDSs. Such mid-IR images of the CDSs have shown that many stars have lost or have been losing their mass in an equatorially-enhanced manner, resulting in toroidal or ring shaped CDSs. For example, the axisymmetric shaping of planetary nebulae is considered to be initiated by an equatorially-enhanced dust-driven wind mass loss at the end of the asymptotic giant branch phase (e.g., \\citealt{ueta00} and references therein). The axisymmetric morphologies of the nebulae around luminous blue variables (LBVs) are explained similarly by the presence of an equatorially-enhanced CDS \\citep{nota95}, which may be generated by stellar rotation near the Eddington limit \\citep{langer99} or the tidal action of the central binary system \\citep{damineli00}.  In this paper, we report our discovery of an extended dust shell around an unidentified IR object, \\iras 18576+0341 (\\iras 18576, hereafter), during our observing run to obtain high-resolution mid-IR images of CDSs around evolved stars. After reviewing previous observations in \\S 2, we present the observed images in \\S 3. The results of the dust radiative transfer model calculations are presented in \\S 4, followed by a discussion on the nature of the source in \\S 5. Finally, our conclusion is presented in \\S 6. ",
        "conclusions": "We have discovered an extended mid-IR emission around \\iras 18576 and presented the mid-IR images at 10.3 and 18.0 $\\um$. The two emission peaks, which are concentrically elongated and symmetrically oriented on the opposite sides of the central peak, are typically the results from an optically thin, limb-brightened edges of the dust torus viewed rather edge-on. We also present new near-IR photometry of the object which demonstrates its variability. Although the source lies in the region of the sky in which a severe Galactic extinction has been observed, we have constructed a model using all of the available information. The derived physical quantities for the central star and dust shell and variability of the object suggest that \\iras 18576 is a possible LBV candidate with $\\lstar = 10^{6.4} \\lsun$ and $\\teff = 15000 \\pm 9000$ K. The uncertainties involved in the distance determination leave possibilities that the object could be a RSG, and further observational work is needed to determine the exact evolutionary state of the star. While the main dust shell is composed of silicate dust grains, an additional \\crich (composed of PAHs) shell likely exists inside of the \\orich shell. The axisymmetric nature of the dust toroid is characterized by the calculated pole-to-equator density ratio of 0.5. While the origin of the equatorial enhancement is still open for further discussion, the axisymmetry seems to be a common feature in mass loss from massive stars as in the case for low to intermediate stars."
    },
    "0010/astro-ph0010266_arXiv.txt": {
        "abstract": "s{ We are undertaking a large, wide-field survey of nearby groups and poor clusters. The main goals are identifying and characterizing the properties of low-surface-brightness dwarf galaxies and determining the galaxy luminosity function for M$_R > -17$. Large areas (typically 0.3--0.5 degree$^2$ per system but up to 7 degree$^2$) of the groups Telescopium, Leo I, Dorado, N5575, HCG 42, HCG 44,  HCG 68 and the poor clusters IC 4765 and Hydra have so far been surveyed in V and I. We present the preliminary results for the photometric study of the groups HCG 42, Telescopium (or NGC6868) and IC4765.  Hundreds of new low surface brightness galaxies are catalogued. Their spatial distributions, colors, types and sizes will be studied as a function of the richness of their environments.} ",
        "introduction": "Dwarf galaxies are the most common type of galaxies in the local universe. In addition, they are thought to be the single systems with the largest dark-matter content, with M/L ratios as high as that of groups and poor clusters \\cite{car88}. Hence their spatial distribution and mass spectrum may give us important insight into the spatial scales over which mass is distributed in the universe.  However, they are also the hardest galaxies to observe, due to their low-surface brightnesses and low luminosities.  Only recently it has become possible to obtain deep, wide-field CCD {\\it photometry} of significant numbers of these galaxies, mostly in clusters. {\\it Spectroscopy} of such low luminosity systems is, however, still a challenge even with large-class telescopes unless emission lines are present.  Dwarf galaxies in nearby clusters and groups have been studied in detail with photographic material \\cite{tho93,san85,imp88,fer89,bot91,fer91}. Only recently, with the advent of large-format CCDs, have these searches become possible in wide-field CCD images of nearby groups and clusters \\cite{one97,jer00}. From these works we learned that the dwarf elliptical (dE/dS0) galaxies dominate the faint-end of the number counts in clusters while the dwarf-irregular population is the more numerous in poor groups and in the field. However, our own Local Group seems to be an exception, with a larger number of dwarf spheroidals than irregular galaxies.  The main goal of our project is to identify the population of low-surface-brightness (LSB) dwarf galaxies in nearby groups to M$_V = -10$ mag and determine the luminosity function in each of the groups. The selected sample is formed by groups with different environments, poor groups like Dorado and NGC 6868 and rich groups like Hydra I and IC 4765 with galaxy populations of different sizes and morphological mixtures. The program was started with the observation of the Dorado group \\cite{car01}, at $cs\\sim1200$ km/s, and  continued with the observation of other seven nearby groups with  $1000<cz<4500$ km/s. The aim of this paper is to present preliminary  results of the study of the dwarf galaxy population in three of the groups observed, HCG42, IC4765 and NGC6868. ",
        "conclusions": "More than a hundred LSB dwarf galaxy candidates with $\\mu_{0}\\ge22$ V mag/arcsec$^{2}$, $h\\ge1.3$'', $\\theta_{lim}\\ge3$'' and with $0.5\\le(V-I)\\le1.7$ were found in the groups HCG42, IC4765 and NGC6868. Our add-galaxy experiment showed that the completeness fraction was, on average, about $\\sim 75$\\% for galaxies with $22\\le\\mu_{0}\\le24$, $V\\le20$ and $\\sim 50$\\% for lower surface brightness galaxies. After correction by completeness and background galaxies, we  found an excess of LSB dwarf galaxies in the fields of the groups that must be group members. Further work will add the data of all groups together to determine the combined luminosity function.  In addition, we plan to study the spatial distributions, colors, types and sizes of the dwarf galaxies as a function of the richness of their environments."
    },
    "0010/astro-ph0010116_arXiv.txt": {
        "abstract": "X--ray and optical observations have shown that the supernova explosions and stellar winds of the starburst in NGC\\,2146 produce an outflow of hot material along the minor axis. This outflow is expected to have a more or less conical shape, on either side of the galactic plane, and cone walls of cooler material where the outflow is in shock contact with halo gas. We attempt to determine the geometry (diameter at the base and opening angle) and the physical parameters (velocity and density) of the material in the cone walls from the optical emission line and radio observations presented here, and from published X--ray, radio, and optical observations. We compare the outflow of NGC\\,2146 with the outflow of M\\,82. ",
        "introduction": "In analogy to other starburst galaxies (Heckman et al.\\ 1990, Leh\\-nert $\\&$ Heckman 1996\\,a), the supernova explosions and stellar winds of the strong star formation activity in NGC\\,2146 produce an outflow of hot gas along the minor axis. The outflow is collimated in the disk by a ring of dense molecular gas confining the star formation region, and hence appears at X--ray and optical wavelengths under ideal conditions as cones on either side of the galactic plane. Such an outflow is especially prominent in galaxies viewed edge--on, as in the case of M\\,82. The hot gas at a temperature of 10$^{7}$ -- 10$^{8}$\\,K, seen at X--rays, is confined to the inner part of the cones; the gas in the cone walls in shock contact with the halo material is at a temperature of $\\sim$\\,10$^{4}$\\,K and emits the [forbidden] recombination lines of H$_{\\alpha}$, [NII], [SII], etc. The outflowing hot gas may drag cool gas and dust out of the disk at the inner edge of the molecular ring and transports this material, and angular momentum, to a considerable height in the halo. The structure of such an outflow is schema\\-tically shown by Heckman et al.\\ (1990, their Fig.\\,1) and clearly seen in M\\,82 at X--ray (Schaaf et al.\\ 1989, Bregman et al.\\ 1995) and optical wavelengths (Axon $\\&$ Taylor 1978, Bland $\\&$ Tully 1988, Shopbell $\\&$ Bland--Hawthorn 1998, McKeith et al. 1995, Devine $\\&$ Bally 1999). Predictions, calculations, and interpretations of these outflows under the aspect of energy input, outflow velocity, density and temperature of the gas, and height of the outflow were published by Chevalier $\\&$ Clegg (1985), Tomisaka $\\&$ Ikeuchi (1988), Umemura et al. (1988), Yokoo et al. (1993), and others.  The X--ray and optical emission of the outflow in NGC 2146 was observed by Armus et al. (1995) and Della Ceca et al. (1999). Ultra--compact/ultra--dense H\\,{\\sc ii} regions containing many O--type stars, and radio supernovae and supernova remnants were recently detected in the starburst region in a combination of MERLIN and VLA observations by Tarchi et al. (2000).   \\begin{figure} \\psfig{figure=minoraxis_bw.ps,height=11.0cm,angle=0} \\caption[]{H$_{\\alpha}$\\,6563\\,\\AA\\ and [NII]\\,6583\\,\\AA\\ measured along the minor axis (pa = 50$\\degr$) of NGC\\,2146. The [vertical] straight line at the very left is a sky emission line. The [horizontal] continuum emission comes from a region near the center of the galactic plane. Note the increased emission from an H\\,{\\sc ii} region complex at $\\sim$\\,2.5\\,kpc NE above the galactic plane (see Fig.\\,4 in Young et al., 1988).} \\end{figure}     Optical spectroscopy observations with the slit oriented along the minor axis show under favourable geo\\-metrical conditions characteristic velocity--split emission lines (M\\,82); conversely, measurement of the line splitting allows a determination of the outflow geo\\-metry and the outflow veloci\\-ty. Unfortunately, for reasons to be explained below, in NGC\\,2146 there is no very clear detection of line splitting and hence the geo\\-metry and the velocity of the outflow is not well known, although the galaxy is seen tilted from edge--on by only 25 -- 30\\,$\\degr$. From a combination of the observations of this paper (optical spectroscopy, CO line interferometer data) and published X--ray, radio, and optical measurements we attempt to determine the geo\\-metry and the physical parameters of the outflow cones: i.e.\\ the dia\\-meter ({\\it b}) of the base in the galactic plane, the opening angle ($\\Theta$), the outflow velocity ({\\it V}), and the (electron) density (n$_{\\rm e}$) of the material in the cone walls. We compare the outflow of NGC\\,2146 with the well studied outflow of M\\,82. ",
        "conclusions": "Using a combination of radio, optical, and X--ray observations, we have tried to derive the geometry and physical state of an {\\it idealized} minor axis outflow in NGC\\,2146: the outflow is conical on either side of the galactic plane; the diameter at the base is {\\it b} $\\approx$ 1\\,000\\,pc, comparable in dia\\-meter with the molecular ring; the opening angle is $\\Theta$ $\\approx$ 60$\\degr$; near the base the hot outflowing material drags gas and dust out of the disk; the outflowing material carries angular momentum into the halo; the cone walls of the outflow can be traced (at optical wavelengths) to a height of {\\it h} $\\approx$ 3\\,kpc; the (electron) density in the cone walls seems to decrease proportional to {\\it z}$^{-2}$, with {\\it z} the height above the galactic plane; the outflow velocity at $\\sim$\\,1\\,kpc height is 250 -- 300\\,km\\,s$^{-1}$.  While these parameters of the outflow give a consistent explanation of the radio, optical, and X--ray observations, the absence of velocity component IV and the low outflow velocity along the cone walls (at least cone wall II) may speak against the proposed interpretation. However, compared with M\\,82, the larger dimension of the starburst region and of the molecular ring, and an apparently smaller number of radio supernovae and supernova remnants seem to produce a larger flaring of the outflow ({\\it b}, $\\Theta$), and by this possibly a less propulsive outflow with a lower velocity. The weak component IV may also partially be hidden by the galactic disk, but its absence may also be in line with the general distortion of the galaxy. The coincidence of ultra--compact/ultra--dense H\\,{\\sc ii} regions, radio supernovae, and supernova remnants with the region of the CO ring and the interior gives confidence in the picture of a superwind driven minor axis outflow. \\\\"
    },
    "0010/nucl-th0010093_arXiv.txt": {
        "abstract": "Using the ETFSI (extended Thomas-Fermi plus Strutinsky integral) method, we have calculated the fission barriers of nearly 2000 exotic nuclei, including all the neutron-rich nuclei up to $A = 318$ that are expected to be relevant to the r-process, and all superheavy nuclei in the vicinity of $N = 184$, with $Z \\le 120$. Our calculations were performed with the Skyrme force SkSC4, which was determined in the ETFSI-1 mass fit. For proton-deficient nuclei in the region of $N$ = 184 we find the barriers to be much higher than previously believed, which suggests that the r-process path might continue to mass numbers well beyond 300. For the superheavy nuclei we typically find barrier heights of 6--7 MeV. \\\\  \\noindent PACS: 24.75.+i; 95.30.Cq\\\\  \\noindent{$^*$Supported in part by FNRS (Belgium) and NSERC (Canada)} ",
        "introduction": "The r-process of stellar nucleosynthesis depends crucially on the masses and fission barriers (among other quantities) of nuclei that are so neutron-rich that there is no hope of being able to measure them in the laboratory (see Refs. \\cite{meyer94,artak99} for reviews discussing the nuclear data required for an understanding of the r-process). It is thus of the greatest importance to be able to make reliable extrapolations of these quantities away from the known region, relatively close to the stability line, out towards the neutron-drip line. Until recently the masses and barriers used in all studies of the r-process were calculated on the basis of one form or another of the liquid-drop(let) model (LDM). However, in an attempt to put the extrapolations on as rigorous a footing as possible we have developed a mass formula that is based entirely on microscopic forces, the ETFSI-1 mass formula \\cite{dut,ton,pea,abo1,abo2}. Calculations of the r-process using the ETFSI-1 masses have already been performed \\cite{how93,ga96}, but they are incomplete in that fission had to be neglected, barriers not yet having been calculated in the ETFSI model. Of course, recourse could have been made to the extensive barrier calculations \\cite{how80,mey89} based on the LDM that had been used in earlier r-process studies, but it would have been inconsistent to use one model for the masses (required for the neutron-separation energies $S_n$ and the beta-decay energies $Q_\\beta$), and another for the barriers.  Here we remedy this deficiency by presenting the results of ETFSI-method calculations of the fission barriers of all of the nearly 2000 nuclei lying in the region of the ($N$,$Z$) plane shown in Fig. 1. This region covers the range $84 \\le Z \\le 120$, and extends from moderately neutron-deficient to extremely neutron-rich nuclei, including thereby not only a large fraction of the nuclei whose barriers are known experimentally, but also all nuclei with $A \\le 318$ whose barriers can reasonably be expected to be relevant to the r-process. Towards the upper limit of our range of $Z$ values the r-process path lies at much higher values of $A$ (if it has not already been terminated by fission), and we restrict ourselves to nuclei lying close to the stability line in the long-expected ``island\" of stability that is now becoming experimentally accessible \\cite{oga99,nin99,ogb99}.  The ETFSI method is a high-speed approximation to the Skyrme-Hartree-Fock (SHF) method, with pairing correlations generated by a $\\delta$-function force, treated in the usual BCS approach (with blocking). There are two parts to the total energy calculated by the ETFSI method, the first consisting of a purely semi-classical approximation to the SHF method, the full fourth-order extended Thomas-Fermi (ETF) method, while the second part, which is based on what we call the Strutinsky-integral (SI) form of the Strutinsky theorem, constitutes an attempt to improve this approximation perturbatively, and in particular to restore the shell corrections that are missing from the ETF part. The way in which we extended the ETFSI method, developed originally for the calculation of ground-state energies, to the large-scale calculation of fission barriers is described in Ref. \\cite{fiss1}, which should be consulted for all details of our calculational methods. However, it is important to recall here that our fission paths are optimized with respect to the the elongation parameter $c$, the necking parameter $h$, and the asymmetry parameter $\\tilde{\\alpha}$ (see Ref. \\cite{fiss1}, and in particular its App. A, for the definition of these quantities, and a description of the nuclear shapes allowed by our parametrization in the case of fission; note that as in that reference we are assuming axial symmetry).  In Ref. \\cite{fiss1} we calculated all the barriers that have been measured in nuclei with $Z \\ge 81$. The comparison of our results with the available data showed that if the primary, i,e., highest, barrier is lower than 10 MeV the error never exceeds 1.4 MeV, with either sign being possible, while for primary barrier heights lying between 10 and 15 MeV we probably overestimate the height by about 1.5 MeV; barrier heights in excess of 15 MeV are overestimated by 2 MeV or more. Since the primary barrier is the only one that is really relevant to the r-process we see that our method is sufficiently accurate for the calculation of barriers that are low enough to be of interest to the r-process (see Section 3 below). The calculations here are performed in exactly the same way, and in particular we use the same force parameters, set SkSC4, which, it should be noted, were determined entirely by the mass fit \\cite{abo1}, and have in no way been modified for the barrier calculations.  \\begin{figure} \\rotatebox{90}{\\resizebox{!}{\\hsize}{\\includegraphics{fiss2fig1.ps}}} \\caption{Schematic representation in the ($N$,$Z$) plane of the primary fission barriers displayed in Table \\ref{barriers}. The energy bins are described in the figure. Thin solid lines represent r-process paths for constant values (in MeV) of $S_a$ (defined in text). The neutron-drip line is shown by the thick solid line}  \\end{figure} ",
        "conclusions": "We have extended the ETFSI-1 mass formula, based on the Skyrme force SkSC4, to the calculation of the fission barriers of all nuclei that can be expected to play a role in the r-process of nucleosynthesis; we recall that the force SkSC4 gives an excellent fit to the mass data, and to the known primary-barrier heights that are lower than 15 MeV. The results that we present here are radically different from the only other such calculations that have been made \\cite{how80,ms99}, in that we obtain much higher barriers for proton-deficient nuclei in the region of $N$ = 184, a consequence of which is that the r-process path might continue to mass numbers considerably in excess of 300 before being brought to a halt by neutron-induced fission. In view of the importance of this result we summarize here the reasons why we believe that our calculations are essentially correct. Our high barriers on the r-path are related both to the shell effects associated with our Skyrme force, and to our much lower value of the symmetry coefficient $J$ (27 MeV as opposed to 36.5 MeV for HM \\cite{how80} and 32.65 MeV for MS \\cite{ms99}); we now examine each of these points in turn.  Our shell effects appear to be fairly robust within the ETFSI framework, being difficult to change significantly while maintaining the fit to the mass data. At the same time, we recall that in extrapolating far from the known region of the nuclear chart out to the highly neutron-rich region that is relevant to the r-process, the isospin dependence of the ETFSI spin-orbit field conforms well to the predictions of relativistic mean-field theory \\cite{ons97,np98}, adding thereby to our confidence in the reliability of the extrapolation.  As for the value of the symmetry coefficient $J$, we have found, after the completion of these calculations, that 28 MeV would have been a better value than 27 MeV, essentially because it allows us to avoid an unphysical collapse of neutron matter, while maintaining a high-quality mass fit. A complete new mass fit, ETFSI-2, is currently being undertaken with the constraint $J$ = 28 MeV, and in principle when this is completed we should repeat the calculations of this paper with the new force determined by this fit. However, we have shown here that the maximum effect of the change of force will be to lower the barriers of the most neutron-rich nuclei by about 3 MeV, and that for nuclei closer to the stability line the effect will be correspondingly smaller. In any case, one can be reasonably sure that the ETFSI-2 fission barriers would lie considerably closer to the ETFSI-1 barriers presented here than to the barriers of either MS \\cite{ms99} or HM \\cite{how80}.  We have also extended our barrier calculations of nuclei with $N$ in the vicinity of 184 down to the stability line, exploring thereby the much studied ``island\" of stability. Here our barriers are in reasonable agreement with other microscopic calculations that have been made in this region: this constitutes an additional check on the overall validity of our calculations, and there is certainly no indication of a tendency for our calculational procedure to overestimate barrier heights.  Nevertheless, we have not yet discussed triaxiality, which could be expected to lower fission barriers to some extent. Several studies of this question have been made, the most extensive probably being that of Ref. \\cite{dutt00} (see that paper for other references), in which the barriers of 15 heavy and superheavy nuclei, some close to the stability line, others highly neutron-rich, were investigated within the framework of the ETFSI method and found, when triaxiality was taken into account, to be lowered by 0.6 MeV on average, with a maximum shift of 1.3 MeV. It was concluded that this effect is probably negligible, given the overall discrepancy between calculation and experiment (see Section 1). Nevertheless, two counter-examples in both of which triaxiality is claimed to lower the barriers by the enormous amount of 4 MeV are to be noted: $^{310}126$ \\cite{cwiok96} and $^{258}$Fm \\cite{ben98}. The latter case, which was regrettably overlooked in Ref. \\cite{dutt00}, is particularly disturbing, since it occurs in a region of the nuclear chart where the barriers of neighbouring nuclei are much less sensitive to triaxiality. The former of these two nuclei is too heavy for us to check with our codes, but we have repeated the latter case using the ETFSI method, and find that triaxiality lowers the barrier by 1.6 MeV. This is a significantly smaller effect than the 4.1 MeV claimed in Ref. \\cite{ben98}, but it is still the largest triaxiality shift that we have found, and really too large to be neglected. One is thus forced to the conclusion that while {\\it in most cases} triaxiality can reasonably be neglected, as far as barriers are concerned, there will be isolated cases where this is not possible. The situation is most unsatisfactory, since such cases can only be identified by first performing the full triaxial calculation, and doing this for the nearly 2000 nuclei considered here would be prohibitively time-consuming. It seems that the best that one can do at the present time is to proceed as we have done here, neglecting triaxiality, and bear in mind that a few of our barriers, including some low enough to be relevant to the r-process, will be considerably overestimated because of this approximation. However, the barriers of our proton-deficient nuclei in the vicinity of $N$ = 184 will remain much higher than previously believed.  \\noindent {\\bf Acknowledgements}  We are indebted to B. Lorazo of the Computing Centre of the Universit\\'e de Montr\\'eal for his help in facilitating our computations, and to M. Farine for calculating the surface properties of the SkSC forces. P. M\\\"{o}ller is thanked for valuable communications. J. M. P. acknowledges the financial support of the FNRS (Belgium) that made possible an extended visit to the Universit\\'e Libre de Bruxelles, and is grateful for the hospitality extended to him at the Institut d'Astronomie et d'Astrophysique by Prof. M. Arnould. M. R. is a Research Associate of the FNRS."
    },
    "0010/astro-ph0010334_arXiv.txt": {
        "abstract": "We present high resolution echelle spectroscopy of a sample of H {\\sc II} galaxies.  In all galaxies we identify different H$\\alpha$ emitting knots along the slit crossing the nucleus. All of these have been isolated and separately analyzed through luminosity and size {\\it vs} $\\sigma$ diagnosis plots. We find that in all cases, for a particular galaxy, the bulk of emission comes from their main knot and therefore, at least for the compact class galaxies, we are dealing with, luminosity and sigma values measured using single aperture observations would provide similar results to what is obtained with spatially resolved spectroscopy. In the size {\\it vs} $\\sigma$ plots as expected there is a shift in the correlations depending on whether we are including all emission in a single point or we split it in its different emitting knots. The problem of a proper determination of the size of the emitting region so that it can be used to determine the mass of the system remains open. From the data set gathered, using the highest surface brightness points as recently proposed by Fuentes-Masip et al. (2000), the best luminosity {\\it vs} $\\sigma$ correlation turns out to be consistent with a Virial model. ",
        "introduction": "H {\\sc II} galaxies are a subsample of dwarf galaxies with a strong star formation activity leading to intense nebular lines easy to detect in objective prism surveys.  The optical properties of H {\\sc II} galaxies are dominated by their strong emission line spectra superposed on a weak blue continuum. The question, posed by Sargent \\& Searle (1970) of whether these are young galaxies forming stars for the first time seems now to have been answered by recent studies.  H {\\sc II} galaxies show an underlying stellar population of intermediate to old ages (Telles \\& Terlevich 1997, Doublier \\etal 1997, Marlowe \\etal 1999, Cair\\'os \\etal 2000). The optical/near-IR colors of their underlying galaxies are similar to late type dwarf galaxies, although their structural parameters derived from brightness profiles indicate that H {\\sc II} galaxies are more compact than late type quiescent dwarfs (Telles \\& Sampson 2000; Telles \\etal 1999).  Their morphological classification splits them into two groups: Type I includes the most luminous H {\\sc II} galaxies, all of which present an overall irregular morphology. Types II's, on the other hand, are more compact and rounder and present a luminosity profile similar to that of dwarf spheroidal galaxies (Telles, 1995).  The few H {\\sc II} galaxies found to have a bright neighbor (maybe by chance) are all Type II's of regular morphology, contrary to what one would expect if galaxy interactions could be held responsible for the morphological disturbances found in Type I's (Telles \\& Terlevich 1995).  { In addition, H {\\sc II} galaxies as a class are not preferentially clustered around faint low mass galaxies (Telles \\& Maddox 2000), but some are found to have HI companions (Taylor 1997). The triggering mechanism of the present starburst in H {\\sc II} galaxies is not at all clear, yet for these dwarfs internal processes must play a major role.}  Here we show echelle data obtained at the William Herschell telescope (4.2m) at the \"Observatorio del Roque de los Muchachos\" (ORM) at La Palma on a sample of type II dwarfs.  Our high resolution echelle data show a strong variation in the line profiles across the emitting region and even in the most compact sources, there is an indication of separate knots of star formation (SF) evolving concurrently within the galaxy nucleus.  { The presence of multiple knots of star formation within the line emitting regions are also observed on high spatial resolution images, in particular in the near-IR.  Telles \\etal (1999, 2000) also identified knots which are possible super stellar clusters (SSC). Their individual properties may impose further constraints on the history of SF of a galaxy as well as indicate possible mechanisms able to trigger the present burst.}   For our sample here we have defined different apertures across the H$\\alpha$ luminosity profile and have extracted the spectra corresponding to different emitting knots in every object. Also the {\\it total} spectra was obtained to mimic single aperture observations (Melnick \\etal 1988), when it includes the ensemble of individual knots.  The structure found in H {\\sc II} galaxies has profound implications on several topics. In particular on issues such as star formation and its possible sequential propagation in H {\\sc II} galaxies, and how is the ISM structured in these galaxies. Another issue central in this field of research is the validity of the interpretation, and use of the empirical correlations, of size and luminosity {\\it vs} their supersonic line widths for high redshift galaxies.  These correlations were first found for Giant H {\\sc II} Regions by Melnick (1979), Terlevich \\& Melnick (1981) and Melnick \\etal (1987) and extended to H {\\sc II} galaxies by Melnick \\etal (1988) and Telles \\& Terlevich (1993).  The fact that Giant H {\\sc II} Regions and H {\\sc II} galaxies have supersonic motions and follow the same correlations have made us use and refer to the whole emsemble of objects as a single family.  This may indicate that process of massive cluster formation is similar in both H {\\sc II} galaxies and Giant H {\\sc II} Regions and it is the main reason as to why we will not differenciate between them in discussing their dynamical properties.  Here we describe in Section~\\ref{data} and ~\\ref{analysis} our selected sample and observational strategy, and Section~\\ref{results} presents full detail of the results for each galaxy. A general discussion and a summary of the results is given on Section~\\ref{conclusion}. ",
        "conclusions": "\\label{conclusion}  Without a doubt, one of the most intricate issues in the field of giant H {\\sc II} regions, and now also of H {\\sc II} galaxies, is the origin of their supersonic velocity dispersion ($\\sigma_{gas}$ $>$ $c_{\\rm H {\\sc II}}$).  It has long been established that a simple agglomeration of a large number of H {\\sc II} regions cannot lead to the dominant supersonic line profiles, and thus, giant H {\\sc II} regions, and H {\\sc II} galaxies, constitute a different class of objects. Different not only because of the size of their emitting regions and the fact that they are powered by recent violent bursts of star formation, but also because of their essence: their peculiar inner dynamics.  It is now well established that the supersonic line width correlates with the size of the emitting region (size $\\propto$ $\\sigma^{2}$) and luminosity ($L$ $\\propto$ $\\sigma^4$) of the ionized regions (Terlevich \\& Melnick 1981, Melnick \\etal 1987,1988, Telles \\& Terlevich 1993). The fact that the correlations are similar to the relations inherent to virialized stellar systems such as globular clusters, spiral bulges and the cores of elliptical galaxies, led Terlevich \\& Melnick to postulate that giant H {\\sc II} regions and H {\\sc II} galaxies are themselves virialized systems and thus that the measured gas velocity dispersion ($\\sigma_{gas}$) should directly relate to their total mass. In their original scenario Terlevich \\& Melnick envisaged a gravitational potential that forced the collective motion of clumps of gas to present the supersonic $\\sigma$ values. However, in our sample and in the recent finding in giant H {\\sc II} regions the massive emitting knots present a relative velocity unable to explain the observed supersonic $\\sigma$ values (see Tenorio-Tagle \\etal 1993 1996, and references therein).  Instead, the brightest knots present an intrinsic value of sigma formerly ascribed to the whole ionized region.  This fine tuning regarding the size of the supersonic $\\sigma$ region and its luminosity has brought the empirical correlations into a much better agreement with what is expected from a virialized system.  On the theory ground, virialization is now believed to have lots to do with the motion of the low-mass stars moving in the gravitational potential of the system while undergoing winds. This enhances their cross-section and allows them to cause the stirring of the gas left-over from the star formation event (see Tenorio-Tagle \\etal 1993).   On the other hand, some authors have claimed that the supersonic line widths could result from a plethora of unresolved expanding shells caused by the mechanical energy of the large number of massive stars powering each of the sources (see Chu \\& Kennicutt 1994 and references therein). In the latter case however, there is no obvious way to explain the empirical correlations and thus it has been argued that the measured line widths follow the correlations simply due to a lack of resolution in the integrated spectra obtained from single aperture observations (but see also Tenorio-Tagle \\etal 1996).  Our results confirm and extend the empirical correlations found for giant H {\\sc II} regions and H {\\sc II} galaxies.  This fact has profound implications both on the observations of H {\\sc II} galaxies and on the interpretation of their supersonic line width.  Enhanced spectral and spatial resolution seems to unveil an intricate structure in H {\\sc II} galaxies.  Note that the previously measured (single aperture) supersonic motions in fact arise from regions of much smaller dimension than that occupied by the full extent of the ionized gas.  As shown in Figure~\\ref{fig3} and Figure~\\ref{fig4} for compact H {\\sc II} galaxies, accurate determinations of the size and luminosity of the region presenting the supersonic $\\sigma$ values nicely outline the correlations. H {\\sc II} galaxies when resolved, present several emitting knots with a variety of shapes, luminosity and $\\sigma$ values.  However, we have shown that in these cases the global integrated value agree very closely with the properties derived for the main emitting knot.  This is simply because the intrinsic properties (luminosity, velocity dispersion) of a galaxy are dominated by the central (core) component.  A fine calibration of these relations for local H {\\sc II} galaxies may be of great importance if used as a distance indicator of galaxies at large redshift, since H {\\sc II} galaxies are easy to find at great distances (see also Melnick \\etal 2000). Particularly, as the global line emitting properties reflect the intrinsic properties of the central core component, observations even with poor spatial resolution could accurately define the luminosity and $\\sigma$ values of the dominant central core in every galaxy."
    },
    "0010/astro-ph0010428_arXiv.txt": {
        "abstract": "Whereas a variation of the solar luminosity, L, will inevitably cause a similar change of the total solar irradiance, S, the opposite is not true. In fact, the bulk of the days to months variations of S can be explained entirely in terms of the passage of active regions across the solar disk. In this case, L remains essentially unchanged.  For the total irradiance variation observed over the solar cycle, the issue is more uncertain. One view explains this modulation primarily as a combined action of active regions and magnetic network. These components would be superposed to an otherwise unchanging photosphere. the other view suggests that the activity cycle modulation of S is primarily produced by a variation of L (both in terms of R and T$_{\\mbox{\\scriptsize{eff}}}$) caused by structural reajustments of the interior of the Sun induced by a changing magnetic field. We will present evidence in support of this second interpretation, and a model for it. We will also present the S variations over the last 5 centuries implied by our model. ",
        "introduction": "There is no question that \\begin{itemize} \\item The hours to months variations of the total irradiance are primarily (totally?) due to active regions. \\item The spots depress the irradiance. \\item The faculae add to the irradiance. \\item The contrasts are sufficiently high to be measured with some confidence. \\item The temporal behavior during one rotation is exactly as expected. \\end{itemize}  \\begin{figure} \\includegraphics[width=8cm]{talkfig1.eps} \\caption{ A composite solar irradiance record from the end of 1978 to the present \\protect\\citep{FL98} and the yearly mean of solar irradiance. \\label{talkfig1} } \\end{figure}  However, it is not the case for the 11-year cycle, as shown in Figure~\\ref{talkfig1} \\citep{FL98}. \\begin{itemize} \\item It is usually {\\bf assumed} that the magnetic network causes most of the modulation. \\item The measured network contrast is insufficient to account for the entire variation \\citep{EBF00}. \\item The precision of the irradiance measurements is less certain because instrument degradation is more significant than that in short timescales. \\item Proxies of the network are designated, and their magnitude is adjusted to minimize residuals with observations. \\end{itemize} From this viewpoint, it is assumed that the background photosphere remains unchanged during the entire cycle.  Of course, an alternative possibility is that most (if not all) of the 11 year variability is due to a change in the ''luminosity'' without the effects of the magnetic network. In order for that to happen, the following are true: \\begin{itemize} \\item The photospheric temperature must change. \\item The internal solar structure must change. \\item The solar radius must change. \\end{itemize} We propose here the explanation that the 11-year modulation of the total irradiance is due to structural adjustments of the solar interior in response to a variable internal magnetic field. ",
        "conclusions": "From what we present above, we reach the following conclusions: \\begin{itemize} \\item The total irradiance variation, and the photospheric temperature variation observed over the 11-year activity cycle can be explained in terms of the variation of an internal solar magnetic field of 20-47 kG located at $r=0.96 R_{\\odot}$. \\item The above result is in agreement with helioseismological data, and with the variations of the solar radius measured with MDI/SOHO. \\item The extrapolation of this process to the past 2-3 centuries produces a change in luminosity of only 0.1\\%, and a radius change of only $0.02$ arcsec. \\item If radius variations of order $0.5$ arcsec do occur, a larger (1.3 MG) variation of a field located below the base of the convection zone is required. \\item The combined effect of both phenomena can yield a $\\Delta L$ of 0.2\\%\\ over many decades. \\end{itemize}"
    },
    "0010/astro-ph0010272_arXiv.txt": {
        "abstract": "We report the discovery using the Parkes radio telescope of binary millisecond pulsars in four clusters for which no associated pulsars were previously known.  The four pulsars have pulse periods lying between 3 and 6 ms. All are in circular orbits with low-mass companions and have orbital periods of a few days or less. One is in a 1.7-hour orbit with a companion of planetary mass. Another is eclipsed by a wind from its companion for 40\\% of the binary period despite being in a relatively wide orbit.  These discoveries result from the use of improved technologies and prove that many millisecond pulsars remain to be found in globular clusters. ",
        "introduction": "Globular clusters are a rich source of millisecond pulsars. Besides the evolution of primordial binary systems, exchange interactions in their core result in the formation of binary systems containing neutron stars. These systems subsequently evolve, spinning up the neutron star through mass accretion (\\cite{sb76}; \\cite{bv91}; \\cite{ka96}).  The millisecond pulsars formed in this way are among the most stable clocks in nature and are valuable for studies of the dynamics of clusters, the evolution of binaries embedded in them, and the interstellar medium (\\cite{phi92}; \\cite{hmg+92}; \\cite{bv91b}; \\cite{fcl+00}).  However, they are difficult to find because the pulsed emission is weak and distorted by propagation through the interstellar medium, and the apparent pulse period may change rapidly because of binary motion.  We have constructed a new high resolution filterbank system to improve the sensitivity to pulsars with high dispersion measure (DM) and we have implemented a new multi-dimensional code to search over a range of DMs and over a range of accelerations resulting from binary motion.  Using these new capabilities, we have undertaken a search of globular clusters for millisecond pulsars. So far, we have discovered four millisecond pulsars in four clusters, none of which had previously known pulsars associated with them. All of these pulsars are members of short-period binary systems, and two of them have relatively high DM values. These detections bring the number of clusters containing known pulsars to 16. ",
        "conclusions": "We have discovered short-period binary millisecond pulsars in four globular clusters which contained no previously known pulsars.  These detections break the long hiatus in such discoveries and will help in the current understanding of millisecond pulsars in globular clusters. In fact, the four clusters presented here are very close and dense, and the absence of millisecond pulsars in their core was rather intriguing.  The new discoveries result from the use of enhanced receiving systems having low system noise, wide bandwidths and good frequency resolution, combined with the use of high-speed data acquisition systems and improved search algorithms operating on powerful computing systems. They demonstrate that many millisecond pulsars, especially those in short-period binary systems, remain to be discovered in the globular clusters of our Galaxy.  Such systems provide important constraints on the formation and evolution of binary systems in clusters and are significant in investigations of cluster dynamics. At present, it is not at all obvious why some clusters (e.g., 47 Tucanae) have large numbers of detectable pulsars, whereas other apparently similar clusters (e.g., NGC 6544) have few or none."
    },
    "0010/astro-ph0010044_arXiv.txt": {
        "abstract": "We have monitored Supernova 1987A in optical/near-infrared bands from a few weeks following its birth until the present time in order to search for a pulsar remnant. We have found an apparent pattern of emission near the frequency of 467.5 Hz -- a 2.14 ms pulsar candidate, first detected in data taken on the remnant at the Las Campanas Observatory (LCO) 2.5-m Dupont telescope during 14-16 Feb.~1992 UT. We detected further signals near the 2.14 ms period on numerous occasions over the next four years in data taken with a variety of telescopes, data systems and detectors, at a number of ground- and space-based observatories.  The sequence of detections of this signal from Feb.~`92 through August `93, prior to its apparent subsequent fading, is highly improbable (${<10^{-10}}$ for any noise source).  We also find evidence for modulation of the 2.14 ms period with a $\\sim$1,000 s period which, when taken with the high spindown of the source (2-3$\\times$10$^{-10}$ Hz/s), is consistent with precession and spindown via gravitational radiation of a neutron star with a non-axisymmetric oblateness of $\\sim$10$^{-6}$,  and an implied gravitational luminosity exceeding that of the Crab Nebula pulsar by an order of magnitude. ",
        "introduction": "We present evidence herein for the presence of an unique, 2.14-ms pulsar, which appears to precess and whose spindown is apparently dominated by gravitational radiation, but one which has also been very difficult to detect over the past three years due to its faintness and the complicated nature of its signal. \\begin{figure} \\plotfiddle{fig1.eps}{3.0in}{-90}{50}{50}{-162}{277} \\caption{The pulse profiles for the 2.14 ms periodicity detected on UT 26 July, 12 Sep.~and 24 Sep.~`93 with the 1-m telescope are plotted against the UT 14-16 Feb.~`92 detection with the LCO 2.5-m telescope.} \\label{fig-1} \\end{figure} A more detailed and exhaustive treatment of this candidate and upper limits for any other signal are given elsewhere (Middleditch et al. 2000). ",
        "conclusions": ""
    },
    "0010/astro-ph0010208_arXiv.txt": {
        "abstract": "We investigate spatial distributions of specific dust components in the circumstellar shells of a proto-planetary nebula candidate, \\inx{HD 179821}, and a planetary nebula, \\inx{BD$+$30$^{\\circ}$3639}, by means of spectral imaging. With high-resolution ground-based images and {\\sl ISO} spectra in the mid-infrared, we can derive ``dust feature only'' maps by subtracting synthesized continuum maps from the observed images at the feature wavelength. Such spatially detailed information will help to develop models for the evolution of dust grains around evolved stars. ",
        "introduction": "With {\\sl ISO\\/} discoveries of a wealth of infrared (IR) emission features from evolved stars (e.g., Waters et al.\\ 1996; Waters et al.\\ 1998) and laboratory measurements of IR spectra for candidate dust species (Speck 1998 for a compilation of references), we are now able to identify individual circumstellar dust components.  Using such detailed information to interpret high-resolution ground-based mid-IR images we can spatially separate dust emission due to different species by means of spectral imaging. This can be done by (1) identify an emission feature and nearby continuum bands, (2) construct a color temperature map from the continuum images, (3) synthesize a continuum image at the feature wavelength from the temperature map taking into account diffraction effects, and (4) subtract the synthesized continuum-only map from the observed feature band image. ",
        "conclusions": ""
    },
    "0010/astro-ph0010514_arXiv.txt": {
        "abstract": "Recent precise observations of solar global parameters are used to calibrate an upgraded solar model which takes into account magnetic fields in the solar interior. Historical data about sunspot numbers (from 1500 to the present) and solar radius changes (between 1715 and 1979) are used to compute solar variability on years to centuries timescales. The results show that although the 11 year variability of the total irradiance is of the order of $0.1\\%$, additional, longer lived changes of the order of $0.1\\%$ may have occurred in the past centuries. These could, for example, account for the occurrence of climate excursions such as little ice ages. ",
        "introduction": "Because of the steady increase of the concentration of greenhouse gases in the Earth's atmosphere due to anthropogenic activity, global warming is potentially the most serious environmental threat of our time. In order to assess the magnitude of the threat, we must understand the sensitivity of the climate system to a variety of external driving processes. In particular, the magnitude of the solar variability component of climate change, potentially very significant, is still relatively uncertain.  One way of assessing the effects of a given driver of climate is by trying to fit the observations (e.g. reproducing the air temperature variations of the last 150 years \\citep{JWW86,HL88} or sea surface temperature \\citep{R91}) with, and without the driver in question operating. For the case of the Sun, we can assume that it does not change, and adjust all the other parameters to reproduce the temperature record. Subsequently, a given model of the solar variability is assumed, and then all the parameters are re-adjusted to reproduce the same temperature record. Obviously, if the solar variability contributes to the observed trend, the model sensivisity to the other processes that were assumed to produce the trend on their own must decrease. Similarly, if the solar component varies in antiphase with the observed trend, the sensitivity to the other processes must be increased. Andronova and Schlesinger carried out such a calculation \\citep{AS00}. They used two models for the solar variability.  One, constructed by \\citet{LBB95}, henceforth denoted by L, was based on sunspot areas and (when available) locations, He 1083 nm emission, group sunspot numbers, and Ca emission from the Sun and solar analogues.  The other model, developed by \\citet{HS93}, (henceforth called HS), was based on the fraction of sunspot area occupied by the penumbra, equatorial rotation rate, length and decay rate of the solar cycle, and the mean level of solar activity.  Upon including solar variations, the model sensitivity to greenhouse gases changed drastically, decreasing by 53 percent for the HS solar variations, and by 44 percent for the L solar variation. Obviously, if either model provides an approximately accurate representation of solar variability over the last 150 years, the solar input to climate variability is very important. ",
        "conclusions": "According to the picture presented in this paper, the solar variations of years to centuries are due to variations of magnetic fields at two different depths in the solar interior.  The 11 year cycle would arise from magnetic fields of the order of 20-40 kG, located at or above $0.96 R_{\\sun}$.  At the same time, there is a stronger field located at just below the base of the convection zone that undergoes changes on longer timescales. In fact, this field affects where the base of the convection zone is located. The temporal characteristics of the deeper field is still unknown, since the radius data currently available have a scant number of data points, poorly distributed in time.  The amplitude of the variations of the solar luminosity produced by the shallower field is of the order of $0.1\\%$, while the deeper field can produce changes of the same order of magnitude. The solar luminosity when both magnetic components are strong would be about $0.2\\%$ higher than when both are weak.  For instance, the luminosity deficit caused by the shallow field at the time of the Maunder minimum is about $0.1\\%$. To cause the decrease of the radius observed between 1715 and 1979, the magnetic field at the base of the convection zone had to decrease from about 1.3 MG to about 600 kG. Since 1979 is at the maximum of Cycle 21, the corresponding magnetic field at the minimum is smaller than one half of the field at the maximum. We thus infer that the magnetic field at the minimum is smaller than 300 kG, which is in agreement with helioseismology \\citep{ACT00}. The luminosity due to the deep field in 1715 is higher than that in 1979 by about $0.1\\%$, but is almost the same as that in 1925. The Maunder minimum terminated around 1715, when both shallow and deep magnetic field components are, according to our model results, strong. We know that the shallow magnetic field was weak during the Maunder minimum, because of the small sunspot numbers during that period. If we assume that the deeper component was also weak, the solar luminosity during the Maunder minimum may have been about $0.2\\%$ lower than that after 1715. Such a large irradiance change may account for the transition between the little ice age and the following warm period.  \\begin{figure}[t] \\figurewidth{8cm} \\includegraphics[width=8.cm]{sofiafig4.eps} \\caption{Solar variability corresponding to the probable change in the solar radius between 1715 to 1979. $R_{\\mbox{\\scriptsize{CZ}}}$ is the location of the base of the convection zone, while $R_B$ is the location of maximal magnetic field. MG stands for Million Gauss. The mass depth is defined as $\\log(1-M/M_{\\sun})$ by the mass coordinate M. The smaller the mass depth, the closer to the surface of the model sun. } \\label{fig:century} \\end{figure}   The solar radius change found to occur between 1925 and 1976 (or 1979) brings up an additional issue. The implication is that the solar luminosity in 1925 was about $0.1\\%$ higher than that in 1976 (or 1979). Of course, the climate consequences of this higher luminosity period depend strongly on the duration of this phenomenon, which at present is unknown.  However, there is a possibility that the solar effect alone would have caused some cooling, and that the global warming caused by the increase of the concentration of greenhouse gases could be more significant than what has been estimated ignoring the solar effect. Because of the complexity of the climate response to solar irradiance, and in view of the serious uncertainties in the current solar data base, we cannot, at present, draw any sweeping conclusions.  What we can conclude with some certainty is that solar variability could be a significant driver of climate change, and that the efforts made to date in addressing this issue are inadequate. The physical model that we have developed, with all its inadequacies (for example, it is unidimesional, whereas the problem is clearly three-dimensional) can constitute a framework which could form the basis for interpreting a variety of historical solar observations."
    },
    "0010/astro-ph0010387_arXiv.txt": {
        "abstract": "We demonstrate that the maximum-entropy method for gravitational lens reconstruction presented in Bridle et al. (1998) may be applied even when only shear \\emph{or} magnification information is present. We also demonstrate that the method can easily handle irregularly shaped observing fields and, because shear is a non-local function of the lensing mass, reconstructions that use shear information can successfully bridge small gaps in observations. For our simulations we use a mass density distribution that is realistic for a $z=0.4$ cluster of total mass around $10^{15} h^{-1} M_{\\odot}$. Using HST-quality shear data alone, covering the area of four WFPC2 observations, we detect $60$ per cent of the mass of the cluster within the area observed, despite the mass sheet degeneracy. This is qualitatively because the shear provides information about the variations in the mass distribution, and our prior includes a positivity constraint. We investigate the effect of using various sizes of observing field and find that $50$ to $100$ per cent of the cluster mass is detected, depending on the observing strategy and cluster shape. Finally we demonstrate how this method can cope with strong lensing regions of a mass distribution. ",
        "introduction": "\\label{intro}  The study of gravitational lensing by clusters of galaxies is important for cosmology as a probe of both dark and luminous matter. `Weak lensing', the very slight shearing and magnification of background galaxies is much more prevalent than the more spectacular `strong lensing' effects, such as multiple or highly distorted images. See Mellier (1999) and Bartelman \\& Schneider (2000) for recent reviews. Many methods have now been proposed for reconstructing the cluster mass distribution from weak lensing observations and many authors have applied these methods to observations in order to reconstruct the mass distributions. The problem splits naturally into two stages. Firstly the problem of estimating a shear and/or magnification maps from observations, and secondly estimating the mass density distribution from the shear and/or magnification maps. In this paper we are concerned with the second problem.  The first non-parametric method for inverting shear data to estimate the mass distribution was proposed by Kaiser \\& Squires (1993) \\& used just the shearing of background galaxies. This method has now been improved by several authors to circumvent many of its original problems (Schneider \\& Seitz 1995; Seitz \\& Schneider 1995; Kaiser 1995; Squires \\& Kaiser 1996). However, Kaiser \\& Squires, Schneider \\& Seitz and others have indicated that even perfect shear information is insufficient to reconstruct the projected mass unambiguously due to the `mass sheet degeneracy'. More recent methods have therefore combined shear information with magnification information in order to overcome this problem (Bartelmann et al 1996; Seitz, Schneider \\& Bartelmann 1998, Bridle et al. 1998). Because these methods iteratively find the mass or gravitational potential distribution which best fit the observations, they also can easily overcome the difficulties of the original Kaiser \\& Squires method.  It has also been noted that magnification data alone are enough to reconstruct the lens (Broadhurst, Taylor \\& Peacock 1995; Dye \\& Taylor 1998). Besides having the potential to avoid the mass sheet degeneracy problem, an advantage of using magnification data alone is that the point spread function of the telescope and the circularising effects of the atmosphere are much less of a problem (Broadhurst et al. 1995). It does however require the sizes and/or number densities of \\emph{unlensed} background galaxies to be known. In addition, one has to know -- or assume -- the luminosity, redshift and spatial distribution of background galaxies (e.g. Dye \\& Taylor 1998).  In an earlier paper (Bridle et al. 1998, hereafter BHLS) we presented a maximum-entropy method for reconstructing the projected mass distribution from both shear and magnification data. In contrast to comparable methods, we reconstruct the projected mass density distribution itself, rather than reconstructing the two-dimensional Newtonian potential and then converting this into the mass distribution. This seems to us to be a more elegant and only slightly more computationally expensive method. As a result, we are then able to use the shape of the probability function in order to estimate the errors on our reconstruction. In addition, the method allows reconstruction outside the observed field, which makes fuller use of the shear information. In this paper we demonstrate the way in which this method may be extended to the cases where no shear information is available, or where no magnification data are available, and illustrate the effect on the resulting reconstructions. An evaluation of the relative merits of using shear versus magnification information, along with a comparison of reconstructions for a two parameter lens model, is given in Schneider, King \\& Erben (2000). We highlight the flexibility of our method in analysing observations of irregularly-shaped patches of sky and reconstructing mass distributions containing strong lensing regions. ",
        "conclusions": "\\label{conclusions}  In investigating how well the BHLS maximum entropy method is capable of deducing the matter distribution of typical massive galaxy clusters at $z\\approx0.4$ from typical shear data, from magnification data and from both, we have demonstrated the following: \\begin{enumerate} \\item Shear data alone provide a useful estimate of the mass of clusters: in simulations with realistic signal to noise and field of view, the algorithm detects $50$ to $100$ per cent of the mass of the cluster despite the mass degeneracy, depending on the size of the observation and the profile of the cluster. \\item The reconstruction method can easily handle irregularly shaped and disjoint observing regions. \\item Because shear information is a non-local function of the mass, small gaps in observations can be bridged successfully when shear information is used in the reconstruction. \\item It is also successful at reconstructing mass distributions from simulated magnification data alone. \\item If the cluster has super-critical regions then the data from these regions can be cut out and the reconstruction will bridge these regions. \\end{enumerate}   \\subsection*"
    },
    "0010/astro-ph0010178_arXiv.txt": {
        "abstract": "Instabilities and strong dynamical interactions between several giant planets have been proposed as a possible explanation for the surprising orbital properties of extrasolar planetary systems. In particular, dynamical instabilities would seem to provide a natural mechanism for producing the highly eccentric orbits seen in many systems. Here we present results from a new set of numerical integrations for the dynamical evolution of planetary systems containing two identical giant planets in nearly circular orbits very close to the dynamical stability limit. We determine the statistical properties of the three main types of systems resulting from the development of an instability: systems containing one planet, following either a collision between the two initial planets, or the ejection of one of them to infinity, and systems containing two planets in a new, quasi-stable configuration. We discuss the implications of our results for the formation and evolution of observed extrasolar planetary systems. We conclude that the distributions of eccentricities and semimajor axes for observed systems cannot be explained easily by invoking dynamical interactions between two planets initially on circular orbits. While highly eccentric orbits can be produced naturally by these interactions, collisions between the two planets, which occur frequently in the range of observed semimajor axes, would result in many more nearly circular orbits than in the observed sample. ",
        "introduction": "The existence of planetary systems around other stars is now well established.  For several years already, we have known many more planets outside than inside our own Solar System (currently $\\sim50$, including planets around radio pulsars).  Several groups with ongoing radial-velocity surveys have reported many unambiguous detections of Jupiter-mass planets around nearby solar-like stars (for recent reviews and updates, see Marcy and Butler 1998, 2000; Hatzes {\\it et al.\\/} 2000; Korzennik {\\it et al.\\/} 2000; Perryman 2000; Vogt {\\it et al.\\/} 2000; Santos et al.\\ 2000).  Astrometry can sometimes help constrain the parameters of the wider systems (e.g., Mazeh {\\it et al.\\/} 1999) and other, newer techniques such as gravitational microlensing (Gaudi {\\it et al.\\/} 2000) and space interferometry (Fridlund 1999) are making rapid progress.  Some of the most exciting recent developments include the detections of planetary transits in HD 209458 (Charbonneau {\\it et al.\\/} 2000; Henry {\\it et al.\\/} 2000) and two sub-Saturn-mass candidates around HD 16141 and HD 46375 (Marcy {\\it et al.\\/} 2000).  The long-term stability of the Solar System, in spite of its chaotic nature (see, e.g., Duncan and Quinn 1993), may have been necessary for the development of intelligent life.  However, it may also be very {\\it atypical\\/}, and may in fact require very special conditions during the early stages of planet formation.  In particular, the existence of a single dominant massive planet in our Solar System (Jupiter), although perhaps essential for long-term dynamical stability, may not be typical of planetary systems that form around other stars. In fact, current theoretical models for the formation of giant planets by accretion of gas from the nebula onto a rocky core, combined with observed mass distributions of protoplanetary disks (e.g., Beckwith and Sargent 1996) suggest that many planetary systems could  form initially with 2--3 Jupiters. In a sufficiently massive protoplanetary disk ($\\ga 0.01M_\\odot$), disk instabilities may also lead naturally to the formation of several giant planets (Armitage and Hansen 1999; Boss 1998). However, one should be careful not to rely solely on theoretical models for planet formation that, for many decades, have been based on only the Solar System for guidance. What the newly detected systems (see Fig.~\\ref{FigEvsA}) clearly tell us is that there exists a much greater variety of other planetary systems than theorists had ever imagined. Some of the unexpected properties of these systems (e.g., highly eccentric orbits) suggest that, unlike our Solar System, they may well have been affected by dynamical instabilities (Rasio and Ford 1996; Weidenschilling and Marzari 1996; Lin and Ida 1997).  \\begin{figure} \\plotone{EvsAwM.ps} \\figcaption{Semimajor axes and eccentricities of all presently known radial-velocity planetary candidates with $m \\sin i< 13 M_{\\rm J}$ ($M_{\\rm J}$ is the mass of Jupiter). The area of each open dot is proportional to the value of $m \\sin i$ for that object.  One group of planets has $a \\la 0.07\\,$AU and nearly circular orbits, while the other, with $a \\ga 0.07\\,$AU, contains many highly eccentric orbits. The data shown here include all planets discovered up to and including the one in $\\epsilon$ Eri (see Hatzes {\\it et al.\\/} 2000 and references therein). Values of $m\\sin i$, $a$ and $e$ for 50 planets were taken from the table compiled by Marcy {\\it et al.\\/} at {\\tt http://exoplanets.org/} as of Oct 28, 2000. \\label{FigEvsA}} \\end{figure}  In a system containing two or more Jupiter-like planets of comparable masses the possibility exists that a dynamical instability will develop, leading to strong gravitational interactions or collisions between the planets (Gladman 1993, Chambers {\\it et al.\\/} 1996). Here we will use numerical integrations of the orbital dynamics to explore the consequences of such dynamical instabilities.  Based on a preliminary set of calculations for systems containing two identical giant planets (Rasio and Ford 1996) we expect a frequent outcome of these instabilities to be a physical collision between two giant planets. Very little mass is lost in such a collision, and the result is therefore a more massive giant planet in a slightly more eccentric orbit.  However, in many cases, the interaction can also lead to the ejection of one planet to a larger distance while the other is left in a slightly smaller, highly eccentric orbit.  If the inner eccentric orbit has a short enough pericenter distance (distance of closest approach to the star), it may later circularize through tidal dissipation, leaving a Jupiter-type planet in a very tight circular orbit around the star, with an orbital period typically of order a few days.  Indeed, all the recently detected planets around nearby stars are Jupiter-mass objects in very tight circular orbits, or in wider eccentric orbits (Fig.~\\ref{FigEvsA}).  The standard model for planet formation in our Solar System (see, e.g., Lissauer 1993) is incapable of explaining them.  According to this standard model, planetary orbits should be nearly circular, and giant planets can only form at large distances ($\\ga1\\,$AU) from the central star, where the temperature in the protostellar nebula is low enough for icy materials to condense (Boss 1995, 1996).  There are many ways for dynamical instabilities to develop in a protoplanetary system.  The simplest scenario, that we will be assuming here, is that two nearly identical giant planets had initially formed (in the conventional way) at a large distance from the central star, and later interacted dynamically (Rasio and Ford 1996). This could happen because their orbital radii evolved secularly at different rates (significant orbital migration is thought to have occurred in the outer Solar System; see Goldreich and Tremaine 1980, Malhotra 1995), bringing them closer together, or because the masses increased as the planets accreted their gaseous envelopes (Lissauer 1993), or both (see Kley 2000, who shows that a system of two identical giant planets still embedded in a protoplanetary disk generically evolves toward a dynamical instability).  The dynamical instability leads eventually to orbit crossing and strong gravitational interactions between the two planets (Gladman 1993).  Other formation mechanisms have been proposed for the 51-Peg-type planets (in very tight circular orbits with orbital periods $\\sim 3-5\\,$d). If these planets had formed, like our own Jupiter, at a large distance from the central star, some angular-momentum-loss mechanism must have brought them in.  A {\\it slow migration\\/} mechanism, such as friction in the protostellar nebula or interaction with a protoplanetary disk, would tend to increase rapidly with decreasing separation.  The dissipation would have had to switch off at a critical moment for the planets to end up so close to the star without being disrupted. Although there exist mechanisms that can provide a barrier at some very short distance from the star (Lin {\\it et al.\\/} 1996; Murray {\\it et al.\\/} 1998; Trilling {\\it et al.\\/} 1998), this always seems to require some fine tuning of the parameters or extreme conditions such as a very massive disk (Murray {\\it et al.\\/} 1998; Rasio {\\it et al.\\/} 1996).  Alternatively, direct {\\it in situ\\/} formation of all 51-Peg-type planets by accretion onto a solid core may also be possible under some conditions (Bodenheimer {\\it et al.\\/} 2000).  Alternative mechanisms for inducing a large eccentricity in a planetary orbit include the gravitational perturbation  by a distant binary stellar companion (Holman {\\it et al.} 1997; Mazeh {\\it et al.\\/} 1997) and dynamical interactions with a gaseous protoplanetary disk (Papaloizou 2000) or with a disk of planetesimals (Murray {\\it et al.\\/} 1998). Perturbations of planetary orbits in a binary star system are well understood theoretically (see, e.g., Ford {\\it et al.\\/} 2000; Innanen {\\it et al.\\/} 1997) and can lead to arbitrarily large eccentricities provided that the relative inclination is large enough and the binary companion is not too far from the planet. While interactions with a disk typically {\\it damp\\/} orbital eccentricities, they may in some special situations lead to modest eccentricity growth. However, within our limited current theoretical understanding of these processes (see, e.g., Nelson {\\it et al.\\/} 2000), it appears unlikely that they would be able to produce eccentricities as large as those observed for many Jupiter-mass objects (with measured eccentricities up to $e\\simeq0.7$; see Fig.~\\ref{FigEvsA}).  The strongest observational evidence to date that extrasolar planetary systems may be affected by dynamical instabilities is provided by the recent detection of two giant planets in wide eccentric orbits around $\\upsilon$~And (Butler {\\it et al.\\/} 1999). These are in addition to the previously discovered 51-Peg-type planet, in a 4.6-day, nearly circular orbit around the star. The presence of one or more additional giant planets {\\it in wider eccentric orbits\\/} in the 51-Peg-type systems is a basic theoretical prediction of our mechanism (see Rasio and Ford 1996), since at least one other planet of comparable mass must have been present to trigger an instability. Moreover, the two outer orbits in $\\upsilon$~And are tightly coupled (ratio of semimajor axes $\\simeq1/3$) and the system is still very close to the edge of dynamical stability (numerical integrations indicate that it may in fact be unstable on timescales $\\sim 10^6-10^7\\,$yr depending on the precise values of the masses and orbital parameters; see Rivera and Lissauer 2000).  This provides further evidence that the present configuration resulted from the evolution of the progenitor planetary system through a phase of violent dynamical instability.  Indeed, systems of multiple planets that become unstable tend to evolve first through a violent phase where energy and angular momentum are quickly redistributed, followed by a much more gradual settling into a more stable configuration that remains very close to the stability edge (Chambers {\\it et al.\\/} 1996).  Additional support for a scenario based on dynamical instabilities comes from the latest detection, around the nearby K2~V star $\\epsilon$ Eri: a planet with mass $m\\sin i = 0.86\\,M_{\\rm J}$, a long orbital period $P = 6.9\\,$yr ($a\\simeq 3.3\\,$AU), and a large eccentricity $e\\simeq 0.6$ (Hatzes {\\it et al.\\/} 2000). The star is not in a binary system and, at this large orbital separation, the planet is unlikely to have had a significant interaction with a protoplanetary disk (which would also have produced significant inward migration). This clearly leaves dynamical interaction with another giant planet (which was likely ejected from the system) as the most natural explanation.  While our study will concentrate on the systematic study of a system containing two giant planets, other groups have performed small numbers of exploratory calculations to determine the consequences of dynamical instabilities in systems containing three planets or more. Weidenschilling and Marzari (1996) have published results for the case of three planets, while Lin and Ida (1997) presented results for systems containing up to 9 planets. With many planets, successive collisions and mergers can lead to the formation of a fairly massive ($\\ga 10\\,M_{\\rm J}$) object in a wide, eccentric orbit.  We feel that it is important to first understand fully the case of two planets. One important advantage of the two-planet case is that the dynamical stability boundary is very sharply defined, and its location known analytically (Gladman 1993). Therefore, the initial value for the ratio of semimajor axes $a_2/a_1$ must be varied only in a very narrow range (right around the stability boundary) for each case. In contrast, for three or more planets, the stability boundary is not well defined (Chambers {\\it et al.\\/} 1996), and a much wider range of semimajor axis ratios would have to be explored. In addition, if the instability is triggered by the increase in the mass of one planet as it accretes its gaseous envelope, we would expect that it would naturally tend to involve only two giant planets, since it appears extremely unlikely that more than two planets would be going through the accretion process at precisely the same time.  Indeed a time difference $\\ga 10^6\\,$yr between the formation of the different giant planets would be expected from the standard scenario (Lissauer 1993). ",
        "conclusions": ""
    },
    "0010/astro-ph0010297_arXiv.txt": {
        "abstract": "The cluster \\object{Abell~2104} is one of the lowest redshift clusters ($z=0.153$) known to have a gravitational lensing arc. We present detailed analysis of the cluster properties such as the gravitational potential using the X-ray data from {\\it ROSAT} (HRI) and {\\it ASCA}, as well as optical imaging and spectroscopic data from the CFHT. The cluster is highly luminous in the X-ray with a bolometric luminosity of $L_{x}\\sim 3\\times 10^{45}$ ergs~s$^{-1}$ and a high gas temperature of $\\sim 10.4$ keV. The X-ray emission extending out to at least a radius of 1.46 Mpc, displays significant substructure.  The total mass deduced from the X-ray data under the assumption of hydrostatic equilibrium and isothermal gas, is found to be $M_{tot}(r<1.46\\mathrm{Mpc})\\sim (8.0\\pm 0.8)\\times 10^{14} M_{\\sun}$. The gas fraction within a radius of 1.46 Mpc is $\\sim 5-10$\\%. The cluster galaxy velocity distribution has a dispersion of $1200\\pm200$ km\\,s$^{-1}$ with no obvious evidence for substructure. The total mass within 1.46 Mpc, deduced from Jean's equation using the observed galaxy number density distribution and velocity dispersion, is found to be $\\sim 6.8\\times 10^{14} M_{\\sun}$ to $\\sim 2.6\\times 10^{15} M_{\\sun}$ marginally consistent with the X-ray deduced total mass. ",
        "introduction": "\\subsection{Observations}  The data were collected in 4 nights at the 3.6 m CFHT Telescope in May 1993. Two 10 minutes exposures in B band and two 15 minutes exposures in R band were obtained. Exposures of 30 to 55 minutes per spectroscopic mask was obtained for 3 separate masks, each containing about 30 slits (Fig.~\\ref{chart}).  The focal reducer MOS/SIS together with CCD Lick2 ($2048 \\times 2048$ pixels of 15 $\\mu$m) were used during the run. This CCD is a thick device having a quantum efficiency of $\\sim 10\\%$ in the blue. The observing configuration provides a pixel size of $0.314\\arcsec$ over a field of view of about $10\\arcmin \\times 10\\arcmin$. The overall image quality was good (stellar FWHM $\\sim 0.9\\arcsec $) although some optical distortions were conspicuous near the edges of the images due to the optics of the focal reducer.  \\begin{figure*} \\resizebox{\\hsize}{!}{\\includegraphics{8809.f1}} \\caption []{ Optical field of \\object{Abell~2104}, observed at the CFHT in R band.  Overlaid are the {\\it ROSAT} HRI contours with levels $(1.7, 2.0, 2.6, 3.2, 3.8, 4.3, 4.6)\\times 10^{-6}$~counts\\,\\,s$^{-1}$\\,arcsec$^{-2}$. The X-ray image was rebinned into $2^{''}$ pixels and smoothed with a $10^{''}$ Gaussian. \\label{superover} } \\end{figure*}  \\begin{figure} \\resizebox{\\hsize}{!}{\\includegraphics{8809.f2}} \\caption{A close up image of the central regions of \\object{Abell~2104} showing the giant arc.} \\label{f:arc} \\end{figure}  \\begin{figure*} \\resizebox{\\hsize}{!}{\\includegraphics{8809.f3}} \\caption []{ Finding chart for galaxies with measured redshifts. The reference numbers are the same as in Table~\\ref{spectro}. The non-member galaxies are marked with a bracket around their reference number. \\label{chart} } \\end{figure*}   \\subsection{Photometric analysis}  The B and R frames were prepared using standard pre-reduction techniques. Since there were only 2 frames per filter, cosmic rays were removed by taking the lower pixel value in cases where a pixel in one frame is significantly higher than the corresponding pixel in the other frame. The photometric analysis was performed by means of the SExtractor package (Bertin \\& Arnouts, 1995) in the same way as Pierre et al. (1997), but adapted to our data. The images were first slightly smoothed to give the same PSF in B and R frames, then the background was estimated using a 64 $\\times$ 64 pixel mesh. Source detections were claimed if at least 9 adjacent pixels were above a threshold corresponding to 1.5 times the local noise level. The CCD Sequence in M 92 (\\cite{Christian}) observed during the same run was used for photometric calibration. Stars VCS1, A, B (probably variable) had to be removed because of obvious inconsistencies.  Estimates of the photometric errors were taken directly from the SExtractor analysis, and are less than $\\sim 0.1$ for R$<22.5$ and less than $\\sim 0.2$ for B$<23.5$.    The catalogue is estimated to be complete to R = 22.5 and B = 23.5. On inspection of the detected objects above the completeness limit, we found those objects with a SExtractor classification $<0.15$ may be assumed to be galaxies, i.e., 275 objects. Changing the threshold does not affect the outcome significantly because most of the galaxies are well separated from stars (3/4 of the objects fall below 0.05 or above 0.95).   Fig.~\\ref{f:cm} shows the colour magnitude diagram for all the galaxies detected in the R-frame, and the corresponding magnitudes in the R and B bands were measured within the same apertures. The band of E/S0 sequence galaxies is discernable in Fig.~\\ref{f:cm}; the spectroscopically confirmed cluster members are shown to fall mostly on the E/S0 sequence confirming that a large fraction of the galaxies on the E/S0 sequence belongs to the cluster. The mean error in B-R colour is $< 0.08$.    \\begin{figure} \\resizebox{\\hsize}{!}{\\includegraphics{8809.f4}} \\caption []{ A colour magnitude diagram for all galaxies detected in the R frame. The filled circles are for spectroscopically confirmed cluster members, and the crosses are for the non-members. The solid line gives the completeness limit. \\label{f:cm} } \\end{figure}    \\subsection{Spectroscopy}  Grism O300 was used for the spectroscopy. It has a zero deviation at 5900~\\AA, covers approximately 4700--7900~\\AA, and gives a dispersion of 3.59~\\AA$/$pixel ($0.314\\arcsec$). The slit has a width of 2\\arcsec, i.e. 6.4 pixels, yielding a resolution of $\\sim$23~\\AA\\ FWHM. Since there was only 1 frame per mask, cosmic rays were picked out individually by eye and replaced by the median of the surrounding pixels.  The internal Helium and Argon lamps was used for wavelength calibration. The subsequent reduction was performed as described in Pierre et al. (1997).  Redshifts were measured by a cross-correlation method implemented in the MIDAS environment following Tonry and Davis (1979). The cross-correlation results for each spectrum were checked independently by eye.  The results from the cross-correlation analysis for all spectra are presented in Table~\\ref{spectro}.  Heliocentric correction has not been applied, but is negligible at this resolution. The absolute error in the velocity calibration is $\\sim 200$\\,km\\,s$^{-1}$.  As a first guess, galaxies are considered to be cluster members if they lie within 3000 km\\,s$^{-1}$ of the central cD galaxy, which selects 47 (the main sample) out of the 60 galaxies. This procedure eliminates most of the foreground and background galaxies without affecting the dispersion measurements significantly.  If we relax the velocity constraint and apply the usual $3\\sigma-$clipping technique then we have 51 cluster members (the extended sample). The cluster redshift distribution for both samples is displayed in Fig.~\\ref{f:zhist}. The histogram includes all galaxies in the redshift range $z\\sim 0.135 - 0.175$ in Table~\\ref{spectro} and a Gaussian corresponding to the velocity distribution of the main sample. The bi-weighted mean and scale for the main sample are $z=0.1532^{+0.0004}_{-0.0006}$ and $\\sigma=1148^{+190}_{-65}$\\,km\\,s$^{-1}$ correspondingly; and $z=0.1538^{+0.0009}_{-0.0006}$ and $\\sigma=1401^{+160}_{-130}$\\,km\\,s$^{-1}$ for the extended sample. It is difficult to find an objective criterion for deciding which galaxies are cluster members. Even with the sophisticated weighting scheme employed by Carlberg et al. (1997), the determination of the weight for each galaxy is still subjective. In Table~\\ref{spectro}, we have marked only the galaxies from the main sample as cluster members.   For the main sample we have enough redshifts to test whether or not the galaxy velocities are drawn from a Gaussian distribution applying various statistical tests for normality (e.g. D'Agostino \\& Stephens 1986; {\\small ROSTAT} -- Beers et al., 1990; Bird \\& Beers, 1993). As a result Anderson-Darling test (A$^2$) accepts the hypothesis for normal distribution at 90\\% significance, the combined skewness and kurtosis test (B1 \\& B2 omnibus test) at 97\\% level and the alternative shape estimators, asymmetry index and tail index based on order statistics, were found to be $-0.21$ and $0.98$ respectively which also show that the velocity distribution is drawn from a Gaussian.   We can obtain a conservative estimate of the errors on the velocity dispersion by comparing the dispersion from the extended and main samples. When we take into account of the uncertainties in cluster membership, a more conservative estimate of the errors should give the velocity dispersion as $1200\\pm 200$ km\\,s$^{-1}$.  We also investigated the presence of substructures in ($\\alpha$, $\\delta$, $z$) space but no obvious signal was detected (see Fig.~\\ref{f:wedge}).  More redshifts are required for a proper statistical analysis.   \\begin{figure} \\resizebox{\\hsize}{!}{\\includegraphics{8809.f5}} \\caption []{ Cluster galaxy redshift histogram (bin size $\\Delta z = 0.0025$). The galaxies considered to be cluster members are marked as solid histogram while the dotted histogram are for the extended sample. The dashed curve is a Gaussian with parameters corresponding to the velocity distribution of the galaxies in the main sample. \\label{f:zhist} } \\end{figure}   \\begin{figure} \\resizebox{\\hsize}{!}{\\includegraphics{8809.f6a}} \\resizebox{\\hsize}{!}{\\includegraphics{8809.f6b}} \\caption[]{Wedge diagrams showing the distribution of galaxies with measured redshifts. \\label{f:wedge} } \\end{figure} ",
        "conclusions": "The rich cluster \\object{Abell~2104} at a redshift of $z=0.1533$ was found to have a high X-ray luminosity ($\\sim 9.0\\times 10^{44}$ ergs~s$^{-1}$ in [0.1-2.4] keV) and temperature ($10.4\\pm0.6$ keV) from {\\it ROSAT} HRI and {\\it ASCA} data. The central cooling time, $t_{cool} \\sim 10^{10}$\\,yr for this cluster indicates the absence of a cooling flow. The galaxy velocity distribution showed that the cD galaxy was at rest at the bottom of the cluster potential. The X-ray image shows significant substructure in the centre of the cluster and an overall elliptical appearance. It appears that the cluster has not yet reached dynamical equilibrium.  As shown in Evrard et al (1996) and Schindler et al. (1996), the total mass deduced from assumptions of dynamical equilibrium are not significantly different from the true values. The total mass deduced from the X-ray data assuming hydrostatic equilibrium is consistent with the dynamic mass deduced from Jean's equation.  However, the current data on the projected galaxy density distribution and our knowledge of the galaxy orbits are limited for studies of cluster dynamics, which allows a wide range of possible parametric functions for the spatial galaxy density distribution without even attempting the non-parametric methods of Merritt and Tremblay (1994) or considering any anisotropic orbits. This can be improved by a deep wide-field observation, to extend the galaxy number density distribution to a large radius (up to 3 Mpc) and to allow a direct measure of the cluster mass from a weak shear analysis. This would allow us to definitively address the issue of whether or not the cluster is in dynamical equilibrium and constrain the range of possible total mass distributions allowed by the wide-range of data from lensing effects to X-rays. In order not to bias the results and incorporate a wide-range of the possible total mass density distributions, a non-parametric method should also be employed. With the launch of XMM and Chandra, we will soon able to obtain a temperature profile and probe the X-ray emission at the edge of the cluster which is crucial to the improvement of the X-ray mass estimates."
    },
    "0010/astro-ph0010068_arXiv.txt": {
        "abstract": "We discuss several effects due to lensing of ultra high energy cosmic rays by the regular component of the galactic magnetic field. Large flux magnification around caustics can be a significant source of clustering in the arrival directions of UHECRs of comparable energy. We also discuss lensing effects in a hypothetical galactic magnetic wind model recently proposed to explain the extremely high energy cosmic rays so far observed as originating from a single source (M87). This model implies large flux magnifications, which reduce the power requirements on the source, and a significant asymmetry in the expected flux between the north and south galactic hemispheres. ",
        "introduction": " ",
        "conclusions": ""
    },
    "0010/astro-ph0010542_arXiv.txt": {
        "abstract": "We have obtained 1-2$\\>$\\AA\\ resolution optical Echellette spectra of the nuclear star cluster in the nearby starburst galaxy NGC\\,4449. The light is clearly dominated by a very young ($6-10\\myr$) population of stars. For our age dating, we have used recent population synthesis models to interpret the observed equivalent width of stellar absorption features such as the H\\,I Balmer series and the Ca\\,II triplet around 8500$\\>$\\AA . We also compare the observed spectrum of the nuclear cluster to synthesized spectra for stellar populations of varying ages. All these approaches yield a consistent cluster age. Metallicity estimates based on the relative intensities of various ionization lines yield no evidence for significant enrichment in the center of this low mass galaxy: the metallicity of the nuclear cluster is about one fourth of the solar value, in agreement with independent estimates for the disk material of NGC\\,4449. ",
        "introduction": "Recent observations with the Hubble Space Telescope (HST) at both optical \\citep{phi96,mat99} and near-infrared \\citep*[NIR,][]{car98,boe99a} wavelengths have revealed that a compact, photometrically distinct star cluster is often present in the center of spiral galaxies of all Hubble types. The formation mechanism of such clusters is puzzling, especially in late-type spirals which by definition do not possess a prominent bulge. In these cases, the gravitational potential is very shallow, and it is far from obvious how the gas could fall all the way to the center to form a dense star cluster.  In this paper, we describe the results of a pilot program to investigate the processes of nuclear cluster formation. More specifically, we have performed 1-2$\\>$\\AA\\ resolution optical Echellette spectroscopy of the nucleus of NGC\\,4449, which is part of a small sample of galaxies with prominent nuclear star clusters that were identified from a recent NIR survey with HST \\citep{boe99a}. The morphological classification of NGC\\,4449 has been the subject of some confusion. It was originally classified as Irr by \\citet{hub36} and \\citet{san61}, but it is listed as Sm in \\citet{san81}. As \\cite{vdb95} points out, the presence of a stellar nucleus might be a useful criterion to distinguish between the two classes, since Irregulars usually do not have a well-defined nucleus - in contrast to late-type spirals, as discussed above. The extent, morphology, and dynamics of its H\\,I gas suggest that NGC\\,4449 has undergone some interaction in the past, a fact that might well account for its unusually high star formation activity \\citep{hun99}. It also has a prominent stellar bar which covers a large fraction of the optically visible galaxy. Both the nuclear cluster and the stellar bar are evident in Figure~\\ref{fig_image}, which shows a ground-based R-band image by \\cite{fre96}.  The fact that the nucleus of NGC\\,4449 is also prominently seen in the far ultraviolet (FUV) \\citep{hil98} and in various emission lines \\citep{sab84,may94,hil98,boe99a} indicates that it is the site of ongoing star formation. NGC\\,4449 is somewhat unusual in this regard, since emission-line surveys have shown that the nuclei of late-type galaxies often are devoid of ionized gas \\citep[e.g.][]{boe99a}. It appears that NGC\\,4449 is one of the few late-type spirals that is ``caught in the act'' of ongoing star formation in its nucleus.  NGC\\,4449 was chosen for a case study because of its high cluster-to-disk contrast which allows us to isolate the light of the nuclear cluster from the underlying galaxy. From the disk-subtracted cluster spectrum, we derive the equivalent width (EW) of a number of stellar absorption features. We then use stellar population synthesis methods to constrain the star formation history of the nuclear cluster and to determine the age of its dominant stellar population. In \\S~\\ref{sec:data}, we describe our observations and the data reduction methods. Section~\\ref{sec:results} describes the quantitative analysis of a number of emission lines and stellar absorption features, as well as our spectral synthesis efforts. From the results of these various diagnostic tools, we deduce an age of about $\\rm 6-10\\,Myr$ for the nuclear cluster. In \\S~\\ref{sec:discussion}, we discuss our results and compare them to a recent study by \\cite*{gel01} which confirms our age estimate from a complementary dataset. We also briefly discuss suggested mechanisms of nuclear cluster formation and their possible impact on the evolution of the host galaxy. ",
        "conclusions": "\\label{sec:discussion} The results of this study imply that NGC\\,4449 can be added to a growing list of late-type spiral galaxies that host a compact, young star cluster in their nucleus. While accurate spectroscopy - and hence a reliable age estimate - of nuclear clusters is available for only a few objects such as the Milky Way \\citep{kra95}, NGC\\,7793 \\citep{shi92}, and IC\\,342 \\citep*{boe99b}, photometric measurements show that many clusters are fairly blue, e.g. those in M\\,31, M\\,33, NGC\\,4242, or ESO\\,359-029 \\citep{lau98,mat99}. This indicates the existence of a young ($\\leq 100\\myr$) stellar population in many of these nuclei.  From an HST survey of intermediate-type spirals, \\cite{car98} find that virtually all galaxies with exponential bulges (i.e. bulges with an exponential surface brightness profile, as opposed to ``power-law'' or ``$R^{1/4}$'' bulges) host compact nuclear clusters and that their luminosity correlates with that of the host galaxy. If one assumes that the mass-to-light ratio of the host galaxies as a whole is roughly the same for all objects, the most straightforward interpretation of this result is that more massive galaxies have more massive nuclear clusters. On the other hand, stellar populations fade with time \\citep[e.g.][]{bru93}. Another possible interpretation therefore is that more massive galaxies host younger clusters. This intrinsic degeneracy makes age dating of nuclear clusters from photometric data alone impossible. Spectroscopic analysis is needed to investigate the age distribution of nuclear clusters and to decide whether their formation is a one time event or not. If nuclear clusters indeed form recurrently, the formation processes likely impact the morphological evolution of the host galaxy.  Of particular interest in this context is a model suggested by \\citet{fri93} and developed further by \\citet*{nor96}. These authors point out that build-up of a central mass concentration - e.g. a nuclear star cluster - can dissolve stellar bars, and lead to the formation of a bulge via collective bending instabilities \\citep{rah91,mer94}. The \\cite{nor96} simulations show that only 5\\% of the combined disk and bar mass in a central concentration is sufficient to destroy a stellar bar on short timescales, leading to a bulge-like distribution of stars. In this scenario, repeated cycles of bar formation --- gas infall --- cluster formation --- bar disruption will build the galaxy bulge over time. If true, the Hubble classification scheme could be naturally explained as an evolutionary sequence from late to early Hubble types.  In the case of NGC\\,4449, the cluster mass of (at least) $4\\times 10^{5}\\msun$ derived in \\S~\\ref{sec:mass} would be sufficient to destroy a bar with a mass up to $10^{7}\\msun$. The unusually prominent stellar bar in NGC\\,4449, which covers a large part of the optically visible galaxy, is probably much more massive than this value. This can be seen from the total luminosity of NGC\\,4449, which is listed in the RC3 \\citep{dev91} as $\\rm V^0_T = 9.53$, which corresponds to $\\rm L_V \\approx 2\\times 10^9\\>L_{\\odot ,V}$. Since the optical image is dominated by the stellar bar, and typical mass-to-light ratios are of order unity, the bar mass should be $\\approx 10^9\\msun$. The fact that it co-exists with the nuclear cluster is therefore not inconsistent with the \\cite{nor96} scenario.  However, as pointed out before, NGC\\,4449 is a special case because of its unusually high overall star formation rate and its complex history with evidence for past interactions. The complex gas flows produced by a past merger might well explain why -- different from most other late-type spirals \\citep[e.g.][]{boe99a} -- the nucleus of NGC\\,4449 contains large amount of ionized gas. Furthermore, there is no indication of an underlying older population of stars in the nuclear cluster of NGC\\,4449, as is the case, e.g., in the Milky Way. We point out, however, that for external galaxies where individual stars cannot be resolved, it is difficult to detect an old stellar population underlying a young starburst because the latter is dominating the luminosity (assuming similar masses for both bursts). It is somewhat unclear at this point whether the processes that govern the star formation in the nucleus of NGC\\,4449 are similar to those in ``typical'' late-type spirals, or whether NGC\\,4449 is an entirely different ``special'' case.  After the present paper had been finished, another study of the nuclear cluster in NGC\\,4449 appeared in the literature \\citep{gel01}. Using ground-based spectroscopy of the $\\rm ^{12}CO(2,0)$ and $\\rm ^{12}CO(3,1)$ absorption features around $2.3\\mum$ and UVI colors from HST imaging, these authors derived a cluster age between 8 and $15\\myr$, in good agreement with our estimate of $6-10\\myr$. Assuming a somewhat lower extinction to the NGC\\,4449 nucleus than derived by us from the Balmer decrement (\\S~\\ref{subsubsec:extinction}), Gelatt \\ea\\ conclude that the color of the nuclear cluster is better matched by evolutionary tracks with a metallicity of Z=0.008, somewhat higher than the average over the NGC\\,4449 disk. However, increasing the extinction by one magnitude (roughly the difference between our extinction and theirs) would easily move the location of the nuclear cluster in their Figure~3 within the range of the lower metallicity tracks. After accounting for the different extinction estimates, the derived V-I color of the nuclear cluster differs by about 0.2 magnitudes in the two studies. This could be the result of the somewhat different aperture sizes used to extract the cluster light ($1.4\\as$ for Gelatt \\ea\\ compared to $2.0\\as$ for our study). Either way, both papers come to very similar conclusions on the age of the nuclear cluster in NGC 4449, with complementary methods.  This study is part of a larger observational program that contains three independent, but related approaches. First, we will use the WFPC2 onboard the Hubble Space Telescope during Cycle 10 to measure the structural properties of all nuclear clusters found in a large sample of nearby, face-on spiral galaxies of late Hubble type. We will then measure the stellar velocity dispersion in the brightest nuclei found in order to derive the cluster mass, as recently demonstrated for the example of IC\\,342 \\citep{boe99b}. Finally, we hope to obtain optical spectra with HST/STIS to derive the stellar population and to constrain the formation history of individual nuclear clusters in much the same way as described in this paper. We aim to study a large number of nuclear clusters to allow a statistical analysis of their properties and dependence on the Hubble type of the host galaxy. Since ground-based observations are possible only for the brightest nuclear clusters, they are likely biased towards the young end of their age distribution. The goal of our study is to eliminate this bias. Measuring the true age distribution of nuclear star clusters will allow us to answer the question whether repeated nuclear starbursts are indeed common in late-type spirals, and, if so, what their duty cycle is."
    },
    "0010/astro-ph0010318_arXiv.txt": {
        "abstract": "The observed iron K$\\alpha$ fluorescence lines in Seyfert~\\uppercase{i} galaxies provide strong evidence for an accretion disk near a supermassive black hole as a source of the emission.  Here we present an analysis of the geometrical and kinematic properties of the disk based on the extreme frequency shifts of a line profile as determined by measurable flux in both the red and blue wings.  The edges of the line are insensitive to the distribution of the X-ray flux over the disk, and hence provide a robust alternative to profile fitting of disk parameters.  Our approach yields new, strong bounds on the inclination angle of the disk and the location of the emitting region.  We apply our method to interpret observational data from MCG-6-30-15 and find that the commonly assumed inclination $30^\\deg$ for the accretion disk in MCG-6-30-15 is inconsistent with the position of the blue edge of the line at a $3\\sigma$ level.  A thick turbulent disk model or the presence of highly ionized iron may reconcile the bounds on inclination from the line edges with the full line profile fits based on simple, geometrically thin disk models. The bounds on the innermost radius of disk emission indicate that the black hole in MCG-6-30-15 is rotating faster than 30~\\% of theoretical maximum. When applied to data from NGC~4151, our method gives bounds on the inclination angle of the X-ray emitting inner disk of $50\\pm 10^\\deg$, consistent with the presence of an ionization cone grazing the disk as proposed by Pedlar et al.~(1993).  The frequency extrema analysis also provides limits to the innermost disk radius in another Seyfert~1 galaxy, NGC~3516, and is suggestive of a thick disk model. ",
        "introduction": "The {\\it Advanced Satellite for Cosmology and Astrophysics} (ASCA) has provided data from over a dozen Seyfert~\\uppercase{i} galaxies to reveal the presence of iron emission lines which are broadened by a considerable fraction of the speed of light --- greater than 0.2~$c$ in some cases (Mushotzky et al. 1995; Tanaka et al. 1995; Nandra et al. 1997a). The observed line profiles are the most direct evidence for the presence of supermassive ($\\sim 10^{8}$~M$_\\odot$) black holes in the centers of these galaxies: the spectra have distinctive skewed, double-peaked profiles which reflect the Doppler and gravitational shifts associated with emitting material in a strongly curved spacetime (Chen \\& Halpern 1989; Fabian et al. 1989; Laor 1991; Kojima 1991).  The data strongly support a model wherein the emission lines are produced by iron K$\\alpha$ fluorescence at 6.4~keV when optically thick, ``cold'' regions of an accretion disk (such that the ionization state of iron is less than Fe~XVII) are externally illuminated by hard X-rays~(George~\\& Fabian 1991; Matt, Perola \\& Piro 1991).  In one bright, well-studied source, MCG-6-30-15, the high signal-to-noise ratio has enabled parameters of a simple geometrically thin, relativistic model to be estimated (Tanaka et al. 1995; Dabrowski et al. 1997; Reynolds, Begelman 1997; Bromley, Miller \\& Pariev 1998).  The model parameters which may be gleaned from line profiles include disk radii, emissivity of the disk, observed inclination angle of the disk~$i$ ($i=0^\\deg$ is face-on and $i=90^\\deg$ is edge-on), and the spin parameter of the black hole, $a_\\ast = J c/G M^2$, where $J$ is the hole's angular momentum. If the hole is rotating, we assume that the disk lies in the equatorial plane of the black hole, as a result of the Bardeen--Peterson (1975) alignment mechanism, and that the disk is corotating.  Of these parameters, perhaps the most problematic is the disk emissivity.  The distribution of the hard X-ray flux which illuminates the disk determines the emissivity of the fluorescing disk material, and therefore has a strong influence on line profiles.  The emissivity is usually assumed to be axisymmetric. This is reasonable if the observed profile is obtained with a long integration time since strong asymmetries presumably would average out. Specific choices for emissivity include a power-law in radius (Fabian et al. 1989; Bromley, Chen~\\& Miller 1997), a form consistent with a point source of illumination (Matt, Fabian~\\& Ross 1993; Matt, Fabian~\\& Ross 1996; Reynolds \\& Begelman 1997), and a function proportional to the total energy flux in the Page \\& Thorne (1974) accretion model (Dabrowski et al. 1997). Dabrowski et al. (1997) also considered a non-parametric form of the emissivity function. Recently, Iwasawa et al. (1999) performed a line-profile fit assuming nonaxisymmtric X-ray illumination of the disk in MCG--6-30-15. Nonaxisymmetric emissivity might be expected for the bright state of this source if the mass of the central black hole is $\\sim 10^8\\,M_{\\odot}$. In this case the integration time of the observations is shorter than the orbital period at a radius of a few times that of the horizon.  The calculation of emissivity may be complicated somewhat by the local physics of the disk as well as the nature of the illuminating source. If the incident radiation is strong it can cause iron to become highly or fully ionized. For iron atoms at ionization stages no higher than Fe~\\uppercase{xvi}, K$\\alpha$ emission occurs at 6.4~keV, however, resonant absorption and successive Auger processes prevent significant line emission by Fe~\\uppercase{xvii} through Fe~\\uppercase{xxiii}. Lithium-- and Helium--like ions, Fe~\\uppercase{xxiv} and Fe~\\uppercase{xxv}, emit at approximately 6.7~keV, while Hydrogen--like ion Fe~\\uppercase{xxvi} produce K$\\alpha$ line at 6.97~keV (Matt et al.~1996; Matt et al.~1993; see also the review by Fabian et al. 2000).  Local anisotropy of rest-frame emission is another effect which can influence an observed line profile. It is often assumed that the emitter is locally isotropic, although Laor~(1991) considered limb darkening and Matt et al.~(1996) considered effects of resonant absorption and scattering which can result in anisotropic emission.  The next level of detail in modeling local disk physics was to consider non-Keplerian flows, as when material falls within the innermost stable orbit around a Schwarzschild black hole (Reynolds \\& Begelman 1997), or when there is turbulence in a disk of finite thickness (Pariev \\& Bromley 1998).  More recently, temporal variations in line profiles (Lee et al., 2000; Iwasawa et al., 1999; Nandra et al., 1999) have been observed. Simulations of line profiles with a time-dependent illumination (Reynolds et al. 1999; Young \\& Reynolds, 2000; Ruszkowski 1999) are also performed.  Knowledge of the line variability can provide information about the size of the inner region of the accretion disk and the mass of a black hole.  To circumvent uncertainties in the emissivity of the disk and yet still obtain estimates and constraints of fundamental disk parameters such as the inner radius, Bromley, Miller \\& Pariev (1998) suggested using the minimum and maximum frequency shifts of a broad emission line as a diagnostic. As first demonstrated by Cunningham (1975), the edges of a line profile from an emitting disk annulus uniquely determine the radius of the annulus and its inclination. Similarly, bounds may placed on these parameters when the emission comes from many such annuli.  These bounds are more robust indicators than estimates derived by fitting the full profile shape when the emissivity is nonaxisymmetric or when the continuum has been poorly modeled in the central regions of the broad lines.  In this work we use the method of line frequency extrema to study the geometric and kinematic properties of relativistic accretion disks. We provide new methods of line edge detection based on polynomial fits to the line profile and continuum, and compare the results to the $\\chi^2$ rejection technique used by Bromley, Miller \\& Pariev (1998). In \\S~2 we describe the minimum and maximum frequency diagnostic and ways to determine the position of a line edge and its error. In \\S~3 we apply the technique to the iron line profiles from MCG-6-30-15, NGC~4151, and NGC~3516, and we compare the results of the edge fitting to the fitting of the whole line profile. Finally, we discuss prospects for our method with data from future X-ray missions. ",
        "conclusions": "In this work we suggest a method which allows us to put bounds on the geometry and kinematics of the line emitting surface of a relativistic accretion disk around a supermassive black hole.  Our technique is not sensitive to models of the illuminating source, the distribution of illuminating radiation over the disk surface, or the angular distribution of the reflected line in the rest frame of the reflecting material.  We use our method to determine the positions of the red and blue edges of observed emission lines and extract information from these positions by comparison with the results of a fully relativistic ray-tracing code.  The code generates maps of line-frequency extrema, and it can be used to demonstrate explicitly the differences which can arise as a result of black hole spin and turbulence in an accretion disk.  Conventionally, the only way to identify black hole rotation is to find emission from within the innermost stable orbit of a nonrotating black hole. Our method illustrates that if one has information about the geometry of the accretion flow from some other observation such as the inclination angle of a jet, along with an estimate of the luminosity relative to Eddington, then a new measure of the hole's spin may be available. We can thus break the ``degeneracy'' that exists between rotating and nonrotating black holes in a new way.   Certainly, our method cannot be a substitute for the fitting of the whole shape of the line assuming a particular emissivity law (power law axisymmetric in most works) and it cannot provide interesting information about the emissivity law itself. However, it can provide bounds on the geometry of the disk and the angular momentum of the black hole, which any emissivity model should satisfy. Therefore, it is important to emphasize the position of the line edges and to test model parameters obtained as a result of line profile fitting against the constraints provided by our method.  A drawback of the method are uncertainties in the determination of the position of the edges. A main focus of this paper is to find a reasonable edge detection algorithm to replace the conservative edge-boundary limits found by Bromley, Miller \\& Pariev (1998). We perform Monte-Carlo simulations of the data set using fits to the line edges with both a nonlinear sharp-edge model and a polynomial model which makes no a priori assumption about the shape of the line profile. We find reasonable agreement between models, both in terms of best-fit values and the error distributions.  Another drawback of the method is possible contamination with iron K$\\beta$ line, nickel K$\\alpha$ line, $6.7\\,\\mbox{keV}$ and $6.97\\,\\mbox{keV}$ K$\\alpha$ lines of highly ionized iron. We made corrections to profiles for the iron K$\\beta$ line using a K$\\beta$-to-K$\\alpha$ yield ratio and for the nickel K$\\alpha$ line using a fiducial yield of $0.06$ relative to iron K$\\alpha$ (see George \\& Fabian~1991). For quality of spectra currently available, these corrections are small.  We illustrate frequency extrema method by applying it to the Seyfert galaxies MCG-6-30-15, NGC~4151, and NGC~3516. The results for MCG-6-30-15 show that the commonly assumed inclination angle of $30^\\deg$ for a thin accretion disk is inconsistent with the position of the blue edge of the line at a $3\\sigma$ level. The thick turbulent disk model can remedy this discrepancy, since it leads to appearance of a smooth blue wing in the line while the changes to the main profile are not so large (Pariev~\\& Bromley, 1998). Recent estimates of the mass of the black hole in  MCG-6-30-15 favor a luminosity ratio $L/L_{\\rm ed} \\sim 0.6$, which is enough to account for the $g_{\\rm max}$ for all observations of the iron line in MCG-6-30-15 published to date. Furthermore, frequency extrema lead us to conclude that the black hole in MCG-6-30-15 must be rotating with at least $a/M=0.26$.  For NGC~4151 our method excludes face~on and edge~on geometries, while giving bounds for the inclination angle of X-ray emitting inner disk of $50\\pm 10^\\deg$. These bounds are consistent with the models of the ionization cone grazing the disk by Pedlar et al.~(1993) and two X-ray emitting disks of Wang et al.~(1999). However, our bounds are not consistent with the $i=65^\\deg$ geometry assumed by Evans et al.~(1993) based on the observation of the biconical geometry of the narrow line [O~\\uppercase{iii}]~$\\lambda$5007 sources. Because of very low ratio of $L/L_{\\rm ed}$, effects of turbulence in this source are negligible.  We find that frequency extrema results combined with line profile fitting by Nandra et al.,~(1999) favors Schwarzschild vs. Kerr model for NGC~3516. Although $L/L_{\\rm ed} \\sim 0.08$ for this source, the determination of the mass of the black hole is very uncertain. The possibility exists that the thick disk model may be relevant for NGC~3516 and could explain the blue wings of the line, observed during monitoring of this source by Nandra et al.~(1999).  Further reverberation mapping results will provide better estimates of the mass of the black holes in nearby Seyferts and allow the thick turbulent disk model to have more predictive power for each individual source. Certainly, going beyond our simplifying assumptions and incorporating results of modeling of X-ray reflection spectra from the disk surface would be the next step to interpret higher quality spectral data from XMM and Constellation-X missions."
    },
    "0010/astro-ph0010632_arXiv.txt": {
        "abstract": "Thanks to the Rossi X-ray Timing Explorer (RXTE), it is now widely recognized that fast X-ray timing can be used to probe strong gravity fields around collapsed objects and constrain the equation of state of dense matter in neutron stars.  We first discuss some of the outstanding issues which could be solved with an X-ray timing mission building on the great successes of RXTE and providing an order of magnitude better sensitivity.  Then we briefly describe the ``Experiment for X-ray timing and Relativistic Astrophysics'' (EXTRA) recently proposed to the European Space Agency as a follow-up to RXTE and the related US mission ``Relativistic Astrophysics Explorer'' (RAE). ",
        "introduction": "In the last few years the predicted sub-millisecond plasma orbital periods and millisecond spin periods of accreting low-magnetic-field neutron stars (NS) have finally been detected by RXTE. The same satellite also found 0.1~kHz Quasi-Periodic Oscillations (QPOs) in black hole candidates (BH) and, the first (and so far only) accreting millisecond pulsar (see van der Klis 2000 for a review).  RXTE has opened up a completely new field of investigation and it is now widely recognized that fast timing provides one of the most powerful tools for such objectives ({\\em a)} to test General Relativity (GR) in strong gravity fields, ({\\em b)} to constrain the fundamental properties of collapsed stars (EOS of NS, mass and spin of BH).  The RXTE discoveries were made possible by its large X-ray detector area ($0.67$ m$^{2}$, sub-$\\mu$s time resolution and high data rates. The RXTE All Sky Monitor (ASM) and the flexibility of operation and rapid response also played a major role in the success of the mission (Bradt et al.  1993).  For X-ray timing, the signal to noise ratio at which non-coherent variability such as a QPO is detected scales as the source count rate (i.e. the detector area) in the strong source limit (for a photon counting experiment).  Based on this, here we describe how an instrument with {\\em ten times} the detector area of RXTE could help to solve some of the outstanding issues in the astrophysics of collapsed stars. ",
        "conclusions": "The need for a follow-up to RXTE has been clearly identified; fast timing studies are now widely recognized as a powerful tool to study the spacetime around collapsed stars, and to derive constraints on the fundamental properties of neutron stars and black holes.  The studies carried out for EXTRA have demonstrated that a mission using solid-state detector technologies providing an order of magnitude improvement over RXTE can be done within the limits of a modest size mission.  It has also shown that there are existing platforms with resources matching the specific needs of such a mission.  More importantly, the large number of active astronomers that gathered around the proposals demonstrates that there is a very large community supportive of such a mission.  It is a real pleasure to thank all the people who have contributed to the EXTRA proposal (J.L. Atteia, T. Belloni, E. Bravo, A. Brunton, S. Campana, D. Chakrabarti, F. Cotin, D. Dal Fiume, C. Done, E.C. Ford, R.P. Fender, C. Hellier, W. Hermsen, G.L. Israel, P. Jean, P. Kaaret, E. Kuulkers, C. Labanti, S. Larsson, N. Lund, J.E. McClintock, S. Mereghetti, C.M. Miller, C. Motch, J.F. Olive, M. Orlandini, S. Paltani, A. Santangelo, R.A. Sunyaev, A. Zdziarski, P. Zycki), and those of you who have given their scientific or engineering support to EXTRA (A. Bazzano, L. Burderi, M. B\\\"oer, H. Bradt, S. Brandt, C. Castelli, A. Castro-Tirado, M. Cochi, T. Courvoisier, M. Cropper, A.M. Cruise, D. De Martino, C. Eyles, M. Gilfanov, J.M. Hameury, G. Henry, E. Kendziorra, W. Kluzniak, L. Kuiper, J.P. Lasota, F.K. Lamb, G. La Rosa, J. Poutanen, R. Rothschild, R. Svensson, M. Tavani, G. Vedrenne, N.E. White, R. Wijnands, J. Wilms, W. Zhang)."
    },
    "0010/astro-ph0010404_arXiv.txt": {
        "abstract": "A particular kind of quintessence is considered, with equation of motion $p_Q/\\rho_Q = -1$, corresponding to a cosmological term with time-dependence $\\Lambda(t) = \\Lambda(t_0) (R(t_0)/R(t))^{P}$ which we examine initially for $0 \\leq P < 3$. Energy conservation is imposed, as is consistency with big-bang nucleosynthesis, and the range of allowed $P$ is thereby much restricted to $0 \\leq P < 0.2$. The position of the first Doppler peak is computed analytically and the result combined with analysis of high-Z supernovae to find how values of $\\Omega_m$ and $\\Omega_{\\Lambda}$ depend on $P$. Some comparison is made to the CMBFAST public code. ",
        "introduction": " ",
        "conclusions": ""
    },
    "0010/astro-ph0010362_arXiv.txt": {
        "abstract": "The initial results from XMM-Newton observations of the rich cluster of galaxies Abell 1795 are presented. The spatially-resolved X-ray spectra taken by the European Photon Imaging Cameras (EPIC) show a temperature drop at a radius of $\\sim 200$ kpc from the cluster center, indicating that the ICM is cooling. Both the EPIC and the Reflection Grating Spectrometers (RGS) spectra extracted from the cluster center can be described by an isothermal model with a temperature of $\\sim 4$ keV. The volume emission measure of any cool component ($<1$~keV) is less than a few \\% of the hot component at the cluster center. A strong \\ion{O}{viii} Lyman~$\\alpha$ line was detected with the RGS from the cluster core. The O abundance and its ratio to Fe at the cluster center is 0.2--0.5 and 0.5--1.5 times the solar value, respectively. ",
        "introduction": "X-ray observations have revealed that the radiative cooling time of the intra-cluster-medium (ICM) around the central galaxy of many clusters is much less than the cluster age (e.g. Edge, Stewart, and Fabian \\cite{edge}). This indicates that cooling of the ICM could strongly affect the cluster thermal structure (e.g. Fabian \\cite{fabian-cf}). Although pre-Chandra/XMM observations established the presence of plasma near the cluster center considerably cooler than that found further out in the cluster (e.g. Allen et al. ~\\cite{allenetal}; Makishima et al. \\cite{makishima}), the detailed spatial and thermal structure of this cool component, and hence its nature, has not yet been fully understood. In addition the measurement of the elemental abundances in this cooler gas provides critical information concerning its formation and evolution.   Here we report the first results from the XMM (Jansen et al. \\cite{jansen}) observations of a rich cluster of galaxies, Abell 1795 (A1795 for short). This cluster, at a redshift of 0.0631, is one of the best targets for spectral study of the cluster center with XMM. This is because of its peaked X-ray emission and a cool (2--4 keV) component at the cluster center (Briel and Henry \\cite{briel}; Fabian et al. \\cite{fabian}; Fabian et al. \\cite{fabian-chandra}). Based on the the deprojection of the X-ray image a strong cooling flow with a mass deposition rate, $\\dot{M}$, of $\\sim 500$~\\msun$/$yr was reported (Edge et al. \\cite{edge}; Allen et al. \\cite{allenetal}). The ASCA spectra showed a smaller level of the cooling flow with $\\dot{M}$ of $\\sim 250$~\\msun$/$yr (Allen et al. \\cite{allenetal}).   We have measured the temperature structure and metal abundances of the central region based on X-ray spectra with high spectral resolution. Throughout this $Letter$, we assume the Hubble constant to be $H_0 = 50$ km s$^{-1}$Mpc$^{-1}$ and use the 90\\% confidence level unless stated otherwise. One arc-minute corresponds to 110 kpc. ",
        "conclusions": "Based on the XMM observations of Abell 1795, we have obtained the following results.  The temperature and metallicity of the ICM are radially uniform at radii from 2\\arcmin--8\\arcmin\\ ($\\sim 200-800$ kpc), with an observed temperature of 6--7 keV and metallicity of 0.2--0.3 solar. This confirms the previous idea that the cluster is close to being dynamically relaxed (e.g., Briel and Henry \\cite{briel}).  Below a radius of $\\sim 2'$ the effective temperature of the ICM starts to decrease towards the center from 6--7 keV to 3--4 keV. Except inside the central 1\\arcmin\\, the observed temperature profile agrees well with that obtained from a deprojection analysis based on the cooling flow model (Edge et al. \\cite{edge}).  On the contrary, we found lack of emission from cool ($<1-2$~keV) gas in the cluster center, compared to the prediction from the strong cooling flow ($\\dot{M} \\sim 500$~\\msun$/$yr). The EPIC and RGS spectra extracted from the cluster center ($r<30$\\arcsec$-60$\\arcsec) can be described by an isothermal model with a temperature of $\\sim 4$ keV. Furthermore, through the absence of clear \\ion{Fe}{xvii} and \\ion{Fe}{xx} line emission in the RGS spectra, the emission measure of any cool component with a temperature in the range $0.4-1.3$~keV is found to be at least a factor of 3 lower than that would be expected from the strong cooling flow. The absence of the predicted amount of cool emission is also supported by the cut-off temperature of $\\sim$2.4~keV found by fitting cooling flow models to the EPIC data. The absence of these emission lines is difficult to model with a uniformly covering absorber, both covering the hot and cool components, and we derive a covering fraction of $<20$\\% for gas of sufficient column density ($10^{21}$~cm$^{-2}$) necessary to ``hide'' the iron lines. We derived an upper-limit to $\\dot{M}$, of $150$ \\msun$/$yr, which is smaller than found by Allen et al. (\\cite{allenetal}; $250\\pm 70$ ~\\msun$/$yr) based on the ASCA spectral analysis.  The present observation indicates that at temperatures greater than 3 keV the ICM is cooling as predicted by the cooling flow picture. Nevertheless, we have no direct spectral evidence that the gas continues cooling below 2 keV.  This result for A~1795 is similar to what has been found in A~1835 (Peterson et al. \\cite{peterson}) and S\\'ersic 159-03 (Kaastra et al. \\cite{sersic}). A full discussion of possible explanations for the lack of observed cool gas is given by Peterson et al. (\\cite{peterson}) in their analysis of the A~1835 cluster.  We detected the \\ion{O}{viii} Ly~$\\alpha$ line from the cluster center with the RGS. The O abundance was found to be 0.2--0.5 solar value. This is similar to the value obtained in the center of the Virgo cluster (Matsumoto et al. \\cite{matsumoto}) based on ASCA spectra, and in the centers of A~1835 (Peterson et al. \\cite{peterson}) and S\\'ersic 159-03 (Kaastra et al. \\cite{sersic}) based on XMM/RGS spectra. The O to Fe ratio at the cluster center is $0.5-1.5$ times the solar ratio. Since oxygen and iron are mainly produced in type II and Type Ia supernovae, respectively, the observed ratio suggests significant contributions from both type Ia and type II  to the metal enrichment at the cluster center. In the core the Fe abundance is a factor of 2 larger than in the outer parts (see Fig.~\\ref{fig:r-ta}). This is consistent with the general idea that the iron abundance is higher around the cD galaxy (Fukazawa \\cite{fukazawa-phd}; Fukazawa et al. \\cite{fukazawa})."
    },
    "0010/astro-ph0010648_arXiv.txt": {
        "abstract": "The uncovering of the large-scale structure that is hidden in the Zone of Avoidance is essential to our quest for the understanding of the dynamics of the nearby universe.  Dedicated and sophisticated observations of the Zone of Avoidance have to be matched by equally sophisticated theoretical tools for the reconstruction of the large-scale structure from sparse, noisy and very incomplete data. A general approach for the estimation of cosmological parameters and the reconstruction of the density and velocity fields is presented here, based on maximum likelihood analysis, Wiener filtering and constrained realizations.  Radial velocity surveys and the IRAS 1.2\\,Jy redshift survey are used to reconstruct the Zone of Avoidance. ",
        "introduction": "The galactic Zone of Avoidance (ZOA) presents an obstacle to our efforts to understand the dynamics of the `local' universe. Namely, the Milky Way prevents a full-sky mapping of the light, and therefore mass, distribution that induces the local velocity field. Thus, a prime motivation of the astronomical study of the ZOA is to survey the galaxy distribution in the regions obscured by the Milky Way. These efforts have indeed uncovered a very rich structure in the ZOA, in particular in the region of the Great Attractor (GA; Kraan-Korteweg \\& Lahav 2000, and references therein).   In spite of all the efforts and successes in unveiling the ZOA, the fact remains that the ZOA is a region more difficult to observe than other parts of the sky.  It follows that the reconstruction of the large-scale structure (LSS) in the ZOA should rely more on sophisticated statistical methods than other fields of the sky.  Here, we shall report on a Bayesian approach to the reconstruction of the LSS from incomplete, sparse  and noisy data sets, such as peculiar velocities and redshift surveys.  Most of the analysis centers on the recovery of the LSS from noisy surveys of radial velocities with very incomplete sky coverage , with some analysis of the IRAS 1.2 Jy redshift survey that is presented as well.  The main focus of the present paper is on the general methodology with an emphasize on the tools and methods relevant to the reconstruction of the LSS. The method is applied to the MARK III (Willick et al. 1995), SFI (da Costa et al. 1996) and the ENEAR (da Costa et al. 2000) surveys of the radial velocities, as well as the IRAS 1.2\\,Jy redshift survey. Most of the work presented here has been done in collaboration mostly with S. Zaroubi and the cosmographical implications of this work are presented in Zaroubi (these proceedings).  Here, the general Bayesian methodology is presented in \\S 2.  The cosmological parameters estimated from radial velocities surveys are summarized in \\S 3.  The LSS and the ZOA that is reconstructed from radial velocity catalogs and the IRAS 1.2\\,Jy redshift survey are presented in \\S 4.  A new algorithm of recovering the tidal field from the full velocity field enables the studying of the LSS on scales beyond those that are directly probed by the data (\\S 5).  A technique of simulating the dynamical role of the ZOA based on N-body simulations that are constrained by the observed LSS is described in \\S 6.  A general discussion concludes this review. ",
        "conclusions": "The recovery of the LSS in the ZOA poses a challenge to our understanding of the dynamics of our `local' universe.  Direct observational efforts of unveiling the ZOA, sophisticated as they might be, need to be complemented by elaborate statistical method for the reconstruction of the LSS that is hidden in the ZOA. Within the standard cosmological model in which structure emerges from a primordial Gaussian perturbation field {\\it via}  gravitational instability, the Bayesian approach presented here provides   the optimal tools for estimating the cosmological parameters and the reconstruction of the LSS.  The algorithm presented here consists of the following four steps. First, given a data set, the cosmological (and other) parameters are estimated by maximum likelihood analysis over a range of models. Then, the consistency of the most probable model with the data is analyzed by the differential $\\chi^{2}$ analysis.  The model that is found by the maximum likelihood and the goodness-of-fit analysis, serves as the {\\it prior} model for the WF/CR reconstruction. Applying the WF to the data, one gets the mean density and velocity fields, thus providing an estimator of the LSS. The CRs constitute an ensemble of Monte Carlo simulations that are designed to reproduce the imposed constraints, namely the observed radial velocity and/or redshift surveys.  The scatter of the CRs around the WF field provides a measure to the robustness of WF reconstruction.  A detailed study of the cosmography recovered from the radial velocity surveys is given by Zaroubi (these proceedings).  Here, this has been extended to study structure beyond the depth of these surveys.  This is done by the tidal field decomposition of the velocity field.  The tidal field induced by the structure beyond a sphere of radius 6000\\kms\\ has a dipole component of 300\\kms\\ towards (galactic) $(\\ell,b)\\approx (300\\deg,15\\deg)$. This direction implies that a substantial fraction of the structure that induces this dipole is hidden in the ZOA. This provides a further motivation for studying the ZOA."
    },
    "0010/astro-ph0010154_arXiv.txt": {
        "abstract": "To better understand the environment surrounding CO emission clumps in the Keyhole Nebula, we have made images of the region in H$_2$ 1--0 S(1) (2.122 $\\mu$m) emission and polycyclic aromatic hydrocarbon (PAH) emission at 3.29 $\\mu$m. Our results show that the H$_2$ and PAH emission regions are morphologically similar, existing as several clumps, all of which correspond to CO emission clumps and dark optical features. The emission confirms the existence of photodissociation regions (PDRs) on the surface of the clumps. By comparing the velocity range of the CO emission with the optical appearance of the H$_2$ and PAH emission, we present a model of the Keyhole Nebula whereby the most negative velocity clumps are in front of the ionization region, the clumps at intermediate velocities are in it and those which have the least negative velocities are at the far side. It may be that these clumps, which appear to have been swept up from molecular gas by the stellar winds from $\\eta$ Car, are now being over-run by the ionization region and forming PDRs on their surfaces. These clumps comprise the last remnants of the ambient molecular cloud around \\mbox{$\\eta$ Car}. ",
        "introduction": "The Keyhole Nebula is part of the Carina Nebula, a star forming/giant molecular cloud (GMC) complex, containing some of the most massive stars known in our galaxy. It is bathed in the UV radiation field from the nearby open star cluster, Tr 16. This cluster contains numerous O-type stars, including three O3 stars \\cite{Walborn95}. It is at a distance of 2.2 kpc \\cite{Tovmassian95} and is centred on one of the most massive stars known -- $\\eta$ Car, which has a present day mass of about 120 \\Msun\\ \\cite{Davidson97}. The optical image of the Keyhole Nebula taken by \\citeN{Malin93} reveals many interesting features such as dark patches and filaments, the most prominent being in the shape of an old-fashioned key hole, which gave the nebula its name (see Fig.~\\ref{multiwavelength}a). \\citeN{Herschel47} was the first to note this peculiar feature and likened it to the shape of a $``$lemniscata\". Some of the smaller features also have bright rims. This is beautifully illustrated in the recent colour-composite image taken with NASA's Hubble Space Telescope.\\footnote{These data were collected by the Hubble Heritage Team, N. R. Walborn, R. H. Barba and A. Caulet and were released  after the submission date of this article (see \\\\ http://oposite.stsci.edu/pubinfo/pr/2000/06/pr-photos.html).}  The molecular gas in this region is highly inhomogeneous, breaking into several clumps which are all associated with the dark optical features \\cite{Cox951}. The clumps, typically a few tenths of a pc across (10--30 arcsec) and with masses of $\\sim 10$ \\Msun\\, are distributed across a range of velocities (see Table~\\ref{co-clumps}). Some of the molecular clumps correspond to those optical features that have bright rims, suggesting the presence of photodissociation regions (PDRs) at the interface with the surrounding ionized gas. PDRs are regions where the far-UV radiation field dominates the heating and chemistry of the gas. They begin at the sharp ionized/neutral interface at the edge of molecular clouds and then extend well into the molecular gas (see \\citeNP{Tielens93}).   \\begin{figure*} \\centering \\mbox{\\subfigure[Three-colour optical emission image, courtesy of David Malin.  This image was produced by combining 3 exposures (of 15 mins) each taken using blue, green and red filters. The Keyhole is the prominent dark band bisecting the image.]{\\psfig{file=fig1a.ps,width=0.56\\textwidth}}\\quad} \\mbox{\\subfigure[H$_2$  2.122 $\\mu$m line emission obtained with UNSWIRF. This image is continuum subtracted (aside from the circular residuals from $\\eta$ Car).]{\\psfig{file=fig1b.ps,width=0.48\\textwidth}}\\quad \\subfigure[PAH 3.29 $\\mu$m emission obtained with SPIREX. This image includes continuum sources (the diffraction spikes from $\\eta$ Car are artifacts).]{\\psfig{bbllx=71pt,bblly=163pt,bburx=361pt,bbury=437pt,file=fig1c.ps,clip=,width=0.48\\textwidth}}} \\caption{\\label{multiwavelength}A collection of multi-wavelength images of the Keyhole Nebula. Features of interest are labelled in the optical and H$_2$ 1--0 S(1) images and are described further in Tables~\\ref{co-clumps} and ~\\ref{h2pah-table}.} \\end{figure*}  To better understand the environment surrounding these clumps we have imaged the Keyhole region at two wavelengths suitable for detecting emission from PDRs. We have used the University of New South Wales Infrared Fabry-Perot (UNSWIRF) to image H$_2$ 1--0 S(1) (2.122 $\\mu$m) emission and the SPIREX/Abu thermal infrared camera at the South Pole to image narrow-band emission at 3.29 $\\mu$m. This wavelength includes one of the more prominent `unidentified' infrared (UIR) emission bands and is generally attributed to the \\mbox{C--H} stretching mode of free polycyclic aromatic hydrocarbon molecules (PAHs) (\\citeNP{Leger84}, \\citeNP{Allamandola85}, \\citeNP{Geballe94}, \\citeNP{Joblin95}).  H$_2$ 1--0 S(1) emission in molecular clouds can be attributed to collisional excitation in the presence of shocks or to UV fluorescence in PDRs (see \\citeNP{Burton92}). In dense gas, $ n \\geq n_{crit} \\sim 10^5$ cm$^{-3}$ where $n$ is the hydrogen nucleus density, UV excitation levels are collisionally redistributed leading to an increase in the population of the $\\nu=1$ level. This subsequently increases the intensity of the $\\nu$=1--0 S(1) line (see \\citeNP{Allen99}). Previous observations of PDRs indicate that 3.29 $\\mu$m PAH emission zones are situated near the very edge of molecular clouds, peaking at the ionization front, while H$_2$ 1--0 S(1) emission zones extend further into the clouds by about 2 magnitudes of visible extinction (e.g. \\citeNP{Sloan97}, \\citeNP{Tielens93}, \\citeNP{Burton89}). So while H$_2$ 1--0 S(1) emission cannot be used alone to identify PDRs, if found to correspond to another PDR tracer (such as PAH emission) it can pinpoint regions of molecular gas that are high in density and exposed to the surrounding far-UV radiation field.  \\begin{table} \\begin{minipage}{\\columnwidth} \\caption{\\label{co-clumps}A list of the $^{12}$CO(2--1) emission clumps measured by Cox \\& Bronfman (1995). The mass measurements may underestimate the true mass of each clump by up to a factor of 3. The type of dark optical feature each clump is associated with is also listed. The velocities are given with respect to the local standard of rest (LSR).} \\begin{tabular}{@{}llll}  \\hline Clump\t  \t& Velocity \t&Mass \t \t&Optical\\\\ &(\\kms)   \t&(M$_{\\sun}$) \t&Feature\\\\ \\hline  C \t&$-7.6$\t\t&6\t&  Faint\\\\ B \t&$-8.5$\t\t&3\t&  Faint\t\\\\ A1\\footnote{Also known as the `Kangaroo Nebula'.}  &$-17.4$  &14  & Bright rim      \\\\ A2 \t& --\t\t& --\t& Bright rim\t\\\\ 4 \t&$-25.0$\t\t&6\t& Bright rim\t\\\\ 5 \t&$-25.1$\t\t&1\t& Bright rim\t\\\\ 3 \t&$-28.3$\t\t&4 \t& Keyhole (dark) \\\\ 1 \t&$-31.9$\t\t&17 \t& Keyhole (dark) \\\\ 2 \t&$-32.0$\t\t&11\t& Keyhole (dark) \\\\ \\hline  \\end{tabular} \\end{minipage} \\end{table} ",
        "conclusions": "We have imaged the Keyhole region of the Carina Nebula in the H$_2$ 1-0 S(1) line at 2.122 $\\mu$m and in the PAH emission feature at 3.29 $\\mu$m, and compared the results to optical, radio continuum and \\CO images of the region. The H$_2$ and PAH images are remarkably similar, indicating they are fluorescently excited, and show a series of emission clumps typically 30--60 arcsec in extent distributed around the edges of the radio emission. When compared to the optical and CO images three categories of clumps are apparent: (i) dark regions of optical obscuration with CO but no H$_2$ or PAH emission (clumps 1, 2, and 3 which form the Keyhole feature), (ii) faint, optically dark clumps with bright rims and H$_2$ and PAH emission (clumps A1, A2, 4 and 5) and (iii) faint optically dark clumps with H$_2$ and PAH emission (clumps B and C).  The clumps are typically of mass \\mbox{10 \\Msun} and appear to have been swept up from ambient cloud material by a wind from $\\eta$ Car. They have been driven in all directions away from $\\eta$ Car, and must now be coasting at speeds of \\mbox{$\\sim$ 10 \\kms} as there is no evidence of any shocked emission. The prominent dark clumps of the Keyhole are coming towards us and are outside the \\HII\\ region. The bright-rimmed clumps A1, A2, 4 and 5 are within the \\HII\\ region and thus are being heated by the radiation field. Clumps such as A1 are seen face-on whereas clumps 4 and 5 are seen edge-on. Finally the faint dark clumps B and C are on the far side of the \\HII\\ region and have been driven away from $\\eta$ Car in the opposite direction to ourselves. The clumps comprise the last remnants of the ambient molecular cloud from which $\\eta$ Car formed."
    },
    "0010/astro-ph0010012_arXiv.txt": {
        "abstract": "Within the framework of the current models for the $slow$ neutron capture ($s$) process in Asymptotic Giant Branch (AGB) stars we explore the uncertainties introduced by the assumptions made on stellar modeling. On the basis of star models constructed with three different evolutionary codes we generate detailed neutron-capture nucleosynthesis post-processing models. The main difference among the codes is that one of them includes an overshooting mechanism. As a result, the neutron fluxes are stronger both during the interpulse periods and within thermal pulses. The efficiency of the \\ctb source is also studied and we find that a linear relationship exists between the initial \\cdb in the intershell and the maximum number of neutrons produced. ",
        "introduction": "The $s$ process had been suggested in the 1950s to account for the nucleosynthesis of elements heavier than Fe \\cite{burb}. The $s$-process path of neutron capture generally follows the valley of $\\beta$ stability, however, for some values of neutron density and temperature, $branchings$ are open and nuclei outside the valley of stability can be produced. It can be shown that in equilibrium conditions the $s$-process distribution follows the rule $\\sigma_A N_A = const$, where $\\sigma_A$ is the neutron capture cross section of the isotope $A$ and $N_A$ its abundance in number. The rule is valid locally, far from neutron magic nuclei which act as bottlenecks of the flux and produce discontinuities in the distribution. An important parameter is the neutron exposure, $\\tau = \\int^t_0 N_{\\rm n} v_{th} dt$, where $v_{th}$ is the thermal velocity of the gas. The neutron exposure determines the final distribution of heavy elements; for higher neutron exposures heavier elements are produced \\cite{clayton68}.  The major astrophysical site for the $s$ process has been recognised to be the deep layers of AGB stars, which are low to intermediate mass stars in a late phase of their evolution. These stars experience recurrent convective instabilities in the region between the H- and He-burning shells (intershell) as a consequence of thermonuclear runaways of the He-burning shell ($thermal$ $pulses$, TP). At the end of a TP, the convective envelope sinks into the intershell and dredges up material to the surface ($third$ $dredge$ $up$, TDU). A large amount of \\nvdb is present in the convective TPs as a product of the chain \\nqagb starting on the abundant \\nqb from the H-burning ashes. However, the temperature in the TPs of low mass stars ($M \\leq 4$ \\ms) is not high enough to significantly activate the \\nvdanb reaction. The \\ctb neutron source is activated at lower temperatures ($\\sim 0.8 \\times 10^{8}$ K), however, an amount of \\ctb higher than that present in the H-burning ashes is needed to reproduce the observed enhancements of heavy elements. To enable the formation of a region rich in \\ct, the \\ctb $pocket$, some protons from the envelope must enter the \\cd-rich intershell. A favorable location for the occurrence of the mixing is the sharp H/He discontinuity which is left over after TDU (see Fig. 1), however, no standard stellar models have found the penetration of protons to occur. Recently, models including time-dependent overshoot \\cite{herwig97}, motivated by hydrodynamical simulations (e.g.\\cite{singh98}), and models with rotation \\cite{langer99} have been able to create a proton-rich layer at the top of the intershell, after the end of the TDU. After less than a few thousand years \\ctb and \\nqb are formed in the region. Before the end of the interpulse period all \\ctb burns ($\\alpha$,n) and the region becomes enriched in \\s-processed material. At the end of the interpulse the pocket is engulfed by the following TP and is thus mixed with material from the previous TP and the ashes from the H-burning shell. Inside the TP, if the temperature at the bottom of the shell reaches 2.5 $\\times 10^{8}$ K, the \\nvdanb reaction is marginally activated and a second neutron flux can occur. After the TP the $s$-process-rich material from the intershell is dredged up to the envelope by the following TDU. This cycle is repeated over all TPs with TDU and the composition of the envelope throughout the AGB phase is changed by TDU and mass loss effects.  Predictions from $s$-process models have to be compared with observational constraints from spectroscopical observations of chemically peculiar stars (see e.g. \\cite{busso98,busso00}) and with laboratory measurements of isotopic anomalies in silicon carbide (SiC) grains recovered from meteorites, and believed to be originated in AGB stars \\cite{gallino97}. Stellar models, together with Galactic Chemical Evolution models, should also reproduce the solar distribution of $s$-process elements \\cite{trava00}. For an extensive review on the $s$ process in AGB stars see \\cite{busso99}. ",
        "conclusions": "\\subsection{The two neutron bursts}  \\begin{figure}[htb] \\begin{minipage}[t]{80mm} \\includegraphics[width=79mm,height=79mm]{figure2NIC.eps} \\caption{\\footnotesize Resulting \\cd, \\ctb and \\nqb profiles in the pocket as a function of the initial number of protons introduced below the H/He discontinuity, after the end of the 10$^{\\rm th}$ TDU (see Fig. 1). Also plotted is the total neutron exposure profile as computed at the end of the interpulse period after all the \\ctb has burnt.} \\end{minipage} \\hspace{\\fill} \\begin{minipage}[t]{75mm} \\includegraphics[width=79mm,height=79mm]{figure3NIC.eps} \\caption{\\footnotesize Maximum neutron exposure in the pocket as a function of the initial \\cd. Squares, hexagons and triangles are results obtained for the interpulse periods following the 1$^{\\rm st}$, 10$^{\\rm th}$ and 17$^{\\rm th}$ TDU respectively. Full symbols indicate the X(\\cd) value as computed by the MSSSP at such interpulses.} \\label{fig:toosmall} \\end{minipage} \\end{figure}  The resulting \\cd, \\ctb and \\nqb profiles in the pocket as a function of the initial X$_p$ are presented in Fig. 2, as computed with the MOSN code. The total neutron exposure profile obtained when all \\ctb has burnt is also plotted. Since the \\nq(n,p)$^{14}$C reaction has a relatively high cross section \\cite{gledenov95}, nuclei of \\nqb are a strong neutron poison during the $s$ process. Some of the protons produced make more \\ctb when recaptured by the \\cdb (recycle effect), others destroy the \\ctb producing more \\nq. As a consequence, the total neutron exposure $\\tau$ grows, together with the \\ctb profile, up to a maximum of 0.43 mbarn$^{-1}$, and then decreases when the initial abundance of \\nqb nuclei is comparable or higher than that of \\ctb nuclei. This result confirms that the maximum value of $\\tau$ in the pocket chosen by \\cite{gallino98} (see their Fig. 3 and 6) is typical for stars of solar metallicity when X(\\cd) $\\sim$ 0.2 in the intershell. We also made some tests changing the initial amount of \\cd, and for different interpulse periods. As shown in Fig. 3 a nearly linear relationship exists between the final maximum value of $\\tau$ in the pocket and the initial \\cd, because the latter determines the \\ctb and \\nqb profiles and how many protons will be recycled. We note that the \\cdb in the intershell varies from pulse to pulse. This linear relationship suffers from some uncertainties. While the temperature at the H/He interface at the end of TDU is almost constant ($\\sim 10^7$ K) for different interpulse periods, the temperature gradient towards deeper layers is much steeper for advanced TDU. The temperature at which \\ctb and \\nqb are formed in our calculations, at a mass $\\simeq 3 \\times 10^{-4}$ \\msb below the discontinuity, is $\\sim$ 1.5, 2.2 and 2.5 $\\times 10^{7}$ K after the 1$^{\\rm st}$, the 10$^{\\rm th}$ and the 17$^{\\rm th}$ TDU respectively. The \\cd, \\ctb and \\nqb profiles can be slightly different and, as shown in Fig. 3, the $\\tau_{max}$ versus \\cdb relationship shows some spread. The smaller the extent of the pocket, the less the change of temperature gradient will effect the results. Future work includes more tests with the MOSN code, and $s$-process calculations with the TOSP code, using a different $\\tau_{max}$ for different interpulse periods.  As for the EVOL code, because of IDU, the initial mass fraction of \\cdb is 0.48, and the resulting abundance profiles produce a $\\tau_{max}$ of 0.88 mbarn$^{-1}$. The discrepancy with the maximum value of $\\tau$ we found for the same initial X(\\cd) with the test profiles, computed with the MOSN code, is possibly due to a slightly lower temperature at which the pocket has been computed, because of the smaller extent in mass, and mostly to the different choice of the \\ct(p,$\\gamma$)\\nqb reaction rate. In the MOSN computation the latest rate by \\cite{king94} has been adopted, rather than the previous \\cite{cf88}, which is 30\\% slower.  \\begin{figure}[t] \\includegraphics[width=16cm,height=9cm]{figure4NIC.eps} \\caption{\\footnotesize The two neutron bursts as described with the different codes: temperature and neutron density as computed by the TOSP code in one of the zones that forms the \\ctb pocket for a typical interpulse period (panel $a$ and $c$) and at the bottom of a typical TP (panels $b$ and $d$). The zero points in time represent: in panels $a$ and $c$ the time from the start of the interpulse period (about 10,000 yr) when $T = 0.8 \\times 10^8$ K; in panels $b$ and $d$ the time when $T_{bottom} = 2.5 \\times 10^8$ K. Note the large differences in time scale, total $\\tau$ and neutron-density peak between the two neutron bursts.} \\end{figure}  To compare the neutron fluxes in the interpulse period and in the convective pulse, we plot in Fig. 4 the temperatures and neutron densities for the two neutron bursts, as computed by the TOSP code on the basis of the three different stellar evolution calculations. This comparison is only of qualitative character because we did not match the same TP number, core mass and envelope mass simultaneously for the three codes. However, the main finding is that the features of \\ctb burning are fairly comparable in all cases (left panels in Fig. 4). All \\ctb burns before the onset of the next TP because the temperature reaches $10^8$ K (panel $a$). This is a very typical situation, however, it has to be noted and taken into account in future work that this might not occur during the earliest interpulse periods (see e.g. Fig. 4 of \\cite{herwig00}). While the total neutron exposure is determined only by the initial amount of \\ctb in any radiative layer of the pocket, the maximum neutron density can vary (panel $c$). However, it is so low that no branchings are activated. In the temperature range of interest the \\ctanb rate we use is about 50$-$70\\% lower than the one adopted in \\cite{nacre}, however, this should not make much difference in the overall result. If \\ctb burns faster the maximum neutron density will be higher, yet still unimportant. In the TPs, the temperature profile has a narrow peak (panel $b$), and so has the neutron density (panel $d$). The neutron exposure is unimportant compared to that in the radiative layers, however, branching points are extremely sensitive to this neutron flux. Main differences show up between the FRANEC/MSSSP and the EVOL case. In the EVOL code, because of IDU, the temperature, and consequently the neutron density, stay high for a longer time and branchings are more active.  \\subsection{$s$-process results}  We computed \\s-process models with the TOSP code over a total of 25 TPs with TDU for FRANEC, 18 for MSSSP and 13 for EVOL. In Table 1 the results for the envelope enrichment of \\s-process elements are compared with spectroscopic observations of AGB stars and their relatives. We consider the parameters `ls', which is the average of light \\s-process elements (Y and Zr) and `hs', which is the average of heavy \\s-process elements (Ba, La, Nd and Sm). The ratios of these parameters with Fe, together with the ratios [hs/ls] and [Pb/Ba] are good indicators of the overall distribution of heavy elements, which is determined by the main neutron exposure from the \\ctb pocket. Predictions are given for the last TP, initial values are solar (i.e. zero). To compare with measurements in SiC grains we consider three isotopic ratios: $^{138}$Ba/$^{136}$Ba, $^{88}$Sr/$^{86}$Sr and $^{96}$Zr/$^{94}$Zr. The first two involve $^{138}$Ba and $^{88}$Sr, which are neutron magic nuclei, and are thus sensitive to the neutron exposure in the \\ctb pocket, while the third ratio involves $^{96}$Zr, which is produced through a branching at $^{95}$Zr, thus it is sensitive to the neutron density in the TPs. The ranges of prediction are given for the last phase of the evolution when the condition for SiC grains to form, C/O$>$1, is satisfied in the envelope (last 5 TPs for the FRANEC, 4 for MSSSP and 7 for EVOL).  \\begin{table}[htb] \\caption{Results from the three computations as compared with some observational constraints.} \\label{table:1} \\newcommand{\\cc}[1]{\\multicolumn{1}{c}{#1}} \\renewcommand{\\tabcolsep}{1.1pc} % \\renewcommand{\\arraystretch}{1.2} % \\begin{tabular}{@{}ccccc} \\hline parameter & measured & FRANEC & MSSSP & EVOL \\\\ \\hline $[$ls/Fe$]^a$ & $0. \\rightarrow 1.5^b$ & 0.94 & 0.95 & 0.52 \\\\ $[$hs/Fe$]^a$ & $0. \\rightarrow 1.5^b$ & 0.55 & 0.52 & 0.84 \\\\ $[$hs/ls$]^a$ & $-0.8 \\rightarrow 0.5^b$ & $-$0.40 & $-$0.43 & 0.35 \\\\ $[$Pb/Ba$]^a$ & $-$ & $-$0.52 & $-$0.53 & 0.06 \\\\ $\\delta(^{138}$Ba$/^{136}$Ba)$^c$ & $\\sim -250^d$ & $-289 \\rightarrow -272$ & $-331 \\rightarrow -332$ & $398 \\rightarrow 647$ \\\\ $\\delta(^{88}$Sr$/^{86}$Sr)$^c$ & 51$ \\pm 103^e$ & $ 208\\rightarrow 294$ & $20 \\rightarrow 149$ & $158 \\rightarrow -293$ \\\\ $\\delta(^{96}$Zr$/^{94}$Zr)$^c$ & $-686 \\pm 187^f$ & $-505 \\rightarrow -372$ & $-718 \\rightarrow -610$ & $46 \\rightarrow 1120$ \\\\ \\hline \\end{tabular}\\\\[2pt] {\\footnotesize $^a$[X/Y]=log[(X/Y)$_{meas}$/(X/Y)$_{\\odot}]$.} {\\footnotesize $^b$Range of values measured in stars of $Z \\sim Z_{\\odot}$ from \\cite{busso98}, typical 2$\\sigma$ error $\\pm 0.25$.} {\\footnotesize $^c \\delta$(x/y)=[(x/y)$_{meas}$/(x/y)$_{\\odot} -$ 1] $\\times$ 1000.} {\\footnotesize $^d$Bulk SiC grains from \\cite{prombo93}.} {\\footnotesize $^e$Average of single grains from \\cite{nicolussi98}.} {\\footnotesize $^f$Average of single grains from \\cite{nicolussi97}.} \\end{table}  Since the results from the FRANEC and MSSSP computations, which use the same \\ctb pocket, are very similar, we can conclude that the important uncertainty in \\s-process predictions is the \\ctb pocket, not the intrinsic differences among different codes (with no overshooting mechanism included). In the EVOL code, the neutron exposure in the pocket is higher and the production of the heavier \\s-process elements, Pb and the magic $^{138}$Ba are favored. The \\s-process element enhancements with respect to Fe, of about an order of magnitude in AGB star envelopes, are reproduced by the models. However, due to the uncertainties of the spectroscopic measurements on the observational side, and of the \\ctb pocket on the theoretical side, we cannot draw strong conclusions about which of the models match the observations better. The ratio $^{138}$Ba/$^{136}$Ba measured in bulk SiC grains indicates that, on average, the maximum neutron exposure in AGB stars should not exceed $\\sim$ 0.4 mbarn$^{-1}$. The ratio $^{96}$Zr/$^{94}$Zr is matched when the neutron density in the TPs is not higher than about $5 \\times 10^8$ n/cm$^3$ for a long time, a condition not satisfied by the EVOL model (see Fig. 4), which produces $^{96}$Zr in a significant amount.  The same MSSSP \\s-process calculation was performed with a different \\ctb profile, giving more of the total mass of the pocket to the zone with higher $\\tau$ ($\\sim 0.4$ mbarn$^{-1}$). Previously, as in \\cite{gallino98}, greater mass was given to the region with lower $\\tau$ ($\\sim 0.15$ mbarn$^{-1}$). We found that [ls/Fe] and [hs/Fe] reached 1.27 and 1.0 respectively, while [hs/ls] and [Pb/Ba] kept to $-$0.27 and $-$0.48. The $\\delta$($^{138}$Ba/$^{136}$Ba) did not change much: $-382 \\rightarrow -368$. The $\\delta$($^{88}$Sr/$^{86}$Sr), however, increased to $447 \\rightarrow 651$, going outside the measured range and supporting the profile weighted on the lower $\\tau$.  \\vspace{2 mm} {\\footnotesize {\\bf Acknowledgements.} We wish to thank Roberto Gallino, John Lattanzio, Maurizio Busso and Onno Pols for discussions and inspiration, Robert Cannon for coding and Brett Hennig for help with graphics and proof reading.}"
    },
    "0010/astro-ph0010538_arXiv.txt": {
        "abstract": "Using all available multi-component radio pulse profiles for pulsars with medium to long periods and good polarisation data, we have constructed a two-dimensional image of the mean radio beamshape. This shows a peak near the centre of the beam but is otherwise relatively uniform with only mild enhancements in a few regions. This result supports the patchy-beam model for emission beams in which the mean beam shape represents the properties of the emission mechanism and observed pulse components result from emission sources distributed randomly across the beam. ",
        "introduction": "Pulsars are generally believed to be rotating neutron stars in which the observed pulses result from one or more emission beams sweeping across the Earth as the star rotates. Observations of radio polarisation \\cite{rc69a} led to the magnetic-pole model in which the emission beam was centred on the magnetic axis of a predominantly dipole magnetic field. On the assumption that the emission is directed radially, the observed pulse profile reflects the variations in emission intensity along a line of constant rotational latitude. Although more than 1000 pulsars are now known \\cite{tml93,lcm+00} the shape of pulsar radio beams remains controversial.  There are two main areas of uncertainty. One concerns the outline shape of the radio beam. Originally assumed to be circular, some investigations (e.g. Narayan \\& Vivekanand 1983)\\nocite{nv83} argued for an elliptical beam extend in the latitude direction with a large axial ratio. In contrast, Biggs (1990)\\nocite{big90b} suggested that the beam was slightly compressed in the latitudinal direction. Most other investigations (e.g. Lyne \\& Manchester 1988; Bj\\\"ornsson 1998; Gil \\& Han 1996)\\nocite{lm88,bjo98,gh96} have concluded that the emission beam is essentially circular, and we will assume this in the present investigation.  The second area of uncertainty concerns the form of the beam pattern. Early observations showed that there are two or more pulse components in many pulsars (e.g. Lyne, Smith \\& Graham 1971; Manchester 1971)\\nocite{lsg71,man71b}.  Ruderman \\& Sutherland (1975)\\nocite{rs75} presented a detailed model for pulsar radio emission in which the beam had the shape of a hollow cone, more intense around the periphery, corresponding to the last open field lines emanating from the polar cap region. The radio emission was attributed to curvature radiation by positrons moving these field lines. Such radiation is linearly polarised in the plane of curvature of the magnetic field and hence the model naturally explained the smooth sweep of polarisation position angle seen in many pulsars, as well as the common occurrence of double-peaked pulse profiles.  Backer (1976)\\nocite{bac76} first discussed the more-or-less central pulse component seen in many pulsars and suggested that it resulted from an axial beam.  Rankin (1983)\\nocite{ran83a} made the distinction between this central or `core' emission and the outer or `conal' emission, and showed that these two components of the profile had rather different properties. At least in short-period pulsars, core emission has a steeper spectrum than conal emission \\cite{ran83a,lm88} and the fluctuation properties of the two types of emission are very different \\cite{ran86}.  This model was subsequently extended to have two or more coaxial conal emission zones, either to account for multiple-component profiles \\cite{gk97,ql98} or for the appearance of components at different apparent radii from the conal axis \\cite{ran93,kwj+94,md99}. These latter analyses depend on a knowledge of the angle between the beam and rotation axes ($\\alpha$), usually computed from the observed width of a `core' component. Unfortunately, the determination of this angle is very model dependent and this results in large uncertainties in the derived radii.  Based on a large sample of pulse shape and polarisation data, Lyne \\& Manchester (1988) found that pulse components in multiple-component profiles were not symmetrically located within the beam boundary. Furthermore, they were generally of very different intensities and, in some cases, missing altogether. Their results were consistent with components being randomly located within the beam boundary, leading to the `patchy-beam' model. Gould (1994)\\nocite{gou94} and Gould, Lyne \\& Smith (2000)\\nocite{gls00} confirmed these conclusions with an even larger data set.  \\input ma348_tab.tex  Manchester (1995)\\nocite{man95b} suggested an interpretation of these results in which the observed pulse profile is the product of a `window function', which is a function of pulse period and radio frequency but common to all pulsars, and a `source function' which determines the strength of the emission at a given point within the beam and is different for every pulsar. For example, the source function may be determined by the density or energy of the plasma beams along different field lines, whereas the window function is determined by the properties of the emission mechanism.   In this paper, we determine the two-dimensional shape of the window function at frequencies around 1 GHz by averaging the observed pulse profiles of multi-component pulsars having reliable polarisation data. This analysis depends only on the observed width of the pulse profile and the `impact parameter', that is the minimum angle between the line of sight and the beam axis, usually given the symbol $\\beta$. In contrast to the beam inclination angle $\\alpha$, the impact parameter is normally well determined from the maximum rate of change of position angle at the profile centre \\cite{lm88}. We assume that emission is seen across the entire polar cap and choose only pulsars with profiles where this appears to be the case. In Section 2 we describe the data set and our methods of determining the two-dimensional beam pattern. Results are presented and discussed in Section 3 and Section 4 summarises our conclusions. ",
        "conclusions": "We have computed the average radio beamshape at frequencies about 1 GHz of pulsars with medium to long periods. This beamshape has a peak near its centre and a mild, broad and rather irregular enhancement at a normalised beam radius of about 0.7, but is otherwise rather uniform.  The decline at the beam edge is gradual.  These results suggest that the presence and location of profile components are determined by a `source function' which varies randomly from pulsar to pulsar. The summing of these randomly distributed components results in a relatively uniform average beam profile which we interpret as the `window function' representing the properties of the emission process common to all longer-period pulsars."
    },
    "0010/astro-ph0010006_arXiv.txt": {
        "abstract": "Numerical simulations of cluster mergers reveal many characteristics of the merging process: shock structure and strength, observational signatures of the dynamical state, effects on the mass determination, turbulence and the evolution of the X-ray luminosity and the magnetic field. In this article I review the results obtained from various simulations over the last years. ",
        "introduction": "Irregular cluster morphologies in many X-ray images of cluster as well as indications from optical observations imply that many clusters are not relaxed. Hence merging is a common phenomenon in clusters of galaxies. Such mergers of subclusters are very energetic events, which affect clusters strongly, e.g. shocks emerge which are important because they are the major heating source for the intra-cluster gas and moreover they can accelerate particles.  While observations provide only snapshots of the different evolutionary stages of mergers, simulations are the only way to follow the evolution. The models reveal where shocks emerge, where they move and how strong they are. With the help of simulated X-ray images at different stages of a merger a connection between X-ray morphology and the dynamical state can be established and the effects on mass determination can be investigated. Magnetohydrodynamic simulations show how the magnetic field is enhanced during mergers and how the overall structure of the magnetic field is changed due to a merger. ",
        "conclusions": ""
    },
    "0010/astro-ph0010230_arXiv.txt": {
        "abstract": "This paper presents an analysis of a Keck HIRES spectrum of the QSO FIRST~J104459.6+365605, covering the rest wavelength range from 2260 to 2900 \\AA.  The line of sight towards the QSO contains two clusters of outflowing clouds that give rise to broad blue shifted absorption lines. The outflow velocities of the clouds range from --200 to --1200 \\kms\\ and from --3400 to --5200 \\kms, respectively. The width of the individual absorption lines ranges from 50 to more than 1000 \\kms.  The most prominent absorption lines are those of \\mgii, \\mgi, and \\feii, and \\mnii\\ is also present. The low ionization absorption lines occur at the same velocities as the most saturated \\mgii\\ lines, showing that the \\feii\\ , \\mgi\\ and \\mgii\\ line forming regions must be closely associated. Many absorption lines from excited states of \\feii\\ are present, allowing a determination of the population of several low lying energy levels.  The populations of the excited levels are found to be considerably smaller than expected for LTE, and imply an electron density in the \\feii\\ line forming regions of $n_e \\sim 4 \\times 10^3$ cm$^{-3}$.  Modelling the ionization state of the absorbing gas with this value of the electron density as a constraint, we find that the distance between the \\feii\\ and \\mgi\\ line forming region and the continuum source is $ \\sim 7 \\times 10^2$ parsec. From the correspondence in velocity between the \\feii\\ , \\mgi\\ and \\mgii\\ lines we infer that the \\mgii\\ lines must be formed at the same distance.  The \\mgii\\ absorption fulfills the criteria for Broad Absorption Lines defined by Weymann et al. (1991). Therefore the distance we find between the \\mgii\\ line forming region and the continuum source is surprising, since BALs are generally thought to be formed in outflows at a much smaller distance from the nucleus. ",
        "introduction": "Many QSO rest-frame UV spectra exhibit blue shifted absorption lines that are indicative of an outflow. The properties of these absorption lines vary over a wide range, from very broad ($>$ 10,000 \\kms) absorption lines, usually observed in high ionization species like \\civ\\ and \\ovi\\ to fairly narrow ($\\sim$ 30 \\kms) lines from the same high ionization species, or from low-ionization species like \\feii\\ and \\cii . The absorption lines are often divided in two classes. The first class contains the Broad Absorption Lines (BALs)(Weymann, Turnshek \\& Christiansen 1985, Turnshek 1988) which are broad absorption troughs, and can span a large range of outflow velocities up to several times $10^4$ \\kms\\ . The second class refers to Associated Absorption Lines (AALs) (Weymann et al. 1979, Foltz et al. 1986), which have individual absorption components less than a few hundred \\kms\\ wide, and span a range of outflow velocities up to several thousand \\kms .  It is not obvious whether this division reflects a real bimodal distribution of absorption line properties caused by two different underlying formation mechanisms. There are many examples of QSOs that exhibit both types of lines simultaneously, or have properties that are a mixture between the two classes, eg. high velocity narrow lines or low velocity broad lines.  It is generally thought that BALs are formed relatively close to the galactic nucleus ($\\ltorder 1$ parsec), although the direct observational evidence for this is not very strong.  The minimum distance separating the region in which broad absorption lines arise (the BALR) from the source of the continuum radiation is set by the observation that the Broad Emission Lines are often partly or completely absorbed by the outflow (Turnshek 1988).  This places the BALR outside the Broad Emission Line Region (BELR), the size of which is now relatively well established with reverberation mapping techniques at $\\sim 0.1$ pc for an average luminosity QSO.  The maximum distance of the BALR from the central source is more difficult to constrain. Adopting typical, if uncertain, physical parameters for the BAL clouds it can easily be shown that unrealistically large mass and energy loss rates are implied if the lines are formed more than a few tens of parsecs from the central engine. Further support for the conjecture that the BALR is not much larger than the BELR comes from consideration of the acceleration mechanisms, which generally require that the flow starts at a distance comparable to the size of the BELR in order to reach the high speeds observed (e.g. Arav \\& Li 1994).  The physical conditions inside the BALR are difficult to constrain because of the extremely wide and deep absorption lines in classical BALQSOs. The width complicates the analysis because the optical depth at a certain wavelength is the convolution of the column density as a function of velocity and the multiplet structure of the line, so that the column density as a function of velocity can only be derived by a deconvolution.  The large depth makes the derived column densities very sensitive to partial covering of the continuum light source.  AALs appear to be more diverse, and could arise from gas located in the AGN environment, the host galaxy, or intervening gas with a smaller cosmological redshift than the QSO. Some recent studies of AALs are Hamann, Netzer \\& Shields (2000), Petitjean \\& Srianand (1999), Srianand \\& Petitjean (2000), Barlow \\& Sargent (1997) and Ganguly et al. (1999).  The analysis of AALs does not suffer as much from the problems of blending and saturation encounterd with BALs, and can give quite detailed results on the physical conditions and geometry of the absorbing gas, as demonstrated in the references above. From these studies it has become clear that the great majority of AALs show the effects of partial covering in their observed multiplet line ratios, (e.g. Hamann et al. 1997, Arav et al. 1999) which argues against a cosmological origin, and is most easily explained if the lines are formed in the AGN environment. However, to our knowledge no firm limits on the distance of the absorbers to the active nucleus have been derived so far.  Here we investigate the z=0.700 QSO FIRST~J104459.6+365605 (hereafter FBQS1044+3656).  This object was discovered in the FIRST Bright Quasar Survey (White et al 2000) which consists of 636 radio-selected quasars distributed over $\\sim 2700$ square degrees.  The absorption lines in the spectrum of this QSO have several unique features, allowing us to constrain the density and ionization parameter in the outflow.  The spectrum contains absorption lines of \\mgi\\ , \\mgii\\ and \\feii . Some of the \\feii\\ lines arise from low-lying excited states. Most of the \\feii\\ lines are not very wide nor heavily saturated, behaving roughly as expected from their relative oscillator strength, making quantitive estimates of the column density possible.  The derived relative populations of several low-lying energy levels of \\feii\\ allow us to put significant constraints on the physical conditions in the absorbing clouds.  The characteristic velocity and width of the absorption lines in FBQS1044+3656 are intermediate between BALs and AALs.  The \\mgii\\ absorption is very broad, and just fulfills the strict criterion applied to the \\civ\\ line by Weymann et al. 1991 to decide whether a QSO is a BALQSO. The lower ionization lines of \\feii\\ and \\mgi\\ are narrower, and more characteristic of AALs. However, we will argue that the excellent correspondence in velocity between the strongest \\mgii\\ components and the \\mgi\\ and \\feii\\ lines leave little doubt that they arise from the same outflowing cloud complex.  Our analysis is similar to the study of BALQSO Q0059-2735 by Wampler, Chugai \\& Petitjean 1995 (hereafter referred to as WCP). Q0059-2735 is a classical BALQSO with very deep and wide BAL troughs in both high and low ionization species.  Its spectrum also contains several narrow, low ionization absorption line systems, such as \\feii, \\Niii, and \\mnii, at velocities of a few hundred \\kms. Another recent study with some similarities to ours considers narrow absorption line systems in the BALQSO APM 08279+5255 (Srianand \\& Petitjean 2000), which show absorption from excited states of \\cii\\ and \\siII\\ . ",
        "conclusions": "\\subsection{The electron density in the outflow}  Our most important result is the finding that absorption from excited levels of \\feii\\ is detected, and that the level populations are smaller than expected for LTE.  This shows that the electron density in the outflowing clouds in FBQS1044 is lower than that in the lower velocity clouds in spectrum of Q0059-2735 studied by WCP, which were found to be in LTE. An accurate electron density estimate based on the observed level populations is difficult to obtain, since only a small fraction of the radiative transition probabilities for the strongly forbidden transitions within and between the $a^6D$, $a^4F$ and $a^4D$ terms are known (Nussbaumer \\& Storey 1980, Quinet, Le Dourneuf \\& Zeippen 1996).  A very rough estimate of the electron density can be obtained by considering only the relative populations of the ground level and the lowest excited level ($E=384\\ \\rm{cm}^{-1}$). This level can only decay radiatively to the ground level and not make other radiative decays with unknown transition probabilities. The electron density can be estimated assuming that there is equilibrium between collisional excitation into this level from the ground level and radiative de-excitation to the ground level. For this transition the radiative decay probability $A_{21}$ is 0.00213 (Quinet, Le Dourneuf \\& Zeippen 1996) , and collision strength $\\Omega_{12}$ is 5.5 (Zhang \\& Pradhan 1995). Taking into account the formal errors on the level populations in Table 2, and constraining the temperature of the absorbing gas to lie in the range between 300 and 10.000 K, we obtain $\\log n_e = 3.55(0.25)$ for cluster 2 and $\\log n_e = 3.0(0.4)$ for cluster 1. Using the observed level populations and the collision strengths from Zhang \\& Pradhan 1995, it is easily verified that for $\\log n_e = 3.55$ and a temperature of 5000 K the total collisional excitation rate out of the lowest excited level is 0.24 times the radiative decay rate to the ground level, so that its neglect is justified for this rough estimate.  We can also derive an estimate of the electron density by using the most recent attempt at modelling the level populations of \\feii\\ as a function of density using a full multilevel calculation by Verner et al. 1999. In these models the unknown transition probabilities have been set to such a small value that they are effectively neglected. Based on these results we can make an estimate of the electron density by comparing the departure coefficient of the ground level (the only level among the ones relevant here for which results are presented in Verner et al. 1999) with the predicted value as a function of electron density. We define an ``observed'' departure coefficient for the ground level as \\begin{equation} b_{0,obs}=\\left({n_{0,obs}\\over{\\sum n_{i,obs}}}\\right) \\left({{\\sum g_i e^{(-E_i/kT)}}\\over{g_0}}\\right) \\end{equation} where the sum is over the levels $i$ given in table \\ref{levpop}, and the temperature was taken to be 5000 K for consistency with the Verner et al. model. For this temperature, the contribution of the higher energy levels to the partition function will be very small because of the Boltzmann factor, and their neglect is well justified. The observed departure coefficients are $\\log b_{0,obs} = 0.33(0.08)$ for cluster 2 and $\\log b_{0,obs} = 0.37(0.17)$ for cluster 1. Using Figure 3 in Verner et al. 1999 we can convert these limits to $\\log n_e = 3.8(0.3)$ for cluster 2 and $\\log n_e = 3.5(0.5)$ for cluster 1. These values are in reasonable agreement with the results obtained above by treating the ground and lowest excited level of \\feii\\ as a two-level atom.   \\subsection{Ionization modelling}  We have used the photoionization code CLOUDY (Ferland 1996) to model the ionization state of the outflow to derive additional constraints on the absorbing gas.  Our observational data yield the following constraints on the outflow in FBQS1044+3656: a \\mgi\\ column density $\\sim 2 \\times 10^{13} {\\rm cm^{-2}}$, a lower limit on the \\mgii\\ / \\mgi\\ ratio of $\\sim 30$, an \\feii\\ column density $\\sim 3 \\times 10^{15} {\\rm cm^{-2}}$, an upper limit on the \\fei\\ column density of $\\sim 10^{13} {\\rm cm^{-2}}$ and an estimate of the electron density of $\\sim 4 \\times 10^3 {\\rm cm^{-3}}$.  It turns out that the lower limit on the \\mgii\\ / \\mgi\\ ratio is not very useful, so we will concentrate on the \\mgi, \\feii, and \\fei\\ column densities.  \\subsubsection{Dust-free models }  We will first consider simple models of the absorbing clouds, investigating their structure as a function of ionization parameter, density and shape of the ionizing spectrum. For the moment we ignore the possible effects of dust, which can introduce significant complications.  The clouds are modelled as thin constant density slabs.  In order to satisfy the observational constraints, the slabs are truncated when a value of the \\feii\\ column density of $\\log N_{Fe~II}=15.5$ is reached. Furthermore, to satisfy the electron density constraint we have adjusted the density of the slab in each model so that the \\feii\\ weighted mean electron density is $4 \\times 10^3 {\\rm cm^{-3}}$. In this way the ionization parameter , the abundances and the shape of the ionizing spectrum remain as parameters that can be adjusted to satisfy the constraints on the \\mgi\\ and \\fei\\ column densities. The parameters and properties of the ionization models presented here are given in Table \\ref{ionmods}.  Figure \\ref{uplot} illustrates the structure of the slab as a function of ionization parameter for gas with solar composition irradiated by a standard Matthews \\& Ferland (1987) AGN spectrum.  When modelling the very low ionization species considered here with the CLOUDY code, one encounters the difficulty that \\mgi, \\fei\\ and \\feii\\ can occur in gas with a temperature below the limit at which CLOUDY results are reliable.  We therefore had to stop the calculation of the cloud structure at the depth where the temperature falls below 1000~K, if this occurs before the required \\feii\\ column density is reached.  At temperatures below 1000~K \\feii\\ is still the main form of Fe.  This means that when the calculation is stopped because of the temperature constraint (which occurs for low ionization parameters), the column densities found are only lower limits.  The run of temperature and electron density through the slab is similar to that found in the detailed \\feii\\ BEL models of e.g. Wills et al. (1985) and Collin-Souffrin et al. (1986), with an almost fully ionized Str\\\"omgren layer at the illuminated side of the side, followed by a partially ionized zone in which most of the column density of \\feii\\ resides. To understand the behaviour of the \\mgi\\ column density as a function of the model parameters it is important to realize that the ionization equilibrium of Mg behaves slightly counter-intuitively, in the sense that the fraction of neutral Mg drops sharply {\\em behind} the hydrogen ionization front. The reason for this behaviour is that the photoionization rate of \\mgi\\ is hardly affected by the front because the ionization potential of \\mgi\\ is considerably lower than that of hydrogen, whereas the recombination rate is much reduced because of the sharp drop in electron density. Therefore, the \\mgi\\ column density of the slab resides almost entirely in the ionized Str\\\"omgren layer.  The three models with $\\log U \\ge -3$ all have \\mgi\\ column densities less than $10^{13}{\\rm cm^{-2}}$ (see Table \\ref{ionmods}), between 3 and 10 times less than observed in FBQS1044+3656. The three models with $\\log U$ equal to --4, --5 and --6 run into the 1000~K temperature limit before a high enough \\feii\\ column density is reached, so that the column densities of \\fei\\ and \\mgi\\ are lower limits only. The model with $\\log U=-6$ is excluded, since it already has an \\fei\\ column density higher than our upper limit when the \\feii\\ column density is only $2.5~10^{14}{\\rm cm^{-2}}$. Thus of the models presented in Figure \\ref{uplot}, only those with $\\log U = -4$ and $\\log U = -5$ may explain the observed column densities in FBQS1044+3656. Although the column densities may be consistent with such a very low ionization parameter model, they are not likely to provide a good explanation for the observed \\feii\\ lines in detail, since most of the \\feii\\ column density would have to reside in the region with a temperature below 1000~K, which is difficult to reconcile with the observed population of the excited state at 1873 cm$^{-1}$ because the Boltzmann factor for this level should be $ < 0.067$. Thus we conclude that simple solar composition ionization models are not able to reproduce the observed column densities.  Model 2 ($\\log U = -2$) only fails to reach the observed \\mgi\\ column density by a factor of 3. Such a relatively small discrepancy may be fixed by changing the abundances from the solar values. The simplest solution is to increase the Mg abundance by a factor of 3, which brings the \\mgi\\ column density very close to the observed value. Increasing the Fe abundance reduces the thickness of the slab for a given \\feii\\ column density, but because most of the \\mgi\\ resides in the Str\\\"omgren layer, this hardly affects the \\mgi\\ column density. Thus with an overall increase in metallicity to 3 times solar (Model 7) the observed column densities can be reproduced.  Although the column densities can be reproduced by this high metallicity model, there are two problems that remain to be resolved. The first is that these models are rather contrived in the sense that extreme fine tuning of the cloud geometry is required to obtain the correct \\feii\\ column density. Most \\feii\\ resides in a thin layer at the back side of the cloud, with a thickness much smaller than the thickness of the Str\\\"omgren layer, a problem also pointed out by WCP (see below).  Since the thickness of the slab is set by the \\feii\\ column density, this problem can be remedied by reducing the Fe abundance by a very significant factor. This leads to slabs with a much thicker partially ionized zone. The larger extent of the partially ionized zone does not contribute appreciably to the \\mgi\\ column density however, so that an enhanced Mg abundance is still required. The second problem is that the \\mgi\\ and \\feii\\ reside in different zones of the slab. In FBQS1044+3656, the gas along the line of sight is broken up into many clouds at different velocities, and we would not expect to see such a good correspondence between the absorption components of \\feii\\ and \\mgi\\ if these lines were not formed in the same gas.  In figure \\ref{spplot} we investigate the sensitivity of these results to our assumptions about the shape of the ionizing spectrum. The AGN spectrum used above has a very strong EUV bump. Some more recent observation (e.g. Laor et al. 1997) suggest that in at least some active galaxies the EUV flux is considerably smaller than predicted by the Matthews \\& Ferland spectrum. To investigate this effect, we have repeated some models with a simple power law spectrum with slope -1.4 between 10 microns and 50 keV, which effectively removes the EUV bump. Figure \\ref{spplot} compares the models with $\\log U$ equal to -2, -4 and -6 for the two choices of ionizing spectrum. Although the temperature and electron density in the partially ionized zone are higher in the case of the power law spectrum, the column densities are very similar, so that our conclusions from Figure \\ref{uplot} are not very dependent on the exact shape of the ionizing spectrum.  \\subsubsection{Models with dust}  From the ionization models above we have found that the low ionization species we have been studying are mainly found in gas with a temperature of a few thousand Kelvin. This means it is very well possible that some form of dust will be present in the outflowing clouds. As we will demonstrate, this can change the ionization models significantly. The presence of dust will have three main effects:  \\begin{enumerate} \\item Fe is likely to be strongly depleted. This means that the hydrogen column density required to obtain the observed \\feii\\ column density is much larger.  \\item It is possible that Fe and Mg are depleted by {\\em different} depletion factors. Thus the column density ratios between Fe and Mg ions will be changed.  \\item The dust absorbs ionizing radiation, which changes the ionization equilibrium in the slab. \\end{enumerate}  The following models are based on the Orion dust model in CLOUDY, which is in turn based on the work on abundances and dust properties in the Orion Nebula H~II regions (Ferland 1996, Baldwin et al. 1991, Osterbrock et al. 1992). The properties of this dust model that are important for this work are the depletions of Fe and Mg ($\\sim$ 0.1 for both) and the dust opacity, which reflects the absence of small grains. The physical conditions in the clouds we consider are more similar to those in H~II regions than those in the local ISM, for which a different depletion pattern and grain size distribution has been derived(e.g Cowie \\& Songaila 1986).  The results of the ionization models including dust are illustrated in Figure \\ref{udust}. The most obvious effect of the presence of dust is the depletion of Fe, which leads to thicker slab models with a much thicker partially ionized zone, as discussed above. However, the most important physical difference between these models and the models without dust is that the \\mgi\\ fraction no longer drops as steeply behind the ionization front, so that the largest contribution to the \\mgi\\ column density now comes from the partially ionized zone rather than the Str\\\"omgren layer. The reason for this behaviour is that the \\mgi\\ ionizing flux (including trapped Ly $\\alpha$) is absorbed by dust, so that the reduction in recombination rate due to the drop in electron density is partially compensated. Thus in the models with dust the \\feii\\ and \\mgi\\ column density reside in the same region of the slab, whereas they are separated in the models without dust. Since the partially ionized zone now contributes to the \\mgi\\ column, the observed column can be explained with a much smaller Mg abundance than in the models without dust.  The largest \\mgi\\ column density for the given \\feii\\ column density is produced for $\\log U = -3$ (Model 13).  Even for this model it is still necessary to assume that Mg is three times less depleted than in the standard Orion dust model (i.e. by a factor of 0.33 instead of 0.1) to reproduce the observed \\mgi\\ column. For lower ionization parameters the models run into the 1000 K temperature limit again, which means that most \\feii\\ would be too cold to explain the \\feii\\ excitation, as discussed above. Higher ionization parameters can also reproduce the \\mgi\\ column density if we allow Mg to be even less depleted.  However, the models with larger ionization parameters are effectively excluded by another constraint, which is that the spectrum of FBQS1044+3656 does not appear to suffer from very strong extinction. The extinction at 2500 \\AA\\ predicted by the models with dust is given in Table \\ref{ionmods}. A comparison of the spectrum of FBQS1044+3656 with the mean QSO template of Weymann et al. 1991 shows that the extinction at 2500 \\AA\\ is not likely to be larger than 1.0 magnitude, unless the spectrum is intrinsically very steeply rising towards shorter wavelengths. This excludes the models with higher ionization parameters than $\\log U = -3$.  All models with $\\log U > -6$ have \\fei\\ column densities that lie well below our upper limit.  We conclude that models with dust can explain the observed properties of the outflow reasonably well, and do not suffer from the problems with an improbable cloud geometry and separation of the \\mgi\\ and \\feii\\ producing zones encountered in the high metallicity models without dust.   \\subsection{The size of the outflow}  For the cosmological parameters $H_0 = 50\\ {\\rm km\\sec^{-1} Mpc^{-1}}$, $\\Omega=1$ and $\\Lambda=0$, the absolute B magnitude of FBQS1044+3656 is $M_B=-26.2$. Assuming further that the shape of the continuum is close to the typical AGN spectrum described by Matthews \\& Ferland (1987), we estimate the luminosity of FBQS1044+3656 to be close to the canonical QSO value of $10^{46} {\\rm erg\\ s^{-1}}$.  In table \\ref{ionmods}, the distance between the source of ionizing radiation and the absorbing cloud corresponding to each model is tabulated assuming this luminosity. Both the high metallicity models without dust and the dusty models require that the outflow is situated at a distance of several hundred parsecs from the source of ionizing radiation, i.e typical of the QSO narrow line region. This distance is about two orders of magnitude larger than the distance derived by WCP for the \\feii\\ absorbing clouds in the BALQSO Q0059-2735.  This large size of the outflow raises several questions. The velocity of the outflow is considerably larger than that of material in the narrow line region, and is more typical of the BEL clouds that are situated at only 0.1 parsec from the nucleus. The combination of a large distance and the very small cloud size implied by partial covering re-emphasizes the problems associated with the stability of small, high speed clouds with a very low filling factor , which have been discussed extensively in the context of the formation of broad absorption lines (Weymann, Turnshek \\& Christiansen 1985, Begelman, de Kool \\& Sikora 1991, Weymann 1997). Although the small implied sizes of the clouds are hard to understand theoretically, we do point out that the small transverse cloud size implied by partial covering is consistent with the physical thickness of $\\sim 10^{16}$\\ cm of the slabs in our ionization models as derived from the hydrogen densities and column densities.  One of the main objections to large distances for BAL-like outflows comes from constraints on the mass and energy involved using the typical parameters of high-ionization BALs.  The low ionization absorption lines studied in this paper, however, have rather low velocity and equivalent width compared to typical high-ionization BALs.  We can estimate the mass loss rate and kinetic energy of the outflow for the specific case of FBQS1044+3656 to see if they are unreasonably large.  The mass involved in the low ionization outflow can be estimated from \\begin{equation} M_{outflow} \\gtorder 10^7 N_{21} R_{21}^2 f_{cov} \\Msun \\quad . \\end{equation} where $N_{21}$ is the hydrogen column density normalized to $10^{21}$\\ cm$^{-2}$, $R_{21}$ the size of the outflow in units of $10^{21}$\\ cm and $f_{cov}$ is the global covering factor of the outflow.  Since \\feii\\ absorption is not very common in QSOs, $f_{cov}$ could be as small as $10^{-2}$ for \\feii\\ absorption-producing outflows. Alternatively the outflows may be short lived phenomena, in which case the constraints on the mass loss rate are also less severe.  Taking the parameters of our preferred Models 7 or 15 and estimating the flow time to be $R/v$, the implied mass outflow rate is $70 f_{cov}$ \\Msun yr$^{-1}$ and the kinetic power of the outflow is $3 \\times 10^{44} f_{cov}$ erg s$^{-1}$.  These are values that do not argue strongly against the large size of the outflow.  \\subsection{Is FBQS1044+3656 a BALQSO?}  As mentioned in the Introduction, the characteristics of the absorption lines in FBQS1044+3656 are intermediate between those of BALs and AALs.  Weymann et al. (1991) classified a QSO as a BALQSO if the \\civ\\ line profile showed a continuous absorption of at least 10\\% in depth spanning more than 2000 \\kms\\ in velocity, discounting absorption closer than 3000 \\kms\\ bluewards of the emission peak.  The \\mgii\\ absorption in FBQS1044+3656 meets these criteria, even when a correction is made for the fact that the doublet separation in \\mgii\\ is larger than in \\civ\\ .  The total width of the \\mgii\\ absorption is about the same as in Q0059-2735 which exhibits extremely strong BALs in the higher ionization lines, and it is an established property of low ionization BALQSOs that the absorption troughs of higher ionization species like \\civ\\ or \\ovi\\ are generally much wider than the \\mgii\\ absorption (Weymann et al. 1991).  Thus, considering the properties of the \\mgii\\ absorption alone FBQS1044+3656 should be classified as a BALQSO. In the standard picture of BALQSOs, the absorption lines are formed in an outflow with a typical size of a few parsec.  The characteristics of the very low ionization lines of \\mgi\\ and \\feii\\ are typical of AALs (Weymann et al. 1979, Foltz et al.  1986), for which no firm distance scale has been established so that the distance of a few hundred parsec found in this paper is perhaps not a surprise. The remarkable fact lies in the excellent correspondence in velocity between the strongest \\mgii\\ features and the \\mgi\\ and \\feii\\ absorption lines, which establishes without a doubt that the very low ionization lines and the \\mgii\\ lines are formed in the same outflow.  The association of the \\feii\\ and \\mgi\\ absorption with the \\mgii\\ BAL is much clearer in FBQS1044+3656 than in Q0059-2735, which only shows a number of relatively narrow \\feii\\ components with outflow speeds of a few hundred kilometer per second, and a very smooth \\mgii\\ BAL profile.  The combined properties of the \\mgii\\ , \\mgi\\ and \\feii\\ lines seem to indicate that FBQS1044+3656 contains an outflow with BAL-like characteristics situated at a distance of a few hundred parsec from the nucleus, a scale very much larger than would have been attributed to the outflow on the basis of the \\mgii\\ lines only. At this time, FBQS1044+3656 is a unique object and we do not know how general these results are.  In view of the importance of establishing the relation between the large scale outflow found here and the classical high ionization BAL phenomenon, the essential next step in the study of FBQS1044+3656 is to obtain a UV spectrum of the source to compare the absorption profiles of the low and high ionization species."
    },
    "0010/astro-ph0010140_arXiv.txt": {
        "abstract": "A numerical code aiming at the automatic detection of spherical expanding HI shells in radio data-cubes is presented. The following five shell parameters are allowed to vary within a specified range: angular radius, expansion velocity and three center coordinates (galactic coordinates $l$ and $b$, and systemic radial velocity $V$). We discuss several factors which can reduce the sensitivity of the shell detection: the presence of noise (both instrumental and ``structural''), the fragmentation and asphericity of the shell and the inhomogeneity of the background and/or foreground emission. The code is tested on four objects: two HII regions and two early B stars. We use HI data from the Canadian Galactic Plane Survey. In all four cases the evidence for an expanding HI shell is found. ",
        "introduction": "Shells are omnipresent in the interstellar medium (ISM) of spiral and dwarf irregular galaxies. In some systems most of the ISM can be described as an ensemble of interacting shells and supershells (Staveley-Smith et al. 1997; Mashchenko, Thilker, \\& Braun 1999, hereafter MTB).  Theoretically, many astrophysical phenomena  may lead to the formation of expanding HI shells: stellar winds (Weaver et al. 1977), supernova explosions (Cox 1972), the combined action of multiple supernovae and stellar winds in OB associations (Bruhweiler et al. 1980), radiation pressure from the field stars (Elmegreen \\& Chiang 1982) and the infall of high-velocity clouds onto the galactic plane (Tenorio-Tagle 1981).  In more general terms, any physical mechanism which is able to drive an expanding supersonic shock wave into the surrounding ISM long enough for radiative cooling in the shell of swept-up gas to become significant, will lead to the formation of a relatively thin ($\\la$10\\% of the shell radius) and cold ($\\sim$ 100--1000~K) expanding HI shell. (See Bisnovatyi-Kogan \\& Silich 1995 for a review of shock wave propagation in the ISM.)  Unfortunately, there are many factors which can complicate the above picture and make the HI shells more difficult to detect. The blowing out of an interstellar bubble caused by the presence of a sharp density gradient in the surrounding medium makes the shell susceptible to Rayleigh-Taylor instabilities, which results in its fragmentation. The propagation of the shock through the clumpy turbulent ISM can be another destructive factor. The shells formed around O and early B stars are expected to be at least partially ionized by the ionizing photons from the star (Weaver et al. 1977). The stratification of the ISM structure and the differential rotation of the galactic disk result in the highly aspherical appearance of evolved supershells (Mashchenko \\& Silich 1994; Silich et al. 1996). The detection of expanding HI shells in edge-on disk systems and in our own Galaxy is further hampered by the presence of confusing emission along the line of sight.  As a result, few HI shells have been detected around stars with strong stellar wind. There are $\\sim$30 HI shell detections for WR stars (Arnal \\& Mirabel 1991, Dubner et al. 1992, Pineault et al. 1996, Arnal et al. 1999, Cappa et al. 1999, Gervais \\& St-Louis 1999, and references therein), and just a few shell candidates for O stars (van der Bij \\& Arnal 1986; Benaglia \\& Cappa 1999 and references therein) and B stars (Dubner et al. 1992).  Pearson correlation coefficient based algorithms can be successfully used for the automatic recognition of complex kinematic structures in the radio data-cubes (Thilker, Braun, \\& Walterbos 1998; MTB). A cross-correlation approach is tolerant to the partial fragmentation and ionization of the shell, to the mild distortion of its shape and to the presence of slowly varying background emission (MTB).  In this article we describe an object recognition code for the automatic detection of the expanding HI shells. The code uses the Pearson correlation coefficient as a measure of similarity between the data and a simple model of the shell. ",
        "conclusions": ""
    },
    "0010/astro-ph0010189_arXiv.txt": {
        "abstract": "s{ Many different and complementary strategies for translating the basic principle of multiple topological imaging into observational analysis are now available, both for three-dimensional and two-dimensional catalogues.} ",
        "introduction": "Observational cosmic topology\\cite{LaLu95,LR99} shares the fundamental observational problems of curvature estimates (evolution of objects, peculiar velocities, etc., see Section 5.3 of ref.\\cite{LR99}), but since the former concerns global geometry and the latter only local geometry, the former requires additional work in classifying the methods from a purely geometrical point of view. Most work is based on the principle of the existence of multiple topological images of single physical objects. ",
        "conclusions": ""
    },
    "0010/astro-ph0010376_arXiv.txt": {
        "abstract": "We present the Arecibo Dual-Beam Survey (ADBS), a ``blind\" 21cm survey for galaxies. The ADBS is the largest to date with the Arecibo radio telescope. We were able to achieve an rms sensitivity of 3-4 mJy in only 7 seconds. The ADBS covers substantially more volume than the Spitzak \\& Schneider (1998) and the Zwaan et al. (1997) surveys. It is surpassed by the Parkes HIPASS survey (Kilborn et al. 1999) but has much better angular resolution and sensitivity. We will discuss the HI sample and our efforts to quantify the completeness and reliability.  This is the first survey to use synthetic sources to establish the relation between the flux, line width, and completeness of the survey. An empirical understanding of these quantities is vital to the determination of the mass function. Even if we do not understand all of the reasons why we do not detect galaxies, the empirical understanding allows us to account for them properly.  In addition to the more general discussion of the ADBS survey, we will discuss galaxies which have large ratios of HI mass to optical area and probably have extended HI disks. We detect a much higher percentage of these high HI mass to optical area galaxies than are seen in optical surveys indicating that optical surveys miss these galaxies. Among these galaxies are also 7 galaxies with HI masses $<$  10$^8$ M$_{\\sun}$. Based on the mass function of Schneider \\& Rosenberg (this volume), the low mass galaxies are  an important part of the general galaxy population. ",
        "introduction": "We present the Arecibo Dual-Beam Survey (ADBS) and describe the original driftscan detection survey, the VLA and Arecibo follow-up observations, and the use of synthetic sources as a way of evaluating sample completeness. We then discuss the indications from the ADBS sample that even low surface brightness optical surveys have overlooked a population of compact and low surface brightness (LSB) galaxies that are found in HI surveys. The ADBS, and other ``blind\" HI surveys provide an important view of galaxy populations that is completely unbiased by star formation. Comparing our HI-selected sample to the dwarf and LSB sample of Schneider et al (1990, 1992), we find that the dwarf and LSB sample misses a significant fraction of the LSB population. The detection and study of this galaxy population is important for our understanding of galaxy formation and evolution. ",
        "conclusions": "We have completed a survey of $\\sim$ 430 deg$^2$ at 21 cm and have detected 265 galaxies, 7 of which are below 10$^8$ M\\sun. In order to understand the significance of these low mass sources and to generally understand the statistics of our detection survey, we have used synthetic sources. These sources allow us to map out our completeness as a function of signal-to-noise.  We have also compared the detection of HI-dominant galaxies in the ADBS to their detection in the Schenider et al. (1992, 1990) dwarf and LSB sample. We find that when we select galaxies by their HI, a method that is completely unbiased by the galaxy's surface brightness, we find a much larger fraction of the galaxy population to be HI-dominant than is found by the LSB and dwarf galaxy sample. In addition, the ADBS galaxies are more extreme in their HI-dominance, surface brightness, and color characteristics than those  detected in the LSB and dwarf galaxy sample. The LSB and dwarf galaxy survey, and probably other optical LSB surveys, is not finding all of the galaxies and it is not detecting them in a manner  which is unbiased by their surface brightness properties. In order to  understand the processes of galaxy formation and evolution we must understand the range of star forming environments for galaxies including the HI-dominant galaxies found in ``blind\" HI surveys."
    },
    "0010/hep-ph0010255_arXiv.txt": {
        "abstract": "The gravitational properties of cosmic strings in the Randall-Sundrum brane world scenario are investigated. Using a gauge in which the brane remains straight, the leading order corrections to the metric on the brane are determined. In contrast to their non-brane equivalents, these strings have an attractive $1/r^2$ potential and a radially dependent deficit angle. These two effects alter many cosmological properties of the string, such as the formation of double images. As a result of the attractive force the string will collect matter as it moves through space. ",
        "introduction": "Cosmic strings are a type of topological defect formed at phase transitions in some grand unified theories (GUTs)~\\cite{vilshel}. In the past their gravitational effects have been considered as a possible explanation for the anisotropies in the cosmic microwave background (CMB). Recent observations~\\cite{CMBdata} and simulations~\\cite{strCMBsim} have raised doubts about this.  It has  been suggested that the physical universe may be a 3-brane embedded in a higher dimensional spacetime~\\cite{BWold}. Ordinary matter is confined to this brane, while gravity is free to propagate in the higher dimensional bulk. To be phenomenologically credible, gravity on the brane must, at least at large distances, resemble 4-dimensional Einstein gravity. In a scenario proposed by Randall and Sundrum the bulk space has a non-zero cosmological constant. This gives a mass to the bulk gravitons, suppressing their effects. The standard 3+1-dimensional gravity is produced by massless gravitons trapped on the brane, with small corrections from the bulk. By tuning the brane tension the effects of the bulk cosmological constant can be cancelled on the brane.  In contrast to a point particle, a cosmic string (in the standard 4-dimensional universe) does not exert a force on nearby matter or radiation. Its gravitational effects result from the the fact that the space around it is approximately conical. If two initially parallel particle paths pass either side of the string, the conical geometry will cause their paths to converge. A related effect is the formation of double images. The string's gravitational lensing means that if an observer is on the opposite side of a cosmic string to a quasar, he will see two images of it~\\cite{vilen,gott}.  The conical geometry also produces a Doppler shift in light emitted from objects moving relative to the  string (assuming the string is between the object and the observer). When a string passes between an observer and the cosmic microwave background this Doppler effect will produce anisotropies~\\cite{strCMB,gott}. It is the CMB anisotropies predicted by this effect which have conflicted with observational data, and so damaged the credibility of cosmic string models.  In this paper the gravitational field around a brane world cosmic string is found, using a weak field approximation. The resulting field is significantly different to that of a string in the standard four dimensional universe. This means that any cosmological properties of the string which arise from its gravity need to be revised in the brane world scenario. This is in contrast to point-like particles, where the brane world gravity is qualitatively the same as in the standard case. The geodesics away from the string are also determined. These are used to give a modified expression for the separation of double images created by the string. ",
        "conclusions": "In this paper I have examined the linearized gravitational field around a cosmic string in the Randall-Sundrum brane world scenario. The brane corrections give the string an attractive $1/r^2$ potential, in sharp contrast to standard cosmic strings. At astronomical distances the force is very weak, and so significant gravitational effects will only occur near the string core. Like the standard strings the space at large radii is conical. However as the string core is approached the deficit angle reduces.  Like their non-brane counterparts the brane strings produce double images of objects behind them. The angle between these images will be slightly larger than in the standard case. The closer the light rays pass to the string, the more noticeable the difference. The brane corrections will also alter the CMB anisotropies produced by the string, although I will not consider this here.  The attractive force of the brane string also means that particles can make circular orbits around the string. If the particles also have momentum in the direction parallel to the string, the corresponding paths will be helical. The stability of these orbits is unclear, and depends on gravitational effects beyond the linear approximation. Whether the orbits are stable or not, the string is expected to collect matter as it moves through the universe. This will be in the form of particles which either orbit the string or are pulled into its core."
    },
    "0010/astro-ph0010410_arXiv.txt": {
        "abstract": " ",
        "introduction": "For many years it has been known that some Young Stellar Objects (YSOs) are sources of radio emission in the cm continuum. The more deeply embedded YSOs, now commonly referred to as Class I and Class 0 protostars, exhibit the `thermal' spectrum of free-free radio emission between 2 and 6~cm. At shorter wavelengths continuum emission from cold circumstellar dust becomes the dominant source of radiation (Rodriguez 1994). The energy source for the ionisation and free-free emission in low mass YSOs is still unknown. As the sensitivity of radio telescopes has improved, free-free emission has been detected from an increasing number of protostars, particularly those with CO outflows and associated Herbig-Haro flows. Rodriguez and Reipurth (1996) suggested that that all such systems may be radio sources.  We have conducted a sensitive survey of a small sample of Class I YSOs in the Taurus molecular cloud which exhibit outflow activity. The aim was to discover whether significant radio emission is indeed a universal feature of such systems and, by comparison with near infrared data, to shed light on the process which produces it. By observing at high spatial resolution we also hoped to probe the structure of the circumstellar matter on the scale of the solar system. The radio emission might trace the structure because shock-ionisation of the outflow where it is collimated by the molecular envelope is a long-standing hypothesis to explain the free-free emission. It transpired that most Class I sources are too faint for detailed mapping to be achieved with present radio receiver technology but some useful spatial information was obtained.  The initial survey and follow-up observations at radio and infrared wavelengths are described in Section 2. The results are detailed in Section 3 and in Section 4 we discuss the possible explanations for the differing radio fluxes of otherwise similar protostars. Our conclusions are presented in Section 5. ",
        "conclusions": "Observation of a small sample of low mass Class I protostars indicates that radio emission at above the level of 0.1~mJy/beam  is far from ubiquitous. We suggest that the radio quietness in sources with active Herbig-Haro flows might be accounted for by a relatively low ionisation fraction, at the level of a few percent. The ionisation fraction could plausibly be related to the collimation of the outflow, particularly if shock-ionisation by interaction with the circumstellar envelope plays a role. In addition, radio variability due to unsteady outflows may also occur and this should be investigated by monitoring known radio emitters.  In 04016+2610 (L1489 IRS) a well defined bow shock is observed in 2.12~$\\mu$m H$_{2}$ radiation along the line of the secondary bipolar cavity revealed in a polarised intensity image. This provides further evidence for a quadrupolar flow. We suggest that the system is the product of a gravitational interaction between two young solar systems, which produced a binary star whose components drive outflows along different axes of rotation. The system may be an unresolved binary (separation $\\ltsimeq 0^{``}.1$ or 14 au), or the secondary component may be obscured by dense material in the core of the system."
    },
    "0010/astro-ph0010626_arXiv.txt": {
        "abstract": "Implementation details and test cases of a newly developed hydrodynamic code, \\AMRA, are presented. The numerical scheme exploits the adaptive mesh refinement technique coupled to modern high-resolution schemes which are suitable for relativistic and non-relativistic flows. Various physical processes are incorporated using the operator splitting approach, and include self-gravity, nuclear burning, physical viscosity, implicit and explicit schemes for conductive transport, simplified photoionization, and radiative losses from an optically thin plasma. Several aspects related to the accuracy and stability of the scheme are discussed in the context of hydrodynamic and astrophysical flows. ",
        "introduction": "Many problems in the numerical simulation of hydrodynamic flows require the use of a high grid resolution in order to describe the evolution of the system properly. In turn, the use of large numerical grids implies high computational costs in terms of both, memory and CPU time. However, in many cases the most important flow features occupy only a small fraction of the computational domain. Those structures are usually flow discontinuities like shock waves or contact surfaces.  The addition of physical processes may lead to the formation of qualitatively new features which, similarly to flow discontinuities, can occupy only a small fraction of the total volume. A proper description of the additional physics may also require a resolution of time scales which might be much smaller than the hydrodynamic time scale.  For some problems it is resolution in mass rather than in space or time which determines the quality of a numerical solution. Here methods exploiting the Lagrangian rather than Eulerian description of the flow might be more suitable.  However, the Lagrangian approach encounters severe difficulties in multidimensional problems due to large grid distortions typical for shearing flows. Sophisticated grid rezoning algorithms \\cite{cale} did not gain much popularity except perhaps for one-dimensional problems \\cite{sage,titan}. Alternatively, the notion of a grid might be avoided completely leading to a meshless code. The Smoothed Particle Hydrodynamics method \\cite{l77,gm77,m92} is a practical example of the realization of this idea. In passing we note, that in case of discontinuous flows with shocks the use of artificial viscosity prevents the SPH method from reproducing the quality of the solutions obtained with modern shock-capturing advection schemes implemented in the majority of Eulerian codes \\cite{sm93,mm97}.  One may also consider a Cartesian method in which the grid zones are consecutively refined to increase the resolution where it is desired from the point of view of numerical accuracy. In this approach the grid refinement is done on the basis of single cells resulting in a grid which usually has to be described using a complicated data structure \\cite{mcobra,fr93,dp93,kh98}. When only one isolated region of the volume has to be refined, an approach with fully nested grids of increased resolution, similar to multigrid methods with local refinements for elliptic \\cite{b77,b84,bb87} or parabolic problems \\cite{cs73,vlugr3}, might be used \\cite{r92,zy97}.  In \\AMRA\\ we adopt a block-structured approach to grid refinement in which the refined grid cells are clustered together to form larger rectangular regions, or mesh patches, overlying parent level grids. The refinement process is recursively applied to newly created fine mesh patches in order to increase the resolution even further. In this way, the final data structure can be seen as a hierarchy of mesh patches (Fig.~\\ref{f:grids}) \\begin{figure} \\begin{center} \\includegraphics[bb =  30 178 563 666, height=7 cm]{grids} \\caption[]{ Hierarchy of grids in Adaptive Mesh Refinement method. The base level covers the whole computational domain with mesh patches G(1,1) and G(1,2). Patches G(2,1) and G(2,2) form the second level and a single mesh patch, G(3,1), is located on the finest level. Notice that the hierarchy is fully nested with finer mesh patches completely covered by patches located on the next coarser level, and that each patch may have more than one parent or/and offspring, and siblings may overlap each other. } \\label{f:grids} \\end{center} \\end{figure} located at different levels and integrated with individual time steps. Each single mesh patch has a logical structure identical to the original numerical grid. This scheme is commonly called adaptive mesh refinement, AMR, and has been originally proposed by Berger and her collaborators \\cite{bo84,bc89,bell+94,amrclaw}. ",
        "conclusions": "\\label{s:discussion}  The implementation details of a newly developed Adaptive Mesh Refinement code, \\AMRA, have been presented together with one- and two-dimensional tests cases. Our experience shows that for a given resolution the AMR technique offers savings in terms of computer time as long as the average (over the whole computation) fraction of the computational domain to be resolved at the finest level (filling factor) does not exceed about one half of the total volume. This statement is valid for simple purely hydrodynamic problems. More substantial savings might be expected for problems involving computationally expensive source terms.  Although the fractional volume occupied by the finest level is the major factor determining the overall efficiency of \\AMRA, speedups obtained for the same problem show a strong dependence on processor architecture (RISC/cache, vector). Our experience on shared memory parallel systems indicates that the parallel overhead (communication, synchronization) is significant, and its reduction is likely to be even more important on systems with distributed memory.  Care has to be taken when simulating flows for which rotation and shear are important since patch corners effectively act as a source of spurious vorticity especially when the change of resolution between levels is substantial. Finally, we note that the AMR approach is not suitable for problems which require uniform high-resolution. These include simulations of convective or turbulent flows.  \\ack  We thank Micha{\\l} R\\'o\\.zyczka for his ongoing support and interest in this project. Konstantinos Kifonidis and Pawe{\\l} Cieciel\\c{a}g made an enormous contribution to this work by sharing with us their impressions following ``real world'' applications of \\AMRA\\ -- we thank them warmly for being patient with us! The work of TP was partly supported by the grants 2-1213-91-01, 2.P304.017.07, 2.P03D.004.13, and 2.P03D.014.19 from the Polish Committee for Scientific Research, and ESO grant A-01-063. The simulations have been performed on the CRAY SV1-1A at the Interdisciplinary Centre for Computational Modelling in Warsaw, the SGI Origin200 at the Nicolaus Copernicus Astronomical Center, Warsaw, and the CRAY J916/512 at the Rechenzentrum Garching."
    },
    "0010/astro-ph0010556_arXiv.txt": {
        "abstract": "We present H$\\alpha$ and [\\ion{S}{ii}] observations for the Local Group irregular galaxy \\object{IC 1613} using the PUMA scanning Fabry--Perot interferometer. Our goal is to analyze the kinematics of the ionized gas in the complex sample of superbubbles located in the whole extension of our field (10\\arcmin ), which includes most of the optical emission of this galaxy, and to study the inter-relationship between young stellar associations and nebulae based on a previous study that we have made on the stellar associations of the central region of this galaxy. The ionized gas in this galaxy is distributed in classical \\ion{H}{ii} regions and in a series of superbubbles (also called giant shells) covering a large fraction of the optical extent of the galaxy.  We present a catalog of kinematical properties of both the \\ion{H}{ii} regions of this galaxy and the superbubbles. We have also compared the kinematics of the ionized gas in \\ion{H}{ii} regions to search for possible dynamic differences between neutral and ionized gas. ",
        "introduction": "Irregular galaxies constitute a substantial fraction of all galaxies. They are also dominant among the actively star-forming galaxies. Because they lack spiral density waves, they are excellent laboratories to examine the star formation process in alternative scenarios. Also, they can provide important clues about the mechanisms of formation of ``bubbles'' (the ring-shaped nebulae with diameters, D $\\la$ 50 pc), ``superbubbles'' or ``giant shells''(ring-shaped structures with D $\\la$ 400 pc) and ``supershells'' (ring-shaped structures with D $\\approx$ 1 kpc). Massive stars can dramatically change the interstellar medium (ISM) of these galaxies by means of supernova explosions and stellar winds. On the other hand, the ISM content and energetics can have important effects on the subsequent mechanisms of star formation. Indeed, the compressions produced by shocks formed by supernova explosions or powerful stellar winds on the molecular clouds could trigger the formation of a second generation of stars at the edges of the shells formed by these events (self-induced star formation).  The Local Group irregular galaxies are exceptional targets for the study of the inter-relationship between gas and stars because they are easily resolved. This means that we could know the stellar content of nebular \\ion{H}{ii} complexes and superbubbles and try to identify for relationships between the interior stellar associations and the  dimensions and kinematics of the superbubbles. This kind of investigation has been performed for the most favorable case, the Large Magellanic Cloud (Rosado \\cite{rosado86}; Oey \\cite{oey96a}, \\cite{oey96b}, amongst other references). Indeed, it is in this galaxy that a clear example of self--induced star formation has been found: the nebular complex N11 (Parker et al. \\cite{parker96}; Rosado et al. \\cite{rosado96}).  In this work, we study the nebulae and massive stellar content of the nearby irregular galaxy: \\object{IC 1613}. This galaxy belongs to the Local Group and it is located at a distance of 725 kpc (Freedman \\cite{freedman88a}, \\cite{freedman88b}). Table \\ref{properties} reviews some of its main  properties. Because of its proximity, it has been the target of several studies. Its stellar content has been analyzed by Baade (\\cite{baade63}), Sandage (\\cite{sandage71}), Sandage \\& Katem (\\cite{sandage76}), Hodge (\\cite{hodge78}, \\cite{hodge80}), Freedman (\\cite{freedman88a}, \\cite{freedman88b}) and Georgiev et al. (\\cite{georgiev99}; hereafter Paper I). Several candidate Wolf-Rayet (WR) stars were detected but only one of them was spectroscopically confirmed over the whole extent of this galaxy (Azzopardi et al. \\cite{azzopardi88}; d'Odorico \\& Rosa \\cite{d'odorico82}; Kingsburgh \\& Barlow \\cite{kingsburgh95}). This WR is a WO--type star embedded in a large nebular complex located at the southern region of this galaxy (Rosado et al. \\cite{rosado00}). The \\ion{H}{ii} regions of \\object{IC 1613} have been studied by Sandage (\\cite{sandage71}), Hodge et al. (\\cite{hodge90}; hereafter HLG), Price et al. (\\cite{price90}) and Hunter et al.  (\\cite{hunter93}). The peculiarity of the kinematics of some of the superbubbles of ionized gas has been analyzed by Meaburn et al. (\\cite{meaburn88}). Their work focused only on the NE region of the galaxy where bright \\ion{H}{ii} complexes are easily detected. Other spectroscopic studies on the nebulae of \\object{IC 1613} mainly examine the only SNR (d'Odorico et al. \\cite{d'odorico80}; Peimbert et al. \\cite{peimbert88}; Dickel et al. \\cite{dickel85}; Goss \\& Lozinskaya \\cite{goss95}; Rosado et al. \\cite{rosado00}). The distribution of the \\ion{H}{i} gas in \\object{IC 1613} has been analyzed by Lake \\& Skillman (\\cite{lake89}). These authors have found that the \\ion{H}{i} content makes up to 20 percent of the total mass (see also Huchtmeier et al. \\cite{huchtmeier81}).  In this work, the results of our kinematic analysis, making use of a scanning Fabry-Perot interferometer (PUMA, Rosado et al. \\cite{rosado95}), are presented. We analyze the relationship between nebulae and the stellar associations studied in Paper I. The observations and data reductions are described in Section 2. In the remaining sections we study the kinematic behavior of the \\ion{H}{ii} regions and superbubbles. In Section 3 we show the nebulosities of this galaxy in several emission lines. We also indicate how our photometric calibration has been done. In Section 4 we study the kinematics of HII regions and superbubbles in \\object{IC 1613}. We obtain the peak velocities and velocity dispersions of the \\ion{H}{ii} regions, as well as their stellar content. We also obtain the \\ion{H}{ii} luminosity function and diameter distribution. In  Section 5 we study the superbubbles found, obtaining their dimensions, expansion velocities, H$\\alpha$ fluxes and studying their stellar content. We also obtain some important physical parameters of the superbubbles and the superbubble diameter distributions. We analyse the possibility that the superbubbles were formed by the combined action of stellar winds and supernovae explosions. In Section 6 we present the radial velocity field in our field of view, obtained from diffuse ionized gas, \\ion{H}{ii} regions and superbubbles. We compare our trends with the results of  \\ion{H}{i} observations. In Section 7 we present our conclusions. ",
        "conclusions": "We have measured -- by means of a Fabry-Perot interferometer -- the kinematical parameters of the \\ion{H}{ii} regions that we have detected within the 10$\\arcmin$ field of view of our observations (which includes most of the optical emission of \\object{IC 1613}). We found that the \\ion{H}{ii} regions of \\object{IC 1613} present similar integrated properties as well as H$\\alpha$ line profiles with simple structure that are fitted by single Gaussian curves. We obtained also the H$\\alpha$ fluxes and luminosities of these \\ion{H}{ii} regions. The luminosity function of these  \\ion{H}{ii} regions can be fitted to a power law function of slope $a=-1.54\\pm 0.1$ in agreement with the value given by HLG. We have also studied the diameter distribution and the cumulative diameter distribution of the \\ion{H}{ii} regions within our field of view. We obtained a characteristic diameter $D_0=65.41\\pm 3$ pc which is slightly larger than the previous determination by HLG, probably because our field is smaller. Our characteristic diameter value does not seem to follow the correlation found by Hodge (\\cite{hodge83}) between $D_0$ and the absolute magnitude $M_B$ for a sample of irregular galaxies.   We have also found that the ISM of the entire galaxy is organized in a complex network of superbubbles. We have catalogued the superbubbles detected in our field of view. We have also calculated the 2-D porosity parameter,  $Q_{2D}$, in the field of view of our observations and seen that the $Q_{2D}$ value is intermediate between the observed values for the SMC and M31.  Most of the superbubbles that we have detected show complex radial velocity profiles (some others are so faint that we cannot measure line-splitting in their velocity profiles). These kind of profiles along the same nebula are indicative of internal gas motions and expansions. We have determined the expansion velocities and other physical parameters of the superbubbles found in this work. In addition, we have used the photometric data published in a previous work (Paper I) and in Hodge (\\cite{hodge78}) in order to know the stellar content of the catalogued superbubbles. We find that almost every superbubble has an interior association containing massive stars, suggesting a physical link between them. Furthermore, we find that the dynamical time scales of the superbubbles are always shorter than the ages of the interior stellar associations.  The mean temporal delay between those ages (about 10 Myr) suggests that only evolved members (i.e., the more massive stars) of these associations are responsible for the formation of the superbubbles, possibly when they evolve to supergiants or when they explode as supernovae.  We have found that all the superbubbles with measured expansion velocities have expansion velocities larger than 25 km s$^{-1}$. We used energy considerations in order to explain the origin of the superbubbles in terms of stellar winds and supernova explosions. We find that the superbubbles with extreme expansion velocities (larger than 36 km s$^{-1}$) require interior stars with winds much more powerful than the average of B0--O9.5 supergiant stars if they are wind blown shells following the ``standard model''.  On the other hand, one or two supernova explosions could explain the mechanical energies involved. However, no other traces of SN explosions (extended X-rays, non-thermal radio emission) are known for these superbubbles except an enhancement of their [\\ion{S}{ii}] emission.  We have obtained the size distribution and the cumulative diameter function of superbubbles detected in our field of view and compared these size distributions with size distributions of superbubbles in other galaxies. Our results are statistically poor. However, the value of the characteristic diameter we obtain, $D_0=91.84\\pm 3$ pc, is in agreement with the value given by Price et al. (\\cite{price90}) for the nebulae in this galaxy.   On the other hand, we have determined the global H$\\alpha$ velocity field of \\object{IC 1613}, inside our 10$\\arcmin$  field of view. The velocity field is contaminated by the internal motions of the superbubbles of this galaxy. Nevertheless, a gradient of velocities is suggested across the SE-NW direction. In addition, we compare the mean velocities obtained from the  \\ion{H}{ii} regions with the  kinematics of the \\ion{H}{i} gas but we do not find a clear agreement between them.  \\begin{figure}[ht] \\special{psfile=ms9979f14.ps angle=0 hscale=70 vscale=70 voffset=-410 hoffset=-80} \\vspace{9.0cm} \\caption{\\object{IC 1613} H$\\alpha$ observed radial velocity distribution. Heavy lines correspond to \\ion{H}{i} isovelocities (Lake \\& Skillman \\cite{lake89}).} \\label{vel_HI} \\end{figure}"
    },
    "0010/astro-ph0010283_arXiv.txt": {
        "abstract": "We have developed a new diagnostic tool for the study of gamma--ray emission lines from radioactive isotopes (such as \\al\\ and \\fe) in conjunction with other multi--wavelength observables of Galactic clusters, associations, and alike objects. Our evolutionary synthesis models are based on the code of \\cite{CMH94}, which has been updated to include recent stellar evolution tracks, new stellar atmospheres for OB and WR stars, and nucleosynthetic yields from massive stars during hydrostatic burning phases and explosive SN~II and SN~Ib events.  The temporal evolution of \\al\\ and \\fe\\ production, the equivalent yield of \\al\\ per ionising O7~V star (\\yal), and other observables are predicted for a coeval population. The main results are: \\begin{itemize} \\item The emission of the \\al\\ 1.809 MeV line is characterised by four phases: stellar wind dominated phase ($\\la$ 3 Myr), SN~Ib dominated phase ($\\sim$ 3--7 Myr), SN~II dominated phase ($\\sim$ 7--37 Myr), and exponential decay phase ($\\ga$ 37 Myr). \\item The equivalent yield \\yal\\ is an extremely sensitive age indicator for the stellar population which can be used to discriminate between Wolf--Rayet star and SN~II \\al\\ nucleosynthesis in the association. \\item The ratio of the \\fe/\\al\\ emissivity is also an age indicator that constrains the contribution of explosive nucleosynthesis to the total \\al\\ production. \\end{itemize}  We also employed our model to estimate the steady state nucleosynthesis of a population of solar metallicity. In agreement with other works, we predict the following relative contributions to the \\al\\ production: $\\sim 9 \\%$ from stars before the WR phase, $\\sim 33 \\%$ from WR stars, $\\sim 14 \\%$ from SN~Ib, and $\\sim 44 \\%$ from SN~II. For \\fe\\ we estimate that $\\sim 39 \\%$ are produced by SN~Ib while $\\sim 61 \\%$ come from SN~II. Normalising on the total ionising flux of the Galaxy, we predict total production rates of 1.5 \\Msol\\ Myr$^{-1}$ and 0.8 \\Msol\\ Myr$^{-1}$ for \\al\\ and \\fe, respectively. This corresponds to 1.5 \\Msol\\ of \\al\\ and 1.7 \\Msol\\ of \\fe\\ in the present interstellar medium.  To allow for a fully quantitative analysis of existing and future multi--wavelength observations, we propose a Bayesian approach that allows the inclusion of IMF richness effects and observational uncertainties in the analysis. In particular, a Monte Carlo technique is adopted to estimate probability distributions for all observables of interest. We outline the procedure of exploiting these distributions by applying our model to a fictive massive star association. Applications to existing observations of the Cygnus and Vela regions will be discussed in companion papers. ",
        "introduction": "OB associations and young open clusters are the most active nucleosynthetic sites in our Galaxy.  The combined activity of stellar winds and core--collapse supernovae ejects significant amounts of freshly synthesised nuclei into the interstellar medium.  Among those, radioactive isotopes, such as \\al\\ ($\\tau=1.04 \\times 10^6$~yr) or \\fe\\ ($\\tau=2.07 \\times 10^6$~yr), may eventually be observed by gamma--ray instruments through their characteristic decay--line signatures. Their observation presents direct evidence of recent nucleosynthesis activity, which can be used as a powerful diagnostics tool for studies of present galactic activity.  Galactic 1.809 \\MeV\\ gamma--ray line emission attributed to the radioactive decay of \\al\\ has been observed by numerous gamma--ray telescopes, and the detailed mapping of the emission distribution by the COMPTEL telescope has clearly identified massive stars as the source of this radio--isotope (see \\nocite{prantzos96}Prantzos \\& Diehl 1996 for a review).  The most convincing evidence for a massive star origin comes from the close resemblance between the 1.809 \\MeV\\ and galactic free--free emission, which links \\al\\ nucleosynthesis to the O star population (Kn\\\"odlseder et al.~1999a\\nocite{knoedl99a}; Kn\\\"odlseder 1999\\nocite{knoedl99}).  Since in general a variety of distinct massive star populations of different ages, sizes, and metallicities contribute to the observed intensities along a line of sight, this indicates that the average properties of the populations are related.  However, the correlation between 1.809 \\MeV\\ and free--free emission also holds for regions far from the Galactic centre where only few massive star associations contribute to the observed emissions.  Examples are the Cygnus and the Vela regions where localised 1.809 \\MeV\\ emission enhancements coincide spatially with maxima of free--free radiation, showing the same relative intensities as the Galaxy as a whole (Kn\\\"odlseder et al.~1999a\\nocite{knoedl99a}).  To fully exploit in a quantitative manner such existing data and in preparation of the upcoming {\\em INTEGRAL} gamma--ray satellite mission the development of new interpretation tools is necessary.  We here present the first results from our modified time--dependent multi--wavelength evolutionary synthesis models.  A similar model was recently presented by \\cite{Pl00}.  To model the gamma--ray luminosities the nucleosynthetic production of the long--lived radio--isotopes \\al\\ and \\fe, has been included in our multi--wavelength code (Cervi\\~no \\& Mas-Hesse 1994\\nocite{CMH94}, Cervi\\~no et al.\\ 2000)\\nocite{CMHK00}.  These isotopes give rise to the 1.809 \\MeV\\ (for \\al) and 1.173 \\MeV\\ and 1.333 \\MeV\\ (for \\fe) gamma--ray lines respectively. The model properly accounts for the accumulation of radioactive elements and their respective decay times.  Results from state--of--the--art stellar atmospheres of massive stars are included to accurately predict the ionising fluxes from these stars, which are at the origin of thermal free--free emission. Together with the other synthesised observables this provides a versatile tool for gamma--ray to radio analysis of massive star forming regions.  For comparisons with individual Galactic star forming regions (e.g.\\ OB associations, clusters, H~{\\sc ii} regions) or ensembles of such objects it is not only imperative to model the temporal evolution of their properties. The effects of small number statistics of the massive star population must also be taken into account (Cervi\\~no et al.\\ 2000b\\nocite{CLC00}).  Various studies, including the present one, treat such effects by means of Monte Carlo simulations (e.g.\\ Cervi\\~no \\& Mas-Hesse 1994\\nocite{CMH94}, McKee \\& Williams 1997\\nocite{MKW97}, Oey \\& Clarke 1998\\nocite{Oetal98}).  Finally for a fully quantitative and objective confrontation with observables an additional step is performed here, to our knowledge for the first time in this context.  The observational constraints (e.g.\\ a known number of stars of given spectral types, derived age, distance etc.) and their uncertainties are included in a Bayesian approach providing probability distributions for all derived properties.  Section 2 describes the ingredients of our synthesis code.  The main predictions from our models are presented in Section 3.  Uncertainties are briefly discussed in Sect.\\ 4.  Our Bayesian approach to model realistic stellar populations is presented in Sect.\\ 5.  The main conclusions are summarised in Sect.\\ 6. ",
        "conclusions": "We have constructed a new diagnostic tool for the study of the radioactive isotopes of \\al\\ and \\fe\\ produced in massive star forming regions. The main aim of this work was to provide a quantitative model for the analysis of multi--wavelength observations of OB associations, open clusters and alike objects covering the range from gamma--rays (e.g.\\ the 1.809 MeV line of \\al, 1.173 and 1.333 Mev lines of \\fe) to radio, and allowing in a fully quantitative manner to account for statistical richness effects of massive star populations and other observational uncertainties.  To achieve this goal we have used the evolutionary synthesis models of \\cite{CMH94}, which have been updated to include recent Geneva stellar evolution tracks, new stellar atmospheres for OB and WR stars, and nucleosynthetic yields from massive stars during hydrostatic burning phases and explosive SNII and SNIb events (see Sect.\\ 2). In particular proper care was taken to combine the stellar models including mass loss with appropriate presupernova and SN models.  The temporal evolution of the ejected quantity of \\al\\ and \\fe\\ produced by a coeval population, other observables like the total ionising flux and the supernova rate, and derived properties is presented (Sect.\\ 3). This yields the following main results: \\begin{itemize} \\item The equivalent O7 V star \\al\\ yield (\\yal), defined as the stellar yield per ionising flux of an O7 V star (Kn\\\"odlseder 1999\\nocite{knoedl99}), shows a particularly strong time dependence where four main phases can be distinguished: stellar wind dominated phase ($\\la$ 3 Myr), SN Ib dominated phase ($\\sim$ 3--7 Myr), SN II dominated phase ($\\sim$ 7--37 Myr), and exponential decay phase ($\\ga$ 37 Myr). The exact age range of each phase is dependent on the evolutionary tracks and the lower mass limit of WR stars. \\item The use of \\yal\\ is a powerful tool to constrain the evolutionary status of star forming regions. This parameter is obtained from a combined observation of the $\\gamma$--ray line and the ionising flux (for example in form of thermal free--free radio emission) of an association. \\item \\fe\\ production starts with a delay of $\\sim$ 2 Myr with respect to \\al\\ production.  The ratio of the \\fe/\\al\\ emissivities is also an age indicator that constrains the contribution of explosive nucleosynthesis to the total \\al\\ production. \\end{itemize}  Calculations for a steady state population (constant star formation; Sect.\\ 3.5) at solar metallicity predict the following relative contributions to the \\al\\ production: $\\sim 9 \\%$ from stars before the WR phase, $\\sim 33 \\%$ from WR stars, $\\sim 14 \\%$ from SN~Ib, and $\\sim 44 \\%$ from SN~II. The large contribution from stellar wind ejection ($\\sim$ 42 \\%) confirms earlier studies of MAPP97\\nocite{MAPP97} and \\cite{knoedl99} using similar yields, who predict contributions of 20--70 \\% and $\\sim$ 40 \\% respectively. For \\fe\\ we estimate that $\\sim 39 \\%$ are produced by SN~Ib while $\\sim 61 \\%$ come from SN~II. Normalising on the total ionising flux of the Galaxy, we predict total production rates of $1.5 \\Msol\\ $Myr$^{-1}$ and $0.8 \\Msol\\ $Myr$^{-1}$ for \\al\\ and \\fe, respectively. This corresponds to 1.5 \\Msol\\ of \\al\\ and 1.7 \\Msol\\ of \\fe\\ in the present interstellar medium.  As for other chemical evolution models, our calculations depend directly on the adopted nucleosynthetic yields, which are affected by considerable uncertainties (see e.g.\\ Prantzos 1999\\nocite{Pran99}). The main uncertainties regarding \\al\\ and \\fe\\ have been discussed in Sects.\\ 2 and 4. In fact important new insight especially on the physics of supernovae is expected from the study of radioactive isotopes such as \\fe\\ and $^{44}$Ti, which are synthesised in deep layers close to the so--called mass cut separating the outer regions from the remnant. Such studies should also benefit from the present models.  Last, but not least, we have presented a Bayesian approach to quantify the predicted observables and their uncertainty related to richness effects of the IMF in terms of probability density functions (Sect.\\ \\ref{sec:realistic}). Subsequently these functions can be used in combination with prior knowledge on observed objects (e.g.\\ age, distance, and their uncertainties) to calculate detection probabilities and alike quantities. We have already successfully applied our models to existing multi wavelength observations of the Cygnus and Vela regions. The results will be published in companion papers (Kn\\\"odlseder et al., in preparation; Lavraud et al., in preparation). Our tools will be ideal to fully exploit the gamma--ray line observations expected from the upcoming {\\em INTEGRAL} satellite."
    },
    "0010/astro-ph0010147.txt": {
        "abstract": "Australian research into variations of the cosmic ray flux arriving at the Earth has played a pivotal role for more than 50 years.  The work has been largely led by the groups from the University of Tasmania and the Australian Antarctic Division and has involved the operation of neutron monitors and muon telescopes from many sites.  In this paper the achievements of the Australian researchers are reviewed and future experiments are described.  Particular highlights include: the determination of cosmic ray modulation parameters; the development of modelling techniques of Ground Level Enhancements; the confirmation of the Tail-In and Loss-Cone Sidereal anisotropies; the Space Ship Earth collaboration; and the Solar Cycle latitude survey. ",
        "introduction": "\\label{sec:1}During the Second World War the Physics Department of the University of Tasmania was heavily involved in the production of optical elements for Australia's defense effort. The Physics Department established the Optical Munitions Annexe which grew to about 200 staff producing roof prisms and photographic lenses. Toward the end of the war it was recognized that there would be an influx of mature age students to the University and preparations were made to accommodate these returned servicemen. In 1945 A.G. (Geoff) Fenton was recalled from his position in charge of quality control at the Optical Munitions Annexe to the Physics Department to develop lecture and laboratory courses.  He taught himself the necessary glassblowing and electronic techniques to build Geiger M\\\"{u}ller counter tubes for laboratory experiments in nuclear physics involving radioactivity. This in turn led to an interest in cosmic rays which make up the largest fraction of the natural background radiation.  For an historical account of the period see A.G. Fenton (2000) and references therein.  From these beginnings a program of observation and discovery of over 50 years has grown. The research based in Tasmania has played, and continues to play, a significant role in our understanding of cosmic radiation.  In the following sections we will look at some of the highlights of that research with particular emphasis on more recent discoveries and plans to continue the legacy into the future. ",
        "conclusions": "\\label{sec:9}A rather small group of researchers based in Hobart over the last half century has had a very great impact on the field of cosmic ray modulation research.  They have been active in neutron monitor, surface and underground muon telescope and airshower experiments and have played an important role in every aspect of the field. Most have now retired but their legacy continues through the impact of their work and their dedication to keep equipment operating for long enough timebases that we may start to understand effects on timescales of a lifetime or more. The research and the researchers are rightly held in high regard worldwide and those very few of us who follow in their footsteps have a sound foundation from which we may continue the journey of discovery they started."
    },
    "0010/astro-ph0010027_arXiv.txt": {
        "abstract": "The time domain is under-exploited in an astrophysical context, because the large photographic plate collections are almost never consulted by young astronomers. This is a result of lack of education of the more recent generations of astronomers, and of the deluge of terabyte-size datasets produced by new observations. Yet astronomical plates represent a unique set of resources and their quality does {\\bf not} improve with time. We point out valuable scientific projects that can be conducted with archival images and describe a first step to realize this at the Uccle Observatory, in Belgium. ",
        "introduction": "Modern astronomy has become a multi-frequency discipline, where simultaneous observations from the $\\gamma$-rays to radio are used to disentangle the physics of celestial sources. Such observations are obtained by ground or space-based platforms and are available to the investigator almost immediately, in a digital format. Modern astronomy has become a science domain where investigations are conducted by manipulating large data files on computers.  Clearly, this was not always the case. Permanent records have always been kept by astronomers, from scratches on pre-historic animal bones indicating the phases of the Moon, through stellar catalogs of Greek and Roman astronomers (culminating with Ptolemy's {\\it Megale Syntaxis} compilation)  and Galileo's telescope drawings. Since more than a century ago, semi-permanent impartial records have been available in the form of photographic plates. The first photograph of a comet (Donati in 1858) is almost 1.5 centuries old (Pasachoff \\& Olson 1995). Most of the images of the sky more than one decade old are not available digitally, with the very few exceptions of the large sky surveys. The scanning and subsequent calibration of the Palomar Sky Surveys (POSS-I and POSS-II) yielded very useful data sets, such as the Guide Star Catalog and the Digitized Sky Survey (DSS and XDSS). These digital images can now be compared with images taken in other spectral bands.  Modern astronomy is, therefore, multi-spectral in that it combines optical information with that collected in other bands. But while the recently-obtained data are available electronically and can be used to explore physical processes currently taking place in the target objects or recent changes in such objects, any long-term behavior remains largely unexplored.  Yet the world astronomical archives contain today more than two million photographic plates, each covering a sky region of one square degree or wider. Most of the plates are relegated to hardly-used astronomical archives; some even represent obstacles to the development of existing observatories, or belong to archives of ``defunct'' observatories. The purpose of this paper is to emphasize the scientific potential of such a rich heritage, to give some examples of recent usage and of possibly interesting future usage, to expound on ways to provide the information to the astronomical community, and to explain about the first steps taken in this direction by any interested group. ",
        "conclusions": ""
    },
    "0010/astro-ph0010357_arXiv.txt": {
        "abstract": "A new tool is proposed for finding out the completeness limit in apparent magnitude of a magnitude-redshift sample. The technique, closely related to the statistical test proposed by Efron \\& Petrosian (1992), presents a real improvement compared to standard completeness tests. Namely, no a priori assumptions are required concerning the redshift space distribution of the sources. It means in particular that neither the clustering nor the evolution of the mean number density of the galaxies do affect the result of the search. ",
        "introduction": "\\label{Introduction}  Extracting the intrinsic characteristics of the galaxies population from magnitude-redshift samples (e.g. the luminosity function, the power spectrum of the spatial density fluctuations) remains one of the major concerns of observational cosmology. The task is somehow complicated due to the presence of selection effects in observation, e.g. a detection threshold in apparent fluxes. Because a part of the population is indeed not observed, standard statistical methods lead in general to biased estimate of the genuine parameters characterizing the population. While correcting for such biases is hardly feasible when intricate selection effects are at work, the problem has been fortunately handled in some special cases.  Samples complete in apparent magnitude obviously deserve mention. Indeed, magnitude-redshift data truncated to a lower flux limit has been extensively studied in the literature. Powerful methods have been developed in such a case for fitting or reconstructing the luminosity function of the galaxies population (for example the $C^-$ method of Lynden-Bell 1976, the maximum likelihood fitting technique of Sandage et al. 1979). If the sample is furthermore complete in redshift, sophisticated methods have been proposed for estimating the power spectrum of the galaxies distribution (e.g.  Heavens \\& Taylor 1997) and for inferring the cosmic velocity field from magnitude-redshift data (e.g. Rauzy \\& Hendry 2000, Branchini et al.  1999).  Flux limited magnitude-redshift samples can be built by primarily  selecting from a parent magnitude sample all the objects brighter than the adopted flux limit. In a second step is collected the redshift of each of these sources (in this case the sample is complete in redshift), or a random subselection of these galaxies. By following such a strategy of observation, the completeness in apparent magnitude is essentially warranted as long as the parent sample is itself complete up to the flux limit. However, the completeness of the parent sample is in general difficult to assess. Undesired selection effects in observation are often at work and to some extent, the selection process depends on how are defined and measured the magnitudes of the galaxies (e.g. isophotal, visual, total magnitudes), how surface brightness threshold  affects the sample, etc... (see for example Sandage \\& Perelmuter 1990, Petrosian 1976, Driver 1999). Moreover, the parent sample may have been selected using criteria involving other observables not straightforwardly related to magnitudes (e.g. diameters, fluxes in a different passband). It turns out that the completeness assumption, a crucial prerequisite for applying any fitting and recontruction methods mentioned above, must be checked thoroughly.  A classical test for completeness in apparent magnitude is to analyse the variation of the galaxies number counts in function of the limiting apparent magnitude (Hubble 1926). This test, which presupposes that the galaxies population does not evolve with time and is homogeneously distributed in space, is however not very efficient. It is difficult to decide in pratice whether the deviation of the counts law is due to the presence of clustering and evolution of the galaxies luminosity function, or is indeed created by incompleteness in apparent magnitude. Including the redshift information, the $V/V_{\\rm max}$ test of Schimdt (1968) has been also used for assessing the completeness of magnitude-redshift samples (see for example Hudson \\& Lynden-Bell 1991), but suffers unfortunately from the same major drawback than the Hubble completeness test.  Efron \\& Petrosian (1992) have thoroughly analysed the statistical properties of magnitude-redshift samples complete in apparent magnitude. They proposed a robust test for independence, free of assumption concerning the spatial distribution of sources, which allows to estimate the cosmological parameters characterizing the geometry of the Universe from a quasar sample (see also Efron \\& Petrosian 1999). It turns out that this statistical test, given a world model, can easily be recycled for testing the completeness assumption. The purpose of the present paper is exactly to take advantage of such a possibility.  The statistical background of the method as well as the test for completeness are presented section 2. An example of application is given in section 3, where the test is used for investigating the completeness in apparent magnitude of the South Sky Redshift Survey  of da Costa et al. (1998). The properties of the new completeness test and its range of application are finally summarised in section 4. ",
        "conclusions": "A new tool has been  proposed for assessing the completeness in apparent magnitude of a magnitude-redshift sample. The technique presents a real improvement compared to standard completeness tests. Namely, no a priori assumptions are required concerning the redshift space distribution of the sources. It means in particular that neither the clustering nor the evolution of the mean number density of the galaxies do affect the result of the search.  The test for completeness presented herein is however reliable if and only if the magnitude-redshift sample verifies the following criteria: \\\\{\\i}) The distances of the galaxies are required, which implies that a cosmological world model has to be specified and that the contribution of peculiar velocities to observed redshifts can be safely neglected. Furthermore, the galactic extinction and the k-correction involved in the definition of the absolute magnitudes are required. \\\\{\\i}{\\i}) The shape of the luminosity function of the galaxies is not allowed to change with time (in practice, this criterion is achieved if the sample is divided  into thin intervals of redshift, i.e. time). \\\\{\\i}{\\i}{\\i}) The luminosity function of the population is assumed to be independent on the spatial position of the galaxies. In particular, environmental effect may affect the results of the completeness test."
    },
    "0010/astro-ph0010482_arXiv.txt": {
        "abstract": "We present a study, done with the Australian LBA, of  \\HI\\ absorption  for two compact radio galaxies (PKS~1549-79 and PKS~1814-63).  In both the radio galaxies, the \\HI\\ appears to give us information about the environment in which the radio sources are embedded, the effect that the ISM can have on the observed characteristics and the possible presence of interaction between the ISM and the radio plasma. ",
        "introduction": "\\HI\\ absorption is often detected in compact/small radio galaxies.  Indeed, the scale of the radio continuum in these objects represents an effective background against which the absorption can be observed.  Although the \\HI\\ absorption is often interpreted as due to the presence of a nuclear torus/disk, the situation can be more complicated in some (or perhaps many) cases.  For example, extended regions of ionised gas have been imaged around compact radio galaxies (e.g.  with HST, see Axon et al.\\ 2000).  If this gas is also a tracer of the neutral gas, this means that \\HI\\ absorption can also be produced in the regions around the radio lobes and not only against the nucleus.  Moreover, compact radio galaxies (e.g.\\ Compact Steep Spectrum sources) are known to have broad (forbidden) optical emission lines likely due to turbulent motions as result of the interaction between the radio plasma and the ISM (Gelderman \\& Whittle 1994).  This process may ``pile up'' gas at the edge of the lobes, thus making the detection of \\HI\\ absorption more likely (see e.g.\\ IC~5063, Oosterloo et al.\\ 2000).  In some cases (although a minority) star formation and far-IR emission are observed.  Thus, these objects could have a particularly rich ISM that could produce the \\HI\\ absorption in/around their radio lobes.  A further complication in the interpretation of the \\HI\\ absorption is the uncertainty in the values of the systemic velocity as obtained from optical data.  To unambiguously interpret the kinematics of the neutral gas, better systemic velocities than available in literature are usually necessary.  We have recently carried out (with the VLA and the ATCA) a survey looking for HI absorption in radio galaxies: 23 galaxies have been observed so far (Morganti et al.\\ 2000a).  Two of them are compact radio galaxies and show \\HI\\ absorption (PKS 1549--79 and PKS 1814--63).  Thus, they have been followed up with high resolution observations with the Australian LBA. ",
        "conclusions": ""
    },
    "0010/astro-ph0010607_arXiv.txt": {
        "abstract": "The evolution of an initially stellar dipole type magnetosphere interacting with an accretion disk is investigated numerically using the ideal MHD ZEUS-3D code in the 2D-axisymmetry option. Depending mainly on the magnetic field strength, our simulations may last several thousands of Keplerian periods of the inner disk. A Keplerian disk is assumed as a boundary condition prescribing a mass inflow into the corona. Additionally, a stellar wind from a rotating central star is prescribed. We compute the innermost region around the stellar object applying a non-smoothed gravitational potential.  Our major result is that the initially dipole type field develops into a spherically radial outflow pattern with two main components, a disk wind and a stellar wind component. These components evolve into a quasi-stationary final state. The poloidal field lines follow a conical distribution. As a consequence of the initial dipole, the field direction in the stellar wind is opposite to that in the disk wind. The half opening angle of the stellar wind cone varies from $30\\degr$ to $55\\degr$ depending on the ratio of the mass flow rates of disk wind and stellar wind. The maximum speed of the outflow is about the Keplerian speed at the inner disk radius.  An expanding bubble of hot, low density gas together with the winding-up process due to differential rotation between star and disk disrupts the initial dipole type field structure. An axial jet forms during the first tens of disk/star rotations, however, this feature does not survive on the very long time scale. A neutral field line divides the stellar wind from the disk wind. Depending on the numerical resolution, small plasmoids are ejected in irregular time intervals along this field line. Within a cone of $15\\degr$ along the axis the formation of small knots can be observed if only a weak stellar wind is present.  With the chosen mass flow rates and field strength we see almost no indication for a flow self-collimation. This is due to the small net poloidal electric current in the (reversed field) magnetosphere which is in difference to typical jet models. ",
        "introduction": "A stellar dipole type magnetic field surrounded by an accretion disk is the common model scenario for a variety of astrophysical objects. Examples are the classical T\\,Tauri stars, magnetic white dwarfs (in cataclysmic variables) and neutron stars (in high mass X-ray binaries). Part of these sources exhibit Doppler shifted emission lines indicating wind motion. Highly collimated jets have been observed from young stellar objects and X-ray binaries, where also quasi-periodic oscillations (QPO) are observed. In general, magnetic fields are thought to play the leading role for the jet acceleration and collimation (Blandford \\& Payne 1982; Pudritz \\& Norman 1983; Camenzind 1990; Shu et al. 1994; Fendt et al. 1995).  Recently, several papers were considering the evolution of a stellar magnetic dipole in interaction with a diffusive accretion disk. Hayashi et al. (1996) observed magnetic reconnection and the evolution of X-ray flares during the first rotational periods. Miller \\& Stone (1997) Goodson et al. (1997) included the evolution of the (diffusive) disk structure in their calculation. In these papers a collapse of the inner disk is indicated depending on the magnetic field strength and distribution. The inward accretion flow develops a shock near the star. The stream becomes deflected resulting in a high-speed flow in axial direction.  The results of Goodson et al. (1997, 1999) and Goodson \\& Winglee (1999) are especially interesting since combining a huge spatial scale (2\\,AU) with a high spatial resolution near the star ($0.1\\rsun$). However, to our understanding it is not clear, how the initial condition (a standard $\\alpha$-viscosity disk) is really developing in their code without any physical viscosity. Secondly, not very much is said about the amount of magnetic diffusivity. The assumption of constant diffusivity cannot really reflect the two component model of disk and coronal out flow.  Time-dependent simulations lasting only a short time scale depend strongly on the initial condition and the calculation of the evolution of such a magnetosphere over {\\em many} rotational periods is an essential step. In particular, this is important point if the initial condition is not in equilibrium. In summary, we note that all calculations including the treatment of the disk structure could be performed only for a few Keplerian periods of the inner disk (and a fraction of that for the outer disk).  At this point, we emphasize that the observed kinematic time scale of protostellar jets can be as large as $10^3 - 10^4$\\,yrs, corresponding to $5\\times 10^4 - 5\\times 10^5$ stellar rotational periods (and inner disk rotations)! For example, proper motion measurements for the HH30 jet (Burrows et al. 1996) give a knot velocity of about $100-300\\,{\\rm km\\,s^{-1}}$ and a knot production rate of about 0.4 knot per year. Assuming a similar jet velocity along the whole jet extending along 0.25\\,pc (Lopez et al. 1995), the kinematic age is about 1000 yrs.  A different approach for the simulation of magnetized winds from accretion disks considers the accretion disk `only' as a {\\em boundary condition} for the mass inflow into the corona. Since the disk structure itself is not treated, such simulations may last over hundreds of Keplerian periods. This idea was first applied by Ustyugova et al. (1995). Extending this work, Romanova et al. (1997) found a stationary final state of a slowly collimating disk wind in the case of a split-monopole initial field structure after 100 Keplerian periods. Ouyed \\& Pudritz (1997, hereafter OP97) presented time-dependent simulations of the jet formation from a Keplerian disk. For a certain (already collimating) initial magnetic field distribution, a stationary state of the jet flow was obtained after about 400 Keplerian periods of the inner disk with an increased degree of collimation. In a recent extension of their work both groups were considering the influence of the grid's shape on the degree of collimation (Ustyugova et al. 1999) and the effect of the mass flow rate (Ouyed \\& Pudritz 1999). Ouyed \\& Pudritz (2000) investigate the problem of jet stability and magnetic collimation extending the axisymmetric simulations to 3D.  In this paper, we are essentially interested in the evolution of the ideal MHD magnetosphere and the formation of winds and jets and not in the evolution of the disk structure itself. Therefore, we do not include magnetic diffusivity into our simulations. The disk acts only as a boundary condition for the corona/jet region. The winding-up process of poloidal magnetic field due to strong differential rotation between the star and the disk would always be present even if a disk diffusivity would have been taken into account. The disk diffusivity will never be so large that a rigid rotation of the magnetosphere in connection with the disk can be maintained.  In that sense our simulations represent an extension of the OP97 model, taking additionally into account the central star as a boundary condition and a stellar dipole type field as initial condition. First results of our simulations were presented in Fendt \\& Elstner (1999, hereafter FE99). Here we give a more detailed discussion together with new results. Movies of our simulations will be provided under {\\sl http://www.aip.de/$\\sim$cfendt}. ",
        "conclusions": "In the following we discuss the results of four example simulations denoted by S2, S4, L3, L5 (see Tab.\\,1). All the simulations presented in this paper consider a rotating star at the center. The stellar rotational period is chosen as $\\Omega_{\\star} = (v_{K,i}/R_i) = 1$, with a magnetospheric co-rotation radius located at the inner disk radius. The evolution of a non-rotating central star is discussed in FE99, although it was not possible to perform the simulation as long as the examples presented here.  As a general behavior, the initial dipole type structure of the magnetic field disappears on spatial scales larger than the inner disk radius and a two component wind structure -- a disk wind and a stellar wind -- evolves. Our main result is the finding of a {\\em quasi-stationary} final state of a spherically radial mass outflow evolving from the initial dipole type magnetic field on the very extended time evolution. For the boundary conditions applied the calculated flow structure show only few indication for collimation.  The general features in the evolution of the system are independent from a variation of the field strength. For strong fields, the evolution is faster. Thus, for simulations which are limited in time due to numerical problems, a decrease in the field strength would not help. Although the numerical life time of the simulation would be accordingly longer, the result for the final time step will be the same.  \\subsection{Four simulations of the long-term magnetospheric evolution -- an overview} The four example simulations basically differ in the mass flow rates from disk and star and the size of the physical domain investigated (Tab.\\,1). All other parameters, $\\beta_i = 1.0 $, $\\delta_i= 100$, $\\mu=-1.0$, $\\rs = 0.5$, $\\gamma = 5/3$, and the numerical mesh of $250\\times 250$ grid elements remain the same. The first simulation (solution {\\bf S2}) considers a rectangular box of physical size of 20x20 inner disk radii (Fig.\\,2, Fig.\\,5, Fig.\\,7). The stellar wind mass flow rate is comparatively large, $\\dot{M}_{\\star}/ \\dot{M}_{\\rm D} = 2$. The stellar wind injection velocity is very low in order to not disturb a possible weak wind solution already by the boundary condition. The disk wind boundary condition is the same as in OP97 and FE99. Due to the relatively large physical size of the computational domain the stability of the initial condition can be observed for several decades of rotational periods. The torsional Alfv\\'en waves leave the grid after about $t= 40$. The second simulation (solution {\\bf S4}) considers a rectangular box of physical size of 10x10 inner disk radii (Fig.\\,3, Fig.\\,5, Fig.\\,7). Now the total mass flow is dominated by the disk wind . This choice of parameters can be directly seen from the simulation by comparing the size of axial flow and the 'bubble' evolving from the disk flow. Clearly, the disk flow is more prominent. The third (solution {\\bf L3}) considers a rectangular box of physical size of only 5x5 inner disk radii (Fig.\\,4, Fig.\\,6). Similar to solution S4, the total mass flow is dominated by the disk wind. This high resolution simulation zooms into the innermost region around the star. In particular, the neutral line is clearly resolved. We finally discuss another example ({\\bf L5}) in which the stellar wind is dominating the disk wind. This simulation perfectly evolves into final stationary state. For L5 the velocity injection profile is chosen differently from the examples discussed above in order to increase the disk flow magnetization. All parameter runs show a similar gross behavior indicating that our run S2 with lowest resolution is sufficient in order to investigate the main features of the flow evolution. In general, a high stellar wind mass loss rate will stabilize the outflow.   \\subsection{The first evolutionary stages} During the first stages of the long-term evolution the magnetospheric structure is characterized by the following main features: The {\\em winding-up} of the dipolar poloidal field, the formation of a {\\em neutral field line}, a {\\em transient axial jet} feature, a two component outflow consisting of a {\\em stellar wind} and a {\\em disk wind}.  \\subsubsection{Winding-up of the poloidal field} The winding-up process of poloidal magnetic field due to differential rotation between star and disk and the static initial corona induces a toroidal field (Fig.\\,4) with a positive sign along field lines located outside the slowly emerging neutral field line. Inside the neutral field line $B_{\\phi}$ has negative sign. This is in difference to OP97 and other simulations assuming a monotonously distributed initial field.  Torsional Alfv\\'en waves propagate outwards distorting the initial field structure. After about $t=40$ these waves reach the outer boundary (Fig.\\,2). The region beyond the wave front remains completely undisturbed. The region between the Alfv\\'en wave front and the flow bow shock is adjusted to a new equilibrium and also remains in equilibrium until it is reached by the generated outflow ( See the density contour lines close to the disk in Fig.\\,A2. The grey scale density plots cannot show this feature ). Within this Alfv\\'en wave front the magnetic field becomes distorted from its initial force-free state (compare the field line structure of the closed loops in Fig.\\,2 for $t \\leq 30$). The distortion of the force-free field due to propagating Alfv\\'en waves results in Lorentz forces initiating an axial `jet' feature close to the axis. As we will see later, this axial jet, however, is a transient feature.  The winding-up of poloidal magnetic field seems to be similar to the effect proposed by Lovelace et al. (1995). However, in our case, this process is initiated by the differential rotation between star and hydrostatic corona. Only later, the wound-up toroidal field is maintained by both, differential rotation between star and disk and the inertia of the outflow.  \\subsubsection{A neutral field line dividing stellar and disk field} The wound-up magnetic field lines stretch forming a neutral field line of vanishing field strength. The matter around this field line is distributed in a layer of low density (Fig.\\,4). Around this layer an expanding `bubble' is formed due to the additional magnetic pressure due to toroidal fields which disrupts the initial dipolar field structure (Fig.\\,3). When the bubble has left the grid, the field lines which are separated by the the neutral line remain disconnected. This is due to the differential rotation between star and disk. The actual appearance of the axial jet and the low density bubble depends mainly on the mass flow rates from disk and star. The bubble is most prominent in simulation S4 where the disk mass flow rate is largest.  \\subsubsection{A transient axial jet feature} In the beginning of the simulations a jet feature evolves along the rotational axis. Its pattern velocity is about $0.2\\,v_{\\rm K,i}$ (S2) or $0.3\\,v_{\\rm K,i}$ (L3). Such an axial jet is known as a characteristic result of MHD simulations performed in the recent literature (Hayashi et al. 1997; Goodson et al. 1997, 1999; Goodson \\& Winglee 1999; Kudoh et al. 1998). It is often claimed that this feature is connected to the real (protostellar) jets observed on the AU-scale. Apart from the fact that the spatial dimension and velocity do not fit with the observations, we will see that the formation of this feature is a result of the adjustment process of the initially hydrostatic state to a new dynamic equilibrium and will disappear on the long-term evolution.  Winding-up of the initially force-free magnetic field by differential rotation during the first time steps immediately leads to a not force-free field configuration. The resulting magnetic forces accelerate the material of the initially hydrostatic corona forming an axial jet. This process works as long as initially distributed coronal matter is present at this location. As time develops the jet feature becomes weaker and weaker. Disk wind and the stellar wind become the dominant flow pattern and the axial jet dies out after about $t>100$ (Fig.\\,2). In comparison, in simulation L1 (FE99), where a stellar wind is absent, the coronal density along the axis decreases until it is below the numerically critical value and the simulation stops.  The intermittent character of the axial jet flow is best seen in the velocity structure (Fig.\\,5). The velocity vectors of the axial flow are largest during the first time steps. However, after sweeping-out the initial corona, a weak wind flow from the stellar surface succeeds the jet.  \\subsubsection{The disk wind} The disk wind accelerates within tens of grid elements from its low injection speed to fractions of the Keplerian speed. The acceleration mechanism is mainly due to the centrifugal force on the disk matter reaching the non-rotating corona, where the gravitational force is balanced by the pressure and not by a centrifugal force as in the disk. The flow is already super Alfv\\'enic due to the weak dipolar field strength in the disk. Thus, magneto-centrifugal acceleration along the inclined dipole type field lines of the initial magnetic field is {\\em not} the acceleration mechanism. More above the disk also the Lorentz force along the field contributes to the acceleration (see Sect.\\,4.4.3.). The inclination angle between field lines (equivalent to the outflow direction) and the disk depends on the mass flow rate. For the parameter range investigated we see no indication for a disk wind collimation because the Lorentz force points away from the axis (see below).  \\subsubsection{The stellar wind} The rotating stellar magnetosphere generates a stellar wind. Due to the strong magnetic field close to the star the flow starts sub-Alfv\\'enic. It is initially magneto-centrifugally driven with a roughly spherical Alfv\\'en surface located at $1.5\\ri$ (L3) or closer (S2, S4, L5) to the stellar surface. The most dominant flow pattern of the stellar wind is in the part with the widest opening angle (Fig.\\,5). Although the Lorentz force points radially inwards no collimation is observed because of a strong pressure gradient. Depending on the mass loss rates the stellar wind evolves faster or slower than the disk wind (Fig.\\,5).  \\setlength{\\unitlength}{1mm} \\begin{figure} \\parbox{200mm} \\thicklines \\epsfysize=40mm \\epsfxsize=40mm \\caption {Evolution of the poloidal velocity in the simulations S2 ({\\it left}) and S4 ({\\it right}). Time steps (from {\\it top} to {\\it bottom}). $t = 10, 20, 100, 500, 2500 $ (S2) and $t = 5, 10, 20, 100, 600 $ (S4). Vectors scale only within each frame. } \\end{figure}  \\setlength{\\unitlength}{1mm} \\begin{figure} \\parbox{190mm} \\thicklines \\epsfysize=49mm \\epsfxsize=40mm \\caption {Highly time-resolved evolution of simulation L3. Poloidal magnetic field lines ({\\it left}) and density contours ({\\it right}). Time steps $t = 200, 201, 202, 203, 204 $ (from {\\it top} to {\\it bottom}). } \\end{figure}   \\subsection{The long-term evolution} The long-term evolution of the flows depends critically on the choice of inflow boundary conditions into the corona. The total mass flow rate into the corona determines how fast the flow will establish a (quasi-)stationary state. The stellar wind - disk wind mass flow ratio determines (i) the opening angle of the outflow, (ii) the opening angle of the cone of the neutral line which is the boundary layer between the stellar wind and the disk wind, and (iii) the stationarity of the axial flow (see Tab.\\,2). In general the outflow undergoes a highly time-variable and turbulent evolution. However, after relaxation of the MHD system from the initial magneto hydrostatic state into the new {\\em dynamical} equilibrium, we observe an outflow from disk and star distributed smoothly over the whole hemisphere and moving predominantly in spherically radial direction.  In simulation S2 the flow structure is highly time-variable over many hundreds of periods. The intermediate region between the two components -- stellar wind and disk wind -- is characterized by turbulent motions of very low velocity. But also the disk wind seems to be unstable. During the intermediate time evolution only the flow pattern along the $z$-axis relaxes. \\clearpage Compared to the evolution of the stellar wind, the disk wind needs definitely more time to establish a stationary structure. However, after all these turbulent evolutionary steps, with few changes in the general appearance of the flow pattern over hundreds of rotational periods, {\\em after about 2000-2500 rotations a quasi-stationary outflow has been established} over the whole grid (Fig.\\,2). Only the region around the neutral line dividing stellar and disk wind and the region along the $z$-axis is subject to small scale instabilities. Interestingly, the flow along the symmetry axis which shows a stable behavior during the intermediate time evolution, finally becomes unstable. A conical flow consistent of a knotty structure evolves with a full opening angle of $30\\degr$. At this time the opening angle of the neutral layer cone dividing stellar wind from disk wind has been increased compared to the intermediate time steps. This de-collimation of the outer flow causes a de-stabilization of the axial flow. In this sense, the stellar wind flow is stabilized by the ambient (disk wind) pressure. The formation of knots and instabilities along the symmetry axis depends on the stellar wind mass flow rate (see below). The full opening angle of `knot flow` is about 30\\degr to 50\\degr. We emphasize that due to the knot size and knot spacing, these features are not correlated to the observed knots of protostellar jets. It seems more related to QPO's observed in X-ray binaries. The knot velocity is about 10\\% of the Keplerian speed at the disk inner radius.  In simulation S4 the quasi-stationary state is reached earlier after about 200 inner disk rotations according to the smaller box size (Fig.\\,3). The region around the neutral field line is resolved better. We observe the formation of a wavy structure with an amplitude of $0.5\\,\\ri$ and a wavelength of $\\ri$. These `waves' travel outwards and leave the grid. The wave generation is somewhat arbitrary in time. Whereas the region enclosing the neutral layer first seems to have reached a steady state at about $t=100$, at $t=600$ the wave structure evolves again. This is related to the evolution along the symmetry axis. Here, in the contrary, at $t=100$ the simulation still showed a wavy pattern, whereas at $t=600$ a smooth steady state has evolved. The flow structure at $t=400$ seems to be completely smooth and stable. The long-term evolution shows that this stability is in fact a transient feature as far as the neutral layer region is concerned (see below).  Simulation L3 shows that the neutral line has a complex structure. Two current sheaths emerge, one from the stellar radius, the other from the inner disk radius (Fig.\\,4). These are indicated by the density and poloidal magnetic field `islands' emitted along this field line (Fig.\\,4, time step 80 and 160). Figure 6 shows the evolution of the solution L3 with a high time resolution $\\Delta t = 1$ after $t=200$. At this time the simulation has not yet evolved into a quasi-stationary state. It can be seen how plasmoids are formed and move outwards along the neutral line. Most probably, this would be a region of on-going reconnection processes. A similar behavior was found first by Hayashi et al. (1997) including also magnetic diffusion in their treatment. However, the simulation lasted only for one inner disk rotational period (with a star at rest). Our long-term simulations show that the formation of such plasmoids will continue. We do not believe that the lack of diffusion in our treatment is a serious problem concerning this point because the time scale given by the plasmoid velocity is smaller than the time scale given by magnetic diffusion.  Figure 5 shows the poloidal velocity vectors of the simulations S2 and S4 at selected time steps (compare to Fig.\\,3 and 4). The general feature is that the axial jet feature seen in the first time steps disappears. The outflow exhibits a two-component structure. Depending on the inflow density profile the `asymptotic' (i.e. close to the grid boundaries) velocity profile changes slightly. High velocities (larger than $v_{\\rm K,i}$) are only observed far from the axis. For solution S2 we obtain an asymptotic speed of $1.5\\,v_{\\rm K,i}$ for both components with a velocity profile decreasing across the neutral field line. The flow velocity along the axis is $0.2\\,v_{\\rm K,i}$. For solution S4 the asymptotic speed of $1.5\\,v_{\\rm K,i}$ for the stellar wind component and with a velocity profile increasing across the neutral field line $2\\, v_{\\rm K,i}$. The latter is due to the low density at this point. The disk wind has not yet accelerated to high velocities. This is due to the wide opening angle of the stellar wind cone and the small physical grid size. The flow velocity along the axis about $0.5\\, v_{\\rm K,i}$. For solution L5 the `asymptotic' speed of the stellar wind is low and $\\leq v_{\\rm K,i}$. The velocity profile is in general similar to that of solution S4.  The duration of our simulation runs (L3, L5, S2, S4) is not limited by numerical reasons. They have been terminated when the flow evolution has reached a {\\em quasi} stationary state. For simulation S2 this means that after about 2500 rotations the main pattern in the flow evolution does not change anymore. In particular, the outer part of the wind flow does not vary in time, while knots are still generated along the axis. For simulation S4, after 400 time steps most of the flow region is in a stationary state. Only the wavy structure along the neutral field line, which is also connected to the formation of knots, continuously evolves and disappears. This structure is {\\em not} a dangerous instability for the outflow (see also Fig.\\,7). For solution L3, the simulation has been terminated when parts of the flow (the outer disk wind flow) had reached a stationary state. The inner solution close to the axis does not reach such a stationary state. In this case, our intention is to investigate the neutral line with better spatial resolution. Simulation L5 was terminated some time after the stationary state has been reached for the whole outflow.  \\subsection{A stationary final state: a radial outflow evolved from an initially dipolar magnetosphere} The main result of our simulations is that the initial dipolar magnetosphere evolves into a spherically radial outflow consisting of two components. Depending on the inflow parameters (mass flow rates, magnetic field strength) our simulations reach a quasi-stationary state. A weak non-stationarity may be present along the neutral field line, which is dividing the stellar wind from the disk wind. Also, for a weak stellar wind flow a turbulent flow pattern may evolve along the axis. Such outflows we call {\\em quasi-stationary}, if the main flow pattern does not change in time. The disk wind and the outer cone of the stellar wind reach a kind of stationary state, where the density profile and field line structure remain almost constant in time.  For the S2 solution the half opening angle of stellar wind cone is about $55\\degr$. This angle remarkably changes during the flow evolution. During the initial evolutionary decades the turbulent region between the stellar wind and the disk wind collimated the stellar wind to a narrower cone. Clearly, such a neutral line is a rather unstable situation. Reconnection will most probably occur which we cannot properly treat with our ideal MHD approach. Increasing the numerical resolution (simulation L3) shows the emission of plasmoids along the neutral line (see below). Figure 7 shows an overlay of three time steps of solution S2 at $t= 2600, 2650, 2700$ clearly indicating the stationary character of disk wind and most parts of the stellar wind together with the non-stationary axial flow and the small-scale wave pattern along the neutral field line.  \\begin{table} \\caption[]{Terminal poloidal velocity of the wind components from the disk $v_{\\rm p, disk}^{\\rm max} $, from the star $v_{\\rm p,star}^{\\rm max} $, along the axis $v_{\\rm p, axial} $, and from the gap $v_{\\rm p, gap}^{\\rm max} $, time when stationary state has been reached $t_{\\rm s}$, and the inclination angle between disk and neutral layer $\\alpha$ for the different simulations.} \\label{table2} \\begin{tabular}{lllllll} \\hline\\hline \\noalign{\\smallskip} & $v_{\\rm p, disk}^{\\rm max}  $ & $v_{\\rm p,star}^{\\rm max} $ & $v_{\\rm p, axial} $ & $v_{\\rm p, gap}^{\\rm max} $ & $t_{\\rm s}$ & $\\alpha $ \\\\ \\hline \\noalign{\\smallskip} {\\bf L1} & 1.0 & ? & ?  & ? & ? & ? \\\\ \\noalign{\\medskip} {\\bf S2} & 1.0 & 1.5 & 0.2 & 1.5 & 2500   & $35\\degr$ \\\\ \\noalign{\\medskip} {\\bf S4} & 1.5 & 2.1 & 0.6 & 1.5 & 400 & $60\\degr$  \\\\ {\\bf L3} & 1.0 & 1.3 & 0.2 & 1.7 & ?   & $50\\degr$  \\\\ {\\bf L5} & 0.9 & 1.0 & 0.2 & 3.0 & 150 & $35\\degr$  \\\\ \\noalign{\\smallskip} \\hline\\hline \\noalign{\\medskip} \\end{tabular}  \\end{table}  In the case of solution S4 the quasi-stationary state is reached earlier after about $t=400$ (Fig.\\,3). The comparatively large stellar wind mass flow rate stabilizes the flow along the axis and no turbulent pattern evolves. On the other hand, due to the low disk wind mass flow rate the half opening angle of the neutral line cone is smaller. It is $35\\degr$ compared to $55\\degr$ for solution S2. Again, Fig.\\,7 demonstrates the stationarity of the simulation with an overlay of three time steps $t= 575, 600, 625$. In comparison with solution S2 now the whole flow pattern is stationary. In particular the region along the axis remains stable. The only time-dependent feature is the neutral line exhibiting a slowly variable wave structure.  \\setlength{\\unitlength}{1mm} \\begin{figure*} \\parbox{100mm} \\thicklines \\epsfysize=86mm \\epsfxsize=70mm \\put(0,75){\\epsffile{9822f7a.ps}} \\put(80,75){\\epsffile{9822f7b.ps}} \\caption {Evolution of simulation S2 (upper) and S4 (lower) on the very long time scale. Shown are overlays of the poloidal field lines ({\\it left}) and density contours ({\\it right}). Three time steps are superposed, $\\tau = 2600, 2650, 2700$ (S2) and $\\tau = 575, 600, 625$ (S4). } \\end{figure*}  Increasing (i) the total mass flow rate and (ii) also the stellar wind to disk wind mass flow ratio will result in an almost perfectly stationary flow (solution L5; see below). Having a similar mass flow ratio, also the opening angle is similar to solution S2. The stationarity of solution L5 will be investigated in more detail below.  The blobs (or rather tori) generated in simulation S2 move with pattern speed of about 0.1 the Keplerian speed at $r_i$. Their size is about the inner disk radius but depends from the mass flow ratio and the numerical resolution. We emphasize that due to the knot size and time scale of knot formation in our simulation, their connection with the jet knots observed in protostellar jets on the large scale distance of tens of AU is questionable. This statement also holds for comparable structures observed in similar simulations presented in the literature (Ouyed \\& Pudritz 1997b, Goodson et al. 1997, 1999; Goodson \\& Winglee 1999).  On the other hand, from the unstable character of the axial flow together with the lack of collimation we may conclude that the model configuration investigated in our paper is unlikely to produce collimated jets. Furthermore, we hypothesize that this behavior may be the one of the reasons why highly magnetized star-disk systems -- containing magnetic white dwarfs or neutron stars -- do not have jets.  In general, the maximum terminal poloidal velocity is of the order of the Keplerian speed at the disk inner radius (see Tab.\\,2). The speed is highest along the field lines from the gap due to the low flow density. The axial flow speed is low. Its mass density depends on the injection parameters and could be relatively large (S4, L5) or low (L3, S2), but is generally less than 10\\% of the density at the inner disk radius. The maximum stellar wind speed is reached along field lines with the largest opening angle and is above the Keplerian speed at the disk inner radius (Fig.\\,5). The same holds for the maximum disk wind speed. However, in this case the maximum speed is reached along the field lines with a foot point at small radius. This is partly due to length of the acceleration distance, partly due to the rapid rotation of the disk material at small radii.  \\subsubsection{The question of dipolar accretion} In all our different simulation runs we never observe a signature of dipolar accretion as it is would be expected from models of young stellar jet formation (e.g. Camenzind 1990, Shu et al. 1994). Instead, a magnetically driven wind develops from the stellar surface. Note that this strong stellar outflow is {\\em permitted} but {\\em not prescribed} by the inflow boundary condition along the star, since the inflow velocity is very low. We emphasize that even in our simulation L1 for a non rotating star (Fendt \\& Elstner 1999), where no stellar wind can develop, no dipolar accretion occurred. Therefore, we believe that it is not the boundary condition which prevails the matter from falling from the disk to the star. In fact, dipolar accretion has never been observed in the literature of numerical MHD simulations considering the star-disk interaction (e.g. Hayashi et al. 1996, Goodson et al. 1997, Miller \\& Stone 1997), but this might also be caused by the comparatively short time evolution in those simulations.  We think that the main reason that hinders the dipolar accretion is the choice of the co-rotation radius equal to the inner disk radius. Only disk material orbiting inside the co-rotation radius could be accreted along the field lines.  It is clear that such an accretion process along converging field lines is difficult to treat numerically. Certainly, this important topic has to be investigated more deeply. We defer this to a future paper.  \\subsubsection{The question of collimation} The quasi-stationary two-component outflow obtained in our simulations show almost {\\em no indication for collimation}. The seems to be in contradiction to the literature (OP97, Romanova et al. 1997). However, the non-collimation of the flows can be explained following the analysis of Heyvaerts \\& Norman (1989). They have shown that only jets carrying a net poloidal current will collimate to a cylindrical shape. However, in our case we have an initially dipole type magnetosphere and the final state of a spherically radial outflow enclosing a neutral line with a poloidal magnetic field reversal. The toroidal field reversal also implies a reversal of the poloidal current density with only a weak {\\em net} poloidal current. In such a configuration, a self-collimation of the flow as obtained by OP97 or Romanova et al. (1997) cannot be achieved. In both of these publications a net poloidal current flows along the {\\em monotonously} distributed field lines.  One might expect to obtain the OP97 results of a collimating jet as a limiting case in the present simulations, just due to the fact that the inflow boundary condition along the disk is the same as in their setup. However, we think that this is not possible since the initial field structure and, thus, the flux distribution in the lower boundary is completely different. A combination of the OP97 field distribution together with a central dipole might do the job. Still, the problem would be the very different field strength of both components, since the dipolar field will decrease by a factor of 10-50 towards the inner disk radius. Thus, the stellar field is always dominating the numerical simulation. We defer the treatment of such a completely new numerical setup to a future paper.  Apart from this argument concerning a flow self-collimation we mention the hypothesis raised by Spruit et al. (1997) claiming that a ``poloidal collimation'' is responsible for the jet structure. Such a poloidal collimation would rely on the magnetic pressure onto the jet flow from the disk magnetic field {\\em outside} the jet. Their condition for poloidal collimation, a disk magnetic field distribution $\\bp \\sim r^{-q}$ with $q\\leq 1.3$, is clearly not satisfied in our case of a dipole type field distribution along the disk (which is conserved from the initial condition because of flux conservation). For a dynamo generated field in the disk this condition is satisfied (R\\\"udiger et al. 1995). This holds also for the disk field distribution of OP97. In this sense, our simulations are consistent with Spruit et al. (1997), although we do not argue that our results support their hypothesis that ``poloidal collimation'' is the main process to produce jets.   \\setlength{\\unitlength}{1mm} \\begin{figure} \\parbox{100mm} \\thicklines \\epsfxsize=40mm \\caption {Properties of the stationary final state for the example solution L5. The time step is $t = 200$ when a stationary state has been reached. Conserved quantities of stationary MHD: magnetic flux function $\\Psi$ (poloidal field lines), mass flux per flux surface $\\eta(r,z)\\equiv \\eta(\\Psi)$, the iso-rotation parameter $\\Omega_F(r,z)\\equiv \\Omega_F(\\Psi)$, the total angular momentum $L(r,z)\\equiv L(\\Psi)$. Density distribution $\\rho(r,z)$, poloidal electric current $j_p(r,z)$, angular velocity $\\Omega (r,z)$, toroidal magnetic field strength $B_{\\phi}$ (positive (negative) values outside (inside) the neutral layer), vectors of poloidal Lorentz force $F_{L,p}(r,z)$ and Lorentz force component parallel to the poloidal field lines $F_{L,||}(r,z)$. (Vectors are normalized to unity). } \\end{figure}  We further to note the results of Ustyugova et al. (1999) who claim that the shape of the numerical box influences the degree of collimation. A rectangular box extended along the symmetry axis would lead to an artificial flow collimation, whereas a quadratic box simulation (as used in our simulation) did not result in a collimated structure. A recent study by Okamoto (1999) also has raised strong arguments against a MHD self-collimation. In particular, he claims that {\\em electric current-closure} will inhibit a self-collimation, a point which is not always considered in MHD jet models. Current-closure, however, is satisfied in our model due to the reversed dipole type initial field.  Nevertheless, strongly collimated astrophysical jet flows are observed. For the moment we speculate that an increase of the {\\em disk field strength} would probably enhance the degree of collimation. So far we doubt whether an increase of the size of the computational box will be sufficient, because in our model the field distribution and mass flow rate decrease strongly with radius.  \\subsubsection{The final steady state - application of stationary MHD} It is a well known from standard MHD theory that an axisymmetric stationary MHD flow is defined by five integrals of motion along the magnetic flux function $\\Psi(r,z) \\equiv 2\\pi \\int \\vec{B}_{\\rm P}\\cdot d\\vec{A}$. Stationarity implies the following conserved quantities along the flux surface $\\Psi$: The mass flow rate per flux surface, $\\eta(\\Psi)\\equiv {\\rm sign}(\\vec{v}_{\\rm P}\\cdot \\vec{B}_{\\rm P}) \\,\\rho v_{\\rm P}/B_{\\rm P}$, the iso-rotation parameter $\\Omega_{\\rm F}(\\Psi) \\equiv (v_{\\phi} - \\eta B_{\\phi}/\\rho)/r $, the total angular momentum density per flux surface, $L(\\Psi) \\equiv r ( v_{\\phi} - B_{\\phi}/\\eta) $, and the total energy density $E(\\Psi)$. Therefore, for a time-dependent axisymmetric simulation evolving into a stationary state, these functions must be constant along the field lines. Figure 8 demonstrates such a behavior for the example solution L5 for the quantities $\\eta(\\Psi)$,$\\Omega_{\\rm F}(\\Psi)$ and $L(\\Psi)$. An overlay of each of these functions with the contour plot of the field lines $\\Psi(r,z)$ would show that the contours are perfectly `parallel'. This clearly proves the stationary character of the final state of simulation L5. For comparison, Fig.\\,8 shows the distribution of the poloidal electric current density, the toroidal magnetic field, and the angular velocity of the plasma.  For a stationary solution it is interesting to investigate the Lorentz force projected parallel and perpendicular to the field lines. The poloidal Lorentz force vectors (Fig.\\,8) show that in the region of the highest poloidal velocities, the Lorentz force is more or less aligned with the field lines. In the region of low poloidal velocity the main component of the Lorentz force is perpendicular to the field lines. As discussed above (Sect.\\,4.4.2.) the perpendicular component of the Lorentz force acts collimating inside the neutral layer and de-collimating outside the neutral layer.   \\setlength{\\unitlength}{1mm} \\begin{figure*} \\parbox{100mm} \\thicklines \\epsfysize=5mm \\epsfysize=60mm \\caption {Density distribution and poloidal magnetic field lines of simulation L5 for initial time step ({\\em left}) and for the final stationary state ({\\em right}) plotted for all four hemispheres. } \\end{figure*}  Another interesting feature is the {\\em direction} of the parallel component of the Lorentz force (see Fig.\\,8). Close to the disk boundary there is a region where this component changes sign and the Lorentz force is therefore {\\em decelerating} the matter\\footnote{In Fig.\\,8, this region is only one vector element wide. However, a better resolution shows that this region extends to about $z=1$ along the outer disk}. This demonstrates that in the stationary final state the Lorentz force is {\\em not} the main driving mechanism of the disk material from the disk into the corona. It is only at a larger height above the disk that the parallel component of the Lorentz force accelerates the plasma.  As a summary of this section we show in Fig.\\,9 the final stationary state of the example solution L5 plotted for all four hemispheres. This figure (and only this one) is {\\em rotated by $90\\degr$} with the $z$-axis in vertical direction. This gives a comparative look how the simulation has evolved from the initial dipolar structure to the spherically radial outflow final state."
    },
    "0010/astro-ph0010094_arXiv.txt": {
        "abstract": "We report a phase-resolved photometric and morphological analysis of $UBV$ data of the Crab pulsar obtained with the 2-d TRIFFID high speed optical photometer mounted on the Russian 6m telescope. By being able to accurately isolate the pulsar from the nebular background at an unprecedented temporal resolution (1  $\\mu$s), the various light curve components were accurately fluxed via phase-resolved photometry. Within the $UBV$ range, our datasets are consistent with the existing trends reported elsewhere in the literature. In terms of flux and phase duration, both the peak Full Width Half Maxima and Half Width Half Maxima decrease as a function of photon energy. This is similarly the case for the flux associated with the bridge of emission. Power-law fits to the various light curve components are as follows; $\\alpha$ = 0.07 $\\pm$ 0.19 (peak 1) , $\\alpha$ = -0.06 $\\pm$ 0.19 (peak 2) and $\\alpha$ = -0.44 $\\pm$ 0.19 (bridge) - the uncertainty here being dominated by the integrated CCD photometry used to independently reference the TRIFFID data. Temporally, the main peaks are coincident to $\\le$ 10 $\\mu$s although an accurate phase lag with respect to the radio main peak is compromised by radio timing uncertainties. The plateau on the Crab's main peak was definitively determined to be $\\leq$ 55 $\\mu$s in extent and may decrease as a function of photon energy. There is no evidence for non-stochastic activity over the light curves or within various phase regions, nor is there evidence of anything akin to the giant pulses noted in the radio. Finally, there is no evidence to support the existence of a reported 60 second modulation suggested to be as a consequence of free precession. ",
        "introduction": "As the youngest known isolated rotation powered neutron star, and having the highest known spin-down flux density, the Crab pulsar is not surprisingly the most efficient source of high energy magnetospheric emission. All theoretical models attempting to elucidate the processes of nonthermal emission must satisfy the empirical constraints provided by the extensive observational datasets available for this object from $\\gamma$-rays to the infrared.  Contemporary theoretical frameworks in existence to explain the nonthermal emission from this and other isolated rotation-powered neutron stars may be broadly divided into two schools. The first places the source of emission close to the polar cap region, e.g. Sturrock (1971), and the second that place the source at a considerable distance above the neutron star surface, in the outer magnetosphere (e.g. Cheng, Ho \\& Ruderman 1986). In recent times, the development of these models has been to some extent restricted to the problem of $\\gamma$-ray emission. These efforts have been predominantly analytical in nature, as in this energy regime primary emission is being sampled, which avoids complications associated with the lower energy sources of emission. In these latter cases, the emission is secondary in nature and a result of interactions that require numerical representation. Thus the polar cap school (e.g. Daugherty \\& Harding 1982) concluded with correct flux estimates but poor light curves, and the outer gap school (Cheng, Ho \\& Ruderman 1986) with reasonable light curves and approximate flux distributions but requiring extraordinary lines-of-sights and magnetospheric physics to justify the emission.  The group at Stanford University lead by R. Romani is generally credited by being the first to rigourously put the existing models to the test numerically. Their conclusions in many ways matched the intuitive premonitions regarding the two model frameworks, with perhaps the `evolved' outer gap model of Cheng, Ho \\& Ruderman (1986) being the more likely candidate (Chiang \\& Romani 1992; Romani \\& Yadagiroglu 1995). Recently Romani (1998) has indicated that it is only by both the development of more advanced numerical simulations that incorporate the full emission physics (from $\\gamma$-rays to the IR) in a truly self-consistent manner and the accurate characterisation of pulsar light curves in the lower energy regimes that one can hope to disentangle the problem of nonthermal pulsar emission.  Optically, spectroscopic, integrated and high speed single-pixel photometry have produced some critically important results in this regard. However, true phase-resolved data acquisition with accurate isolation of the pulsar's emission from that of the nebula requires high time resolved ($\\sim$ 1 $\\mu$s) 2-d photometry to confirm unambiguously earlier conclusions and to determine the true nature of the magnetosphere's emission in the range $\\sim$ 1 - 10 eV.  In a previous paper (Golden et al. 2000), we presented an analysis of such observations of the Crab pulsar in which we isolated and theoretically speculated upon the unpulsed component of the pulsar's optical emission. These observations were made in January 1996 with the TRIFFID 2-d high speed photometer in three colourbands ($UBV$) using the 6m BTA of the Special Astrophysical Observatory in the Russian Caucasus. Here we present a thorough photometric and temporal analysis of the same data. Following a review of the pulsar's relevant empirical and theoretical characteristics, we briefly outline the observational and data reduction methodology - the reader is encouraged to consult Golden et al. (2000) for a more rigorous treatment of this process. We then detail the results of the photometric and temporal analysis, and conclude with a discussion on the implications of these results for current thinking on this pulsar, with wider ramifications for pulsar emission theory in general. ",
        "conclusions": "Definitive optical observations of the Crab and other pulsars have to date been restricted - either by poor temporal resolution both in terms of resolution and with respect to absolute UTC, and also by the limited spatial information which is crucial in distinguishing the pulsar's emission from the background. Using the TRIFFID high speed photometer, we have obtained datasets uncompromised by these previous limitations, and in addition to confirming many of the previous conclusions of Eikenberry et al. (1997), we have determined several new results uniquely due to the technology we have implemented, including the fluxing of the unpulsed component of emission, the absolute arrival times of the three light curves with respect to one another, and the spatial extent of the plateau on the main peak. Our photometric analysis was however limited by the reference data we used to renormalize the flux ratios thus determined. It is clear that further analysis using more photometrically accurate reference data combined with further observations at extended wavebands, such as the $UV$ or $R/I$ would yield fluxes that would provide excellent leverage to the least-squares fits and thus produce more significant power-law exponents. In terms of a more rigorous temporal analysis of the light curves, deeper observations in $UBV$ would place limits on the extent of the main peak's plateau, and possibly indicate if it is energy dependent. Furthermore, a greater sample of photons per pulse period may allow us to probe the possible correlation between the radio giant pulses and high energy emission mechanisms. We note that the incorporation of polarimetric technology into this nascent field of 2-d high speed photometry would yield critical data which would undoubtedly reveal more clues to puzzle over, and thus contribute towards the eventual solution to this three decade old mystery of how exactly pulsars function."
    },
    "0010/astro-ph0010577_arXiv.txt": {
        "abstract": "s{ A new facility to measure neutrino mass differences and mixing angles and neutrino--nucleus cross sections, such as the proposed ORLaND facility at Oak Ridge, would contribute to the experimental determination of vacuum mixing parameters and would provide an experimental foundation for the many neutrino--nucleus weak interaction rates needed in supernova models. This would enable more realistic supernova models and a far greater ability to cull fundamental physics from these models by comparing them with detailed observations. Charged- and neutral-current neutrino interactions on nuclei in the stellar core play a central role in supernova dynamics, nucleosynthesis, and neutrino detection. Measurements of these reactions on select, judiciously chosen targets would provide an invaluable test of the complex theoretical models used to compute the neutrino--nucleus cross sections. } ",
        "introduction": "Core collapse supernovae are among the most energetic explosions in the Universe, releasing $10^{53}$ erg of energy in the form of neutrinos of all flavors at the staggering rate of $10^{57}$ neutrinos per second and $10^{45}$ Watts, disrupting almost entirely stars more massive than 8--10 M$_{\\odot}$ and producing and disseminating into the interstellar medium many of the elements in the periodic table. They are a key link in our chain of origins from the Big Bang to the formation of life on Earth; a nexus of nuclear physics, particle physics, fluid dynamics, radiation transport, and general relativity; and serve as laboratories for physics beyond the Standard Model and for matter at extremes of density, temperature, and neutronization that cannot be produced in terrestrial laboratories.  Current supernova theory centers around the idea that the supernova shock wave---formed when the iron core of a massive star collapses gravitationally and rebounds as the core matter exceeds nuclear densities and becomes incompressible---stalls in the iron core as a result of enervating losses to nuclear dissociation and neutrinos. The failure of this ``prompt'' supernova mechanism sets the stage for a ``delayed'' mechanism, whereby the shock is reenergized by the intense neutrino flux emerging from the neutrinospheres carrying off the binding energy of the proto-neutron star\\cite{w85,bw85}. The heating is mediated primarily by the absorption of electron neutrinos and antineutrinos on the dissociation-liberated nucleons behind the shock. This process depends critically on the neutrino luminosities, spectra, and angular distributions, i.e., on the multigroup (multi-neutrino energy) neutrino transport between the proto-neutron star and the shock. This past decade has also seen the emergence of multidimensional supernova models, which have investigated the role convection, rotation, and magnetic fields may play in the explosion \\cite{hbc92,mwm93,hbhfc94,bhf95,jm96,mcbbgsu98a,mcbbgsu98b,s98,fh99,khowc99}.  Realistic supernova models will require extremely accurate neutrino radiation hydrodynamics, but unless advancements in neutrino transport are matched by equally important advancements in nuclear and weak-interaction physics, the efficacy of the former must be called into question. In particular, we must move forward to the use of ensembles of nuclei in the stellar core rather than a single representative nucleus when computing electron and electron neutrino capture during the critical core collapse phase. The use of a single representative nucleus has been the standard until now in virtually all supernova models. Moreover, in computing the electron and electron neutrino capture rates, and in general all of the neutrino--nucleus cross sections, detailed shell model computations must replace the parameterized approximations that have been used in the past. For example, recent work on electron capture up to mass 65 has shown that these parameterized rates can be orders of magnitude in error\\cite{lmp00}. The electron capture rate is dominated by Gamow--Teller resonance transitions, with the Gamow--Teller strength distributed over many states. Parameterized treatments place the resonance at a single energy, and this energy is often grossly over- or underestimated relative to the Gamow--Teller centroid computed in realistic shell model computations, leading in turn to rates that are often grossly too small or too large, respectively. (If the centroid is underestimated, more electrons can participate in capture. The reverse is true if the centroid is too high.)  Whereas generations of nuclear structure models will afford ever greater realism in the calculation of stellar core properties and the interactions of core nuclei with the neutrinos flowing through the core, nuclear experiments must be designed and carried out that will serve as guide posts for the theoretical predictions that must be made for the countless rates that enter into any realistic supernova or supernova nucleosynthesis model. In particular, we must have neutrino--nucleus cross section measurements that will help gauge neutrino capture and scattering predictions during stellar core collapse and during the p-, r- and $\\nu$-processes after core bounce.  The plan of this paper is as follows. First we give an overview of the state of the art in one- and two-dimensional supernova models, confining our discussion to issues of neutrino transport and convection. For discussions of the role of general relativity, rotation, and magnetic fields in supernova models, the reader may begin with the papers by Bruenn {\\it et al.}\\cite{bdm00}, Liebend\\\"{o}rfer {\\it et al.}\\cite{l00},\\cite{lmtmhb00}, Fryer and Heger\\cite{fh99}, Khokhlov {\\it et al.}\\cite{khowc99}, and MacFadyen and Woosley\\cite{mw99}. Our focus will then turn to the nuclear and neutrino science that is input to these models and important for supernova neutrino detection, with an eye toward measurements that could be made at a stopped-pion facility such as ORLaND. ",
        "conclusions": ""
    },
    "0010/astro-ph0010563_arXiv.txt": {
        "abstract": "{Recently Pen  and Spergel (1997) have shown that  a universe whose energy density is dominated by a frustrated network of non-Abelian TeV-scale cosmic strings could account for a broad class of cosmological observations. In this paper we consider the effects of such a string network on the massive black holes widely believed to inhabit the centers of many galaxies.  As these black holes traverse the universe together with their host galaxies, they would intersect a large number of string segments.  We argue that such segments would become stuck to the black hole, and be stretched by the hole's motion.  Stretching the strings would cause significant deceleration of the black holes.  Although the black holes would probably not be removed from the galaxies completely, they would be  noticeably displaced from the galactic center of mass -- by at least 5kpc.  This displacement seems to be is contradiction to the observational evidence. ",
        "introduction": "Recently, Pen and Spergel (1997) proposed that the energy density of the universe is dominated by the contribution from a frustrated network of non-Abelian cosmic strings. Because the network is frustrated, the string energy  density scales as only the inverse square of the scale factor. This results in a non-decelerating  (though also non-accelerating) universe,  and so an alternative cosmology intermediate between dark matter dominated Open CDM and cosmological-constant dominated $\\Lambda$ CDM. \\cite{S} presents all the basic features of the frustrated string dominated models. This model is also discussed in \\cite{BS}.  The basis of Pen and Spergel's proposal  is that in some non-Abelian field theories, there are cosmic strings which do not intercommute, {\\it i.e.} pass through each other, effectively. Consequently, the string network can become frustrated. It then ceases to effectively relax to its minimum energy configuration, it ``freezes out.\" The vacuum energy density of the strings can then come to dominate the energy density of the universe.  Pen and Spergel refined this scenario with numerical simulations of the evolution of a network of non-Abelian cosmic strings. These simulations showed that for $\\Omega_0 \\sim 0.4-0.6$ and $H_0 \\sim 60-70$ km/s/Mpc their model is consistent with the current observations, including CMB fluctuations, the shape of the galaxy power spectrum the amplitude of the mass power spectrum, and limits on the age of the galaxies. It was also consistent with observations of high redshift supernova and gravitational lensing statistics.  If the universe today is string dominated, the strings must have formed at an energy scale near the electroweak scale ($\\sim 1$ TeV), implying an energy per unit length of approximately $5\\times10^{-5}$g/cm. The characteristic comoving separation between   strings is approximately $10^{-3}$ of the bubble size during the phase transition, corresponding to $L\\simeq0.1$ A.U. today. (An A.U.  is the distance between the earth and the sun.)  This density of strings  is surprising -- there would be many strings inside our Solar System. However, because the mass per unit length of each string is so low, they would be very difficult to detect. For example the gravitational bending angle due to the string is only $(M_W/M_{Pl})^2 \\sim 10^{-32}$ radians, where $M_W$ is the electroweak symmetry breaking scale and $M_{Pl}$ is the Planck scale. Even if the string can catalyze baryon decay, its cross section is so small that the string could pass through the sun, the earth, or even a person, completely undetected. The probability that a string would pass through a specially designed detector in a reasonable time is extremely low.  There is an exception to the innocuousness of the electro-weak scale strings. If a black hole encounters a string at small enough impact parameter that part of the string enters the event horizon, then, as we will show, the black hole will capture the string. If the black hole horizon is large compared to the string cross-section, then the string will remain threaded through the black hole horizon. The exterior ends are unlikely to reconnect for an enormously long time.  Let us assume that such a network of ``frustrated strings\" really exists in our universe. We will show by simple arguments that such a dense string network would severely impact the motion of black holes in the centers of galaxies.  There is convincing evidence that almost all galaxies contain very massive compact objects, probably black holes, at their centers. Analysis of data such as galactic rotation curves, luminosity and mass-to-light ratio indicates the existence of a central extreme mass concentration in many galaxies, including our Milky Way\\cite{Ghez}, M87 \\cite{M87} , and others \\cite{3377} - \\cite{4486B}. The most convincing case for a black hole {\\it per se} is from our own Galaxy\\cite{Ghez}, for which the stellar rotation curves about Sag A$^\\star$ have been measured to approximately $0.01-0.03$pc, and preclude the presence of a stable cluster of stars, white dwarfs, neutron stars, or even stellar-mass black holes. Other than a massive ($2.6\\times 10^6M_\\sun$) black hole the only possible explanations are a super-dense cluster of either very light ($<0.02M_\\sun$) black holes or weakly interacting dark matter. If anything, each of those seems more far-fetched than a single massive black hole.  The range in mass of these black holes is $10^6 - \\mathrm{few}\\times10^9M_\\sun$, implying a range in Schwarzschild radii of $10^9-\\mathrm{few}\\times10^{12}$m. Especially for the more massive black holes, this is larger than the  inter-string separation. Thus  as the black hole moved through the universe it would encounter many strands of cosmic string. These strands would be stretched by the  motion of the black  hole through space. As the strings are stretched, they reduce the kinetic energy of the black hole, slowing it.  Although each string has a minimal effect on the black hole, even integrated over the lifetime of the universe, the cumulative effect of the large number of strings that are being stretched, can be very significant. Typically we find that, despite the dynamical response of the host galaxy, a typical galactic core black  hole would be observably shifted from the center of the galaxy.  As we argued briefly above, the  motions of ordinary stars are not likely to be affected -- the strings pass right through even a neutron star. Stellar mass  black holes will be affected, albeit to a lesser extent than galactic black holes. This is because the cross-sectional area of a black hole, and hence the  number of strings it is expected to capture, is proportional to the square of the black hole mass. The total black hole deceleration due to string drag is therefore proportional to the first power of the black hole mass. ",
        "conclusions": "The possibility that a frustrated network of non-Abelian strings dominates the energy density of the universe is a fascinating one. Although there are models of such strings which agree remarkably well with a broad class of observations, we have shown that with the string and network parameters required to play the desired cosmological role, the  string network would severely affect the relative motion of central galactic black holes with respect to the rest of the galaxy, if the black hole mass is greater than about $10^8M_\\sun$. The black hole would be noticeably displaced from the center of the galaxy even in the last galactic dynamical time. This does not necessarily rule out a cosmologically important frustrated non-Abelian string network.  It could for example be that the superdense objects at the centers of galaxy are not black holes, but some other exotic objects which would interact far more weakly with the string network. One might also be able, by altering the field-theory model which gave rise to the strings, to alter the strings' linear density or network properties and thereby reduce the dynamical effects on the galactic black holes. In any case, these effects must be considered in any ongoing attempts to incorporate cosmic string networks into cosmology."
    },
    "0010/astro-ph0010049_arXiv.txt": {
        "abstract": "High precision VLA flux density measurements for the lensed images of 0218+357 yield a time delay of $10.1^{+1.5}_{-1.6}$ days (95\\% confidence). This is consistent with independent measurements carried out at the same epoch (\\cite{biggs99}), lending confidence in the robustness of the time delay measurement. However, since both measurements make use of the same features in the light curves, it is possible that the effects of unmodelled processes, such as scintillation or microlensing, are biasing both time delay measurements in the same way. Our time delay estimates result in a confidence intervals that are somewhat larger than those of Biggs et al., probably because we adopt a more general model of the source variability, allowing for constant and variable components. When considered in relation to the lens mass model of Biggs et al., our best-fit time delay implies a Hubble constant of $H_o = 71^{+17}_{-23}$~km~s$^{-1}$~Mpc$^{-1}$ for $\\Omega_o = 1$ and $\\lambda_o=0$ (95\\% confidence; filled beam). This confidence interval for $H_o$ does not reflect systematic error, which may be substantial, due to uncertainty in the position of the lens galaxy. We also measure the flux ratio of the {\\it variable} components of 0218+357, a measurement of a small region that should more closely represent the true lens magnification ratio. We find ratios of $3.2^{+0.3}_{-0.4}$ (95\\% confidence; 8~GHz) and $4.3^{+0.5}_{-0.8}$ (15~GHz). Unlike the reported flux ratios on scales of $0.1''$, these ratios are not strongly significantly different. We investigate the significance of apparent differences in the variability  properties of the two images of the background active galactic nucleus.  We conclude that the differences are not significant, and that time series much longer than our 100-day time series will be required to investigate propagation effects in this way. ",
        "introduction": "Gravitational lenses provide unique opportunities to make cosmological measurements and to study distant sources and lensing galaxies. One such measurement is that of cosmological distance, without the assumptions inherent in intermediate distance indicators (\\cite{refsdal64}; \\cite{refsdal66}). The difference in the light travel time of the paths to the images is proportional to the angular diameter distance to the lens (if the mass scale in the lens can be determined independently) or to an ``effective'' distance, a combination of the angular diameter distances between the lens, observer, and source (\\cite{narayan91}). The difference in the light travel time can be found from the time delay between flux variations in each lensed image. A mass model predicts the relationship between time delay and angular diameter distance. Thus, a time delay and a mass model can be used to determine independently the angular diameter distance to the lens. Another prediction of a gravitational lens model is the magnification ratio. This quantity can be used as a model constraint or as a means to investigate source structure on very small scales (see, for example, \\cite{conner92}). Finally, studies of the variability of lensed images, particularly any differences in the variability properties, present a new tool for investigating propagation effects.  Gravitational lens monitoring observations were first carried out for the gravitational lens 0957+561, with time delays reported by Florentin-Nielsen (1984), Vanderriest et al. (1989), Leh\\'ar et al. (1992), Press, Rybicki, and Hewitt (1992a\\&b) and Oscoz et al. 1997, among others. (For a comprehensive review of time delay measurements in 0957+561 see \\cite{haarsma97}.) The results were conflicting and somewhat controversial, and the discrepancies have only recently been resolved (\\cite{kundic97}, \\cite{haarsma99}). Though these measurements, particularly those of Kundi\\'c et al., determine the delay with high precision, uncertainties in the lens model at present limit the accuracy of a distance determination in this system (\\cite{falco91}; Grogin \\& Narayan 1996a\\&b; \\cite{bernstein97}; \\cite{barkana99}; \\cite{bernstein99}). Recently time delays have been reported in other lens systems, including PG~1115+080 (\\cite{schechter97}), PKS~1830-211 (\\cite{lovell98}), HE~1104-1805 (\\cite{wisotzki98}), B1608+656 (\\cite{fassnacht00}), and B1600+434 (\\cite{koopmans00}). For most of these lenses as well, the modeling currently limits the determination of angular diameter distance. In order to refine significantly gravitational lens investigations of cosmography, it will probably be necessary to consider measurements of many lens systems as well as to improve individual lens models. With this goal in mind a number of lens monitoring programs are under way.  The radio source 0218+357 was first proposed as a gravitational lens by Patnaik et al. (1993) and O'Dea et al. (1992). It consists of two images of a background BL~Lac object, including a radio Einstein ring, and an extended jet and lobe emission  that are not multiply imaged (see Figures \\ref{xmap.fig} and \\ref{umap.fig}). The radio ring has an unusually small diameter, only $0.35''$. The small ring, the detection of absorption lines attributed to neutral hydrogen (\\cite{carilli93}) and various molecules (\\cite{wiklind95}), the large rotation measure observed in linear polarization, and the lens optical colors and redshift  (\\cite{keeton98}) suggest that the lensing object is a gas-rich spiral galaxy. The images of the radio core reveal a flat-spectrum component that has been imaged on VLBI scales (\\cite{patnaik95}, \\cite{patnaik99}). The lens redshift, based on optical (\\cite{browne93}) and radio absorption lines, is 0.685; the likely source redshift is 0.96 (\\cite{lawrence96}).  VLA monitoring observations of 0218+357 have been carried out by Biggs et al. (1999), and they report a time delay of $10.5\\pm0.4$~days (95\\% confidence). They also report significantly different flux ratios of $3.57\\pm0.01$ and $3.73\\pm0.01$ (68\\% confidence) at 8~GHz and 15~GHz, respectively. These results are based on data spanning approximately 100~days. The experience of 0957+561 has taught us that parameter fitting on a relatively small dataset may fail to be robust, with different analysis techniques yielding conflicting results.  Delays based on a single feature may be unreliable due to the effects of unmodeled processes, such as scintillation or gravitational microlensing, which may bias the results. Therefore, repeated measurements and different analysis techniques should be brought to bear on the gravitational lens monitoring experiments. The analysis of Biggs et al. included features found in the polarization light curves as well as the flux density light curves, which in effect represent repeated mesurements of the delay. We report in this paper an investigation of independent VLA measurements of the flux density of 0218+357, in which we used data reduction and parameter fitting techniques that differed from those of Biggs et al. We also adopt a more general model of the source variability, including the possibility of constant and variable components with different magnification ratios. A comparison of the results addresses the issue of the robustness of the parameter estimates. We caution, however, that since our monitoring observations occured at the same epoch as those of Biggs et al., unmodeled scintillation or microlensing may be biasing both mesurements. ",
        "conclusions": "VLA flux density measurements for the lensed images of 0218+357 have been carried out with errors of 1\\% or less, even with only five to ten minutes of dwell time on source. This precision is important in time delay measurements because the variability of the background object is often limited. We find a time delay of $10.1^{+1.5}_{-1.6}$ days (95\\% confidence). We find magnification ratios associated with the {\\it variable} part of the light curves of $3.2^{+0.3}_{-0.4}$ (8~GHz), $4.3^{+0.5}_{-0.8}$ (15~GHz), and $3.4^{+0.2}_{-0.4}$ (averaged). We find best-fit values (which are rather poorly determined) of the difference in the constant part of the light curves of $110^{+80}_{-110}$~mJy (8~GHz) and $180^{+100}_{-140}$~mJy (15~GHz). Our results, when compared to those of Biggs et al., indicate that the time delay measurement is very robust in this system. The magnification is relatively poorly determined because it is difficult to separate the variable and constant part of the light curves; however, this quantity is the one properly associated with the true magnification of the lens and is needed for lens modeling. Our values for the magnifications at the two frequencies are different at about the $3\\sigma$ level.  However, given the uncertainties involved in determining confidence intervals, we do not feel this is strong enough evidence to claim frequency-dependent magnification ratios. Monte Carlo simulations of structure function estimation have shown that the structure functions are also poorly determined. Thus, examining any differences in the variability properties of lensed images, which would be interesting in the investigation of propagation effects, would require much longer time series than are available in this system at present. We have demonstrated numerically that this uncertainty in the structure functions does not significantly impact the time delay estimation, as expected from theoretical considerations (Press, Rybicki and Hewitt 1992a,b). For the lens model of Biggs et al., the measured time delay implies a Hubble constant of $H_o = 71^{+17}_{-23}$~km~s$^{-1}$~Mpc$^{-1}$ for $\\Omega_o = 1$ and $\\lambda_o=0$ (95\\% confidence, filled beam). However, this confidence interval does not reflect systematic error, which may be substantial, due to the uncertainty in the position of the lens galaxy.  Gravitational lenses provide a way to determine the value of $H_{o}$ directly, without any dependence on intermediate distance indicators.  This requires more data and modeling than currently exist. Of particular value would be to combine angular diameter distance measurements for many different lensed systems.  As more lenses are monitored for longer periods, the time delays inevitably will become far more accurate and numerous.  This leaves the lens models as the limiting factor in individual angular diameter distance measurements.  Each lens is unique in this respect, but in general models can be improved by extensive observing at various frequencies and resolutions. Such measurements improve the flux density and positional information about the multiply imaged source components and the lensing mass, and provide a greater number of modeling constraints. In the lens system 0218+357, there is a full Einstein ring in addition to the doubly imaged radio core.  This ring is difficult to observe in detail because it is small (about 350 mas in diameter) and very faint compared to the radio cores.  However, an accurate map of the fine structure in the ring will add many more constraints to any existing models of this system and greatly improve their accuracy.  This is probably the best way to improve the measurement of the angular diameter distance to this lens."
    },
    "0010/astro-ph0010425_arXiv.txt": {
        "abstract": "Atmospheres having a significant radiative support are shown to be intrinsically unstable at luminosities above a critical fraction $\\Gamma_\\mr{crit} \\approx 0.5-0.85$ of the Eddington limit, with the exact value depending on the boundary conditions.  Two different types of absolute radiation-hydrodynamic instabilities of acoustic waves are found to take place even in the electron scattering dominated limit.  Both instabilities grow over dynamical time scales and both operate on non radial modes.  One is stationary and arises only after the effects of the boundary conditions are taken into account, while the second is a propagating wave and is insensitive to the boundary conditions. Although a significant wind can be generated by these instabilities even below the classical Eddington luminosity limit, quasi-stable configurations can exist beyond the Eddington limit due to the generally reduced effective opacity.  The study is done using a rigorous numerical linear analysis of a gray plane parallel atmosphere under the Eddington approximation. We also present more simplified analytical explanations. ",
        "introduction": "Luminous objects shining close to the Eddington limit are of special interest since they occur in a wide range of astrophysical contexts. As such, they have been a popular topic for observational and theoretical investigations. Luminosities in the range of the Eddington limit appear in the very luminous stars populating the top of the HR diagram, in novae during the critical phases of the eruption and also in accreting objects of both galactic and extra-galactic scales. These very luminous objects usually display a great diversity of phenomena ranging from erratic variability to spatial structures and winds. Consequently, the theoretical modeling requires the incorporation of hydrodynamics, radiation and sometimes even magnetic fields.  Undoubtedly, much of the interesting behavior stems from the large, dynamically important luminosity of these objects -- luminosities that are close to the Eddington limit.  The classical Eddington limit is the upper limit possible to the luminosity of an object in hydrostatic balance \\citep{Edd}.  It is the luminosity for which the radiative force upwards balances the gravitational force downwards assuming the opacity of the (fully ionized) gas attains its minimal possible value, that is, the Thomson scattering opacity for non relativistic electrons.  For a homogeneous static configuration, it is the upper luminosity limit because above it there is no consistent solution to the hydrostatic equations.  An atmosphere that violates this upper limit will then simply blow itself apart -- an optically very thick wind is then unavoidable and the object will evaporate on a relatively short time scale.  When the opacity is larger than the Thomson scattering one, as is often the case, one can define a modified Eddington limit that is smaller than the `classical' value (e.g., \\citenp{HD84}, \\citenp{Lam}, and \\citenp{App}).  The modified Eddington luminosity was used to explain the upper luminosity limit observed for stars and the variability that these stars exhibit.  The meaning of the modified Eddington luminosity should however be taken cautiously, because as one approaches the modified Eddington limit, the increased scale height and reduced density imply that in any fully ionized atmosphere, the opacity law is reduced to that of Thomson scattering \\citep{HD94}. Namely, as the modified limit is approached, it itself approaches the classical limit such that the modified Eddington limit is probably not an absolute luminosity limit.  On top of the interesting questions on the nature of atmospheres when their luminosity approaches the Eddington luminosity, exists the fundamental question: `Is the Eddington luminosity the maximal luminosity under all conditions?'. Part of the answer was already answered in the negative by Shaviv (1998). Here we elaborate and extend the analysis and demonstrate that the classical Eddington limit is the limit under a very restrictive set of assumptions. Still we would like to emphasize that a significant factor in turning the Eddington limit into a major tool in setting astrophysical limits is the simplicity of the classical expression, namely $\\Le=4 \\pi c G M / \\sigma_\\mr{Thomson}$.  There are several reasons why the study of radiative instabilities is important.  Most, if not all of the objects shining close to the Eddington limit show variability.  Novae for example, not only show very high (occasionally super-Eddington) luminosities over extended periods of time, they also show clumpy ejected material (e.g., the high resolution HST images, \\citenp{Shar}) which could be explained using radiatively driven instabilities.  Another example are the super-Giants populating the top of the HR diagram.  Quite a few stars with ${M}\\sim 100 {M}_{\\sun}$ and ${L}\\sim 10^{6}{L}_{\\sun}$ exist across a wide range of spectral types and exhibit various degrees of variability.  For example, several of the numerous O3 type stars at 30 Doradus are so bright that their luminosity is clearly very close to their Eddington limit (e.g., \\citenp{MH}).  Less blue (and less young) are the Luminous Blue Variables (LBVs) termed so because of their often eruptive behavior in which their luminosity and rate of mass ejection can increase while their effective temperature decreases (see for example the review by \\citenp{HD94}).  When these objects are in their eruptive state, at least some of them clearly surpass the Eddington limit for significant lengths of time.  For example, $\\eta$ Car in its eruption between 1830 and 1860 released roughly $10^{49}~ergs$ of light over a period much longer than its dynamical time scale (e.g., \\citenp{vGT}), yielding an average luminosity that is more than an order of magnitude higher than the Eddington limit (provided that the mass of $\\eta$ Car is of the order of $100 M_{\\odot}$).  The variability of these objects which is probably a consequence of the high luminosity, is therefore interesting to understand.  An additional type of interesting objects are `very' and `super massive objects' (VMO's and SMO's) that might have existed as massive pop III stars in the young universe.  These massive stars are supposed to shine at luminosities very near the Eddington limit with the more massive of them at virtually the Eddington limit itself. Since the metalicity of these objects is extremely small, the analysis carried out here, which assumes that the opacity is dominated by Thomson scattering, is particularly relevant to these objects.  Super-critical accretion disks are an altogether different type of relevant luminous objects.  Irrespective on whether they advect most of the excess accreted mass and energy above the Eddington rate into the black hole or whether the excess mass is blown away, each element of these disks should shine very close to local effective Eddington limit as these disks are always radiation pressure dominated.  Various works have shown that atmospheres shining close to the Eddington limit exhibit a wide range of instabilities as well as other phenomena such as the generation of strong winds. For example, a large radiation pressure contribution reduces the adiabatic index. The relatively `loose' structure becomes unstable against convection when the adiabatic index decreases due to ionization (e.g., \\citenp{Led}, \\citenp{SC}, and \\citenp{Wag}).  \\citet{JSO} analyzed the appearance of convection at high stellar luminosities assuming that the condition for convective instability is the classical {\\sch} condition, namely the temperature gradient must be higher than the adiabatic one.  The main concern of Joss et al.~was the energy flux and its relation to the Eddington luminosity. Here we are predominately interested in the {\\em conditions} for the stability under high radiative fluxes.  The main result of Joss et al.~was that as the energy flux through an atmosphere is increased, convection arises before the Eddington limit is reached.  Moreover, the convective flux will always grow enough such that the remaining radiative flux is less than Eddington, {\\em as long as convection is efficient}.  Namely, the interior of stars which are dense, will never witness super Eddington fluxes.  It is only near the surface where super Eddington fluxes can be reached.  Since the optical depth below which convection becomes inefficient is relatively high, the wind of a super Eddington object should be very thick and the mass loss should be enormous if nothing happens above the convective zone (cf \\citenp{Sha00}).  When the radiative force term becomes as important as the gravity or the gas pressure gradient terms, it can be a source of mechanical types of instabilities.  For example, \\citet{GK} have shown that strange modes (which are `opacity' modified acoustic waves) suffer an instability in which mechanical work by the radiation is pumped into the waves.  This instability arises from the interaction of the acoustic waves with the non local nature of the diffusive radiation equations.  Another instability that requires special opacity laws is the $\\kappa$-mechanism.  Unlike the $s$-mode instability, however, it grows on time scales much slower than the dynamical time scale, it does not involve work done by the radiation, and exists only when the system is neither fully adiabatic nor fully in the opposite (NAR) limit. However, it can exist when the system is far from Eddington, because it involves the transfer of heat and not work by the radiation.  A related instability arises when the radiation field transfers energy into acoustic waves due to high radiation pressures, as was shown by Hearn (1972, 1973\\nocite{Hea72,Hea73}) and elaborated by \\citet{BPR}, \\citet{Sp}, \\citet{Car} and \\citet{Asp98}.  The instability arises when the opacity increases under compression and the change in the opacity is synchronized with the radiative force. This instability is not a real instability (aka `an absolute instability') in the sense that an infinitesimal perturbation will not grow to become nonlinear, however, as it is a secular drift instability (aka `a convective instability'), it can take finite amplitude waves (excited by convection for example) and increase their amplitude by many $e$-folds during one crossing time of the system.  (A counter propagating wave will be dissipated as fast, thus adding up to no net amplification of a standing wave unless the conditions for the strange mode instability are satisfied, e.g., \\citenp{G94}, \\citenp{Pap97}).  An additional term arises when the Doppler variations of the radiative absorption (arising from wave motion) are taken into account \\citep{Sha99}.  A third term is the driving through line absorption.  This term which utilizes the Doppler effect as well, is often very important in wind acceleration from hot stars (Owocki \\& Rybicki 1984, 1985, 1986, 1991, \\citenp{OR4}, \\citenp{FPP}).  The modes in systems with this type of simple drift instabilities cannot be used however to compose unstable standing waves in a simple Thomson scattering dominated atmosphere.  The problem of the behavior of luminous atmospheres has also been compared to fluidized beds \\citep{PS}.  Intuitively, the kind of bubbling observed in fluidized beds should also manifest itself in radiatively supported atmospheres.  Additional effects arise with the introduction of strong magnetic fields.  Specifically, when strong magnetic fields are present such that hydrodynamic motion perpendicular to the magnetic field lines is frozen, a linear analysis shows that `photon bubbles' arise at a high luminosity, facilitating the transfer of energy \\citep{Aro}.  The original context of accretion onto magnetized neutron stars was also used to describe accretion disks as well \\citep{Gam}.  Magnetic fields are neglected in the present work.  As we shall see, a plethora of instabilities can already manifest themselves when the magnetic fields are dynamically not important.  In a recent work by \\citet{ST99}, it was shown using a global non-radial linear mode analysis, that atmospheres can have unstable standing modes (as opposed to the secular drift instability, or convective instability) even in the case of pure Thomson scattering (as opposed to strange mode instability).  These unstable modes were shown to be horizontally propagating waves that amplify over a time scale much longer than the dynamical times scale of the system.  We have tried to recover this finding but could not succeed.  We report in the Appendix how such an instability could artificially arise. Moreover, the two other soon be described instabilities are dynamically much more important.  In addition to the intrinsic variability and the generation of winds, perhaps the most interesting consequence of the instabilities is the change induced in the Eddington limit. \\citet{Sha98} has shown that in the presence of inhomogeneities, the spatially averaged flux through an atmosphere can increase without increasing the time averaged force on it. This arises because of the tendency of radiation to find the paths of least resistance when escaping through an atmosphere. Under some opacity laws, this will translate into having more radiation in regions where there is less mass and a smaller average force will be exerted. Thus, the luminosity can exceed the Eddington luminosity while the average force on a parcel of mass can remain less than the critical one.  In summary, the classical Eddington limit is restricted to complete homogeneous systems and is frequently not valid in inhomogeneous ones.  A fundamental question of how such inhomogeneities are formed and what form they accept is discussed only briefly in Shaviv (1998). Here we expand on this question with emphasis on Thomson scattering dominated atmospheres (`Thomson atmospheres' hereafter).  Evidently, a picture which relates all the different instabilities together and gives sound physical reasoning to their origin is mostly absent even though it is of paramount importance to the understanding of very luminous objects. It is therefore, our goal in this paper to address the following points:   \\begin{enumerate} \\rightskip 0pt \\item We wish to unravel all the absolute instabilities that Thomson atmospheres (which are the simplest of all) exhibit as they approach the Eddington limit.  This is done using a rigorous linear analysis (that is similar to but not the same as that of \\citenp{ST99}) which assumes nothing save a gray slab atmosphere under the Eddington approximation.  \\item We wish to understand the physical basis behind the instabilities found and how systems with these instabilities behave.  \\end{enumerate}   Two instabilities will be found and described. Additional instabilities arise  when the effects of a varying opacity are taken into account.   As the Eddington limit is approached, these instabilities generally become less important as the opacity becomes mostly that of scattering. However, a future `tabulation' of all the instabilities of luminous atmospheres should include them as well.  A few clear extensions, such as a complete nonlinear analysis using hydrodynamic simulations or a systematic study of the parameter space, will be analyzed in forthcoming publications. These will also include a more elaborate global analysis of the non-trivial opacity law triggered instabilities.  The paper is organized in the following way: The governing radiative hydrodynamic equations are presented in \\S\\ref{sec:equations}.  To clarify the physical picture, we discuss in \\S\\ref{sec:dimen} the parameter space of the problem and show where and when the various limits appear.  We then proceed in \\S\\ref{sec:fulllinear} to analyze the problem in a full global analysis assuming only a gray plane parallel atmosphere, calculate the eigenmodes, and find the characteristics of the instabilities found.  We continue in \\S\\ref{sec:TypeI} to describe the first instability and in \\S\\ref{sec:TypeII} the second instability.  We end with a discussion and summary of conclusions.  An appendix is given to relate the results found here to those found by \\cite{ST99}. ",
        "conclusions": "\\label{sec:discussion} \\label{sec:nonlinear}  We have seen that atmospheres with a significant radiative support always become unstable if the radiative pressure is increased sufficiently and becomes dominant. Both instabilities found are intrinsically non radial. Table \\ref{table:instabilities} summarizes some of the characteristics of the two instabilities found and emphasizes the differences between them and other known instabilities -- convection and $s$-modes which are both dynamic instabilities, and the $\\kappa$-mechanism which is a Carnot type instability.   \\begin{table*}[tbh]  \\caption{Comparison between the instabilities found here and three others.}  {\\small \\begin{center} \\begin{tabular}{| l | c c c c c |} \\hline Instability:                                & Type I & Type II & Convection & $s$-mode &  $\\kappa$-mechanism \\\\ \\hline Radial (exists for $k_x\\rightarrow 0$)      &   -    &   -     &    -       &     +    &        +           \\\\ Propagating ($\\Re(\\omega)\\neq 0$):          &   -    &   +     &    -       &     +    &        +           \\\\ Dynamic time scales ($\\Im(\\omega)\\sim{\\cal O}(1/\\tau_\\mr{dyn})$) &   +    &   +     &    +       &     +    &        -           \\\\ Exists in adiabatic limit ($\\tau_\\mr{cool} / \\tau_\\mr{dyn} \\rightarrow \\infty $) &   +    &   +     &    +       &     -    &    ${^a}$  \\\\ Exists in NAR limit  ($\\tau_\\mr{cool} / \\tau_\\mr{dyn} \\rightarrow 0 $) &   +    &   +     &    -       &     +    &    ${^a}$   \\\\ Exists in diffusion limit  ($\\tau_\\mr{diff} / \\tau_\\mr{dyn} \\rightarrow 0 $) &   +    &   +     &    +       &     +    &    ${^a}$   \\\\ Exists in one fluid limit  ($\\tau_\\mr{diff} / \\tau_\\mr{dyn} \\rightarrow \\infty $) &   -    &   -     &    +       &     -    &    ${^a}$    \\\\ Requires large $p_\\mr{rad}/p_\\mr{gas}$      &   +    &   +     &    -       &     +    &        -          \\\\ Requires special opacity laws               &   -    &   -     &    -       &     +    &        +           \\\\ Significantly affected by boundaries        &   +    &   -     &    -       &     -    &        -           \\\\ \\hline \\end{tabular} \\end{center}  ${^a}$ The $\\kappa$-mechanism per se requires that the gas will {\\em not} be fully in the adiabatic limit. Other similar Carnot type instabilities can exist when the gas is {\\em not} fully in the other three limits. Namely, Carnot type instabilities can potentially exist if the gas is not in one of the four corners of fig.~1.} \\label{table:instabilities} \\end{table*}  Using a Toy model, we can understand the Type I instability. The anti-correlation between gas density and total pressure, dominated by the radiation, reduces the effective speed of sound. If the reduction is large enough, the speed of sound squared becomes negative and we have a purely imaginary frequency. This explains why $\\Im(\\omega)$ was found to be proportional to $k_x$ at long wavelengths.  We have also seen with the Toy model that in addition to the radiative pressure term, changes to the temperature can contribute towards an instability. In Thomson atmospheres, this term is never large enough. However, when opacity variations can be very large, as is the case with dust formation, the temperature perturbation can by itself make the speed of sound become imaginary. In such a case, the lower limit for an instability is the lowest flux for which oscillations on the dynamical time scale are still not adiabatic.  The `temperature' term contribution to the effective speed of sound can become important in a second type of cases in which the equation of state becomes significantly `softer' than that of an ideal gas. That is, when $\\partial \\ln p /\\partial \\ln T |_\\rho \\gg \\partial \\ln p /\\partial \\ln \\rho |_T$. For example, in a system in which the fluid is composed of highly coupled radiation and plasma and the role of the radiation is played by a strong $\\nu$ flux, then given temperature variations will now be more important in setting the effective speed of sound. When this happens, we can again expect to see an instability even in systems which are far from the Eddington flux (as long as oscillations on dynamical time scales are not adiabatic).  The second type of instabilities found, the Type II, were found to be similar to $s$-modes.  The difference however is that Type II does not require special opacity laws on one hand, but on the other hand, it does require non radial oscillations.  The latter allows for a differential relation between the radiation pressure perturbations and the gas density which can result with a non self-adjoint dispersion equation for $\\omega^2$.  Once triggered, the unstable modes describe highly distorted horizontally propagating waves as is depicted in Fig.~\\ref{fig:bubbles}.  Interestingly, if a standing wave in the horizontal direction is composed out of two oppositely propagating waves, the result appears as `bubbles' propagating {\\em downwards}!  That is, it appears as `anti-bubbles'.  The common conception that a `bubbly' phenomenon should arise as the Eddington limit is approached assumes that the `vacated' bubble through which it is easier for the radiation to escape can be held together by something (\\citenp{Sp77}).  This is true in the case of `photon bubbles' in highly magnetized media where the material is forced to move only radially by vertical magnetic fields (\\citenp{Aro}).  When no magnetic field is present, nothing is found to hold the material (\\citenp{Sp77}), and as a result, these bubbles probably do not exit at all.  What we find instead is a very dynamic picture in which the structure or `phase speed' is actually downwards, such that the material in the `vacated' regions is constantly changing on dynamical time scales. This is the opposite of bubbles. Of course, this pattern movement downwards does not involve net motion of material, and it facilitates the transfer of radiation upwards.  What the exact form that this `anti-bubbles' will look like, when in the nonlinear regime, is of course an interesting open question that requires a full numerical simulation.  A very important question which should be addressed in future work is the effects that the assumption of trapping of modes has. That is, it was assumed in the form of the boundary conditions, that acoustic waves are reflected at the boundaries above and below the layer. This need not necessarily be the case. If the atmosphere does not have a corona for example, where the temperature and speed of sound increase with height, any sound wave traveling through the atmosphere will just continue upwards and dissipate itself after forming shocks (which could in principle generate the required corona).  In any case, we expect both instabilities to be important because they amplify on a short dynamical time scale -- a mode does not need to travel back and forth to be amplified significantly. Moreover, the imaginary effective speed of sound in the Type I instability will be shown in a subsequent publication to be a signature of a phase transition in which the homogeneous state of the atmosphere is no longer the preferred or lowest energy state, that is, the instability found is just a manifestation of the system trying to get into a new equilibrium condition which is not the homogeneous one.  Although a lengthy description of the nonlinear behavior is part of a forthcoming publication, a few words are in order about how such systems are expected to behave once these instabilities start to set in.  The results of \\S\\ref{sec:modes_gamma} have shown that either the Type I or the Type II instabilities will necessarily set in before the Eddington limit is reached.  Thus, as a luminous object approaches the Eddington limit, convection will become important in the inner parts of the object where the density is high enough and the cooling time scale long (\\citenp{JSO}), while the outer parts will be unstable to the Type I or II instabilities where the two fluid limit is relevant, even if the opacity is Thomson dominated.  Once either one of the instabilities takes place, it is clear that the atmosphere will not remain homogeneous.  If the nonlinear pattern that is eventually formed propagates, as is the case in the Type II instability, each gas element will experience periodic density variations.  While the peak force experienced by each mass element may exceed the Eddington limit, the time average force may be significantly below it \\citep{Sha98}. In this case, an atmosphere can transfer through it a luminosity greater than the Eddington luminosity without blowing itself apart. This will occur in the region of the atmosphere where perturbations of order the scale height become optically thin.  From this region upwards the effective opacity is again the microscopic one and the luminosity becomes effectively super Eddington.  A predictable wind will then be blown from that region upwards (\\citenp{Novae}).  In the case of the Type I instability, on the other hand, the pattern formed is stationary and it resembles that of `chimneys'.  The gas elements present in the lower density regions can witness a constant (in time) super-Eddington flux.  Although the average over the entire atmosphere can be a sub-Eddington luminosity, some gas elements in this case can be accelerated upwards through the chimneys, driving turbulence in the shear layers along the sides of the chimneys. However, this material will probably not be able to escape to infinity because it is unlikely that the escape speed will be attainable in the chimneys.  This will require strong shears that presumable dissipation by shocks will prevent it from occuring. Thus, once the Eddington flux is surpassed in the chimneys, it will likely generate `fountains' in which material accelerates upwards until the top of the chimneys is reached, then the material will fall back.  If the average luminosity is super Eddington, the material need not stagnate and it can be easily driven off to infinity as a wind.  In both cases, the clear role that the classical Eddington limit plays in the homogeneous case is lost.  On one hand, a strong wind can exist already below the Eddington limit, driven for example b the energy dissipated from the unstable modes; while on the other hand, quasi-stationary configurations (with a wind) can exist above the Eddington limit.  A more detailed treatment of the nonlinear behavior, which includes toy models as well as numeral simulations is underway in a forthcoming publication.       The two main results found in this paper are the existence of two types of dynamically important instabilities that can take place in Thomson atmospheres.  The two instabilities exist in addition to the possibility of convection.  Following is a summary of the main results pertaining to these two important instabilities which will govern the behavior of atmospheres shining close to the Eddington limit.  \\begin{enumerate} \\rightskip 0pt \\item Under a linear analysis of a slab Thomson atmosphere, two dynamically important instabilities were found to exist for the first time.  \\item Both instabilities operate in the two fluid limit of optically thick atmospheres -- when the diffusive time scale is shorter than the dynamical time scale of the system. Hence, they operate from the photosphere down to where convection can become efficient.  \\item The Type I instability is discovered once a nonlocal analysis of the radiation field behavior is carried out. A fixed flux boundary condition quenches it off.  \\item The Type I instability does not propagate. It corresponds to the opening of `chimneys'. It appears in modes that have the radiation density on average anti-correlated with the gas density. The anti-correlation reduces the square of the effective speed of sound. When $c_{\\mr{eff}}^{2}$ is negative, the instability appears and grows over a dynamical time scale.  \\item The Type I instability can take place also for $k_x \\rightarrow 0$, however, the growth rate then tends to vanish: $\\eta \\propto k_x \\rightarrow 0$.  \\item A second instability, the Type II, appears under more general boundary conditions, but at somewhat higher $\\Gamma$'s than the Type I.  \\item The mathematical origin of the Type II instability is similar to that of $s$-modes. Since the hydrodynamic equations are coupled to a diffusion equation for the radiation, the relation between the pressure and density is differential. In the Type II instability, it arises from the term responsible for generating perpendicular radiative fluxes.  \\item The Type II instability describes horizontally propagating waves that amplify over dynamical time scale.  If a standing mode is composed out of two oppositely propagating modes, it has the appearance of `bubbles' moving downwards. This arises because the phase velocity of the wave standing in both the horizontal and vertical direction is pointing down.  This is opposite to the intuitive `photon bubble' picture, in which the structure appears to be moving upwards.   \\item As the flux through an atmosphere increases towards the Eddington limit, Thomson scattering dominated atmospheres become unstable above a critical fraction of the Eddington flux. This instability is independent of and comes on top of the possibility of convective instability. The first mode to become unstable can be either Type I or Type II unstable. Type I appears at $\\Gamma\\gtrsim 0.5$. However, the exact value depends on the boundary conditions and thermal gradient. The Type II instability appears under more general boundary conditions but at $\\Gamma \\approx 0.85$.  \\item The most unstable horizontal wavelength for the type I instability is the longest wavelength in the system at $\\Gamma_\\mr{crit}$, but it becomes smaller as $\\Gamma$ is increased. The most unstable mode for the type II instability is about $0.9 \\left<l_p^{-1}\\right>$. It doesn't change by much as $\\Gamma$ is increased from $\\Gamma_\\mr{crit}$.  \\item The analysis assumed thus far that the modes can be trapped around the layer under consideration. However, since the two instabilities operate over a dynamical time scale, they are probably important even if the acoustic oscillations are not trapped, as is the case with convection for example.  \\item The unstable modes always tend to reduce the effective opacity of the medium (if the luminosity is sufficiently close to the Eddington luminosity), thereby increasing the effective Eddington luminosity. The radiation and the opacity must therefore be solved simultaneously and self consistently. Treatment of the inhomogeneities on a large scale (say scales larger than a few optical lengths) requires a new formulation of the equation of state because of the process of averaging over the inhomogeneities. This problem is deferred to a later paper.  \\end{enumerate}  All the results presented in this work assume that the mass opacity is constant. In general however, the opacity can be a function of the gas variables. This can be shown to not only change quantitatively the results but also to introduce the effects of additional instabilities, some of which have not yet been described. Moreover, it should be stressed that since the temperature of the matter and the radiation at any given point are not necessarily identical, non equilibrium effects cannot be neglected. These effects should be discussed in the future.  Finally, if we are also to fully understand the behavior of the instabilities, we should analyze their nonlinear behavior. It is this non linear behavior that will set many of the observational properties of systems shining close to their Eddington limit."
    },
    "0010/astro-ph0010613_arXiv.txt": {
        "abstract": "I discuss the energetics of synchrotron-emitting outflows, increasingly found to be present in many different classes of X-ray binary systems. It is shown that the outflow is likely to be comparable in power to the integrated X-ray luminosity, traditionally taken to be an indicator of the global mass-transfer rate. This is especially found to be the case in the (low/)hard states of black hole candidate systems. I conclude that jets are extremely important, energetically and dynamically, for the accretion process in the majority of known X-ray binary systems. ",
        "introduction": "Radio emission as a property of X-ray binary systems (XRBs) has developed in recent years from being a rare and unusual facet of their broadband spectrum to being recognised as a ubiquitous property of several classes of XRBs.  In the early 1990s, it was still (almost) possible to discuss individually radio-emitting X-ray binaries (e.g. Hjellming \\& Han 1995), and numbers of systems in which collimated outflows, or jets, were directly observed could be counted on the fingers of one hand.  As the 1990s drew on, more and more of these systems became resolved into such jets, and in the middle of the decade the apparent superluminal motions observed from GRS 1915+105 (Mirabel \\& Rodriguez 1994; see also Fender et al. 1999) and GRO J1655-40 (Tingay et al. 1995; Hjellming \\& Rupen 1995) established beyond doubt that XRBs could accelerate powerful flows to bulk velocities in excess of $0.9c$. By 2000 we are beginning to realise that jets may be rather more ubiquitous than previously imagined (e.g. Mirabel \\& Rodriguez 1999; Fender \\& Hendry 2000; Fender 2000), and in Figs 1(a),(b) I summarise our (or at least my) current understanding of the empirical relation of radio emission to X-ray `state' of the neutron star (NS) and black hole candidate (BHC) systems respectively. In addition to the relations indicated in these figures, it seems that discrete, radio-emitting ejection events (in which the radio spectrum rapidly becomes optically thin) are associated with state transitions (including outbursts of transients).  \\begin{figure} \\centering\\epsfig{figure=fender_fig1a.ps,width=12cm,height=10cm,angle=0} \\caption{(a): A qualititative sketch of the relation of jets to accretion in the three `types' of neutron star XRB. In the low-field `Atoll' sources the accretion rate is believed to be $<10$\\% Eddington, except possibly during rare transient outbursts (e.g. Aql X-1); the evidence is marginal so far but it appears such sources are `radio on' when in the `Island State' (IS) in the X-ray colour-colour diagram (CD). The Z sources are believed to be accreting at a much higher rate, near Eddington, and are `radio on' when on the `Horizontal Branch' and maybe also, at a lower level, the `Normal Branch' in the CD. Note that for both Atoll and Z sources the estimates of surface magnetic field are very uncertain. Finally, in the high-field X-ray pulsars no radio synchrotron emission has ever been detected; possibly this is due to truncation of the accretion disc a long way from the neutron star.} \\end{figure}  \\begin{figure} \\setcounter{figure}{0} \\centering\\epsfig{figure=fender_fig1b.ps,width=10cm,height=12cm,angle=270} \\caption{(b): The qualitative relation of radio emission to X-ray `state' in BHC XRBs: the low/hard state is found to produce a steady, flat- (or inverted-)spectrum jet, the soft state produces no detectable radio emission. The relation of $\\dot{m}$ to these states is not certain (e.g. Homan et al. 2001); nor are the radio characteristics of the relatively rare `Intermediate/Very High' states well determined. } \\end{figure}  While some areas of the coupling between accretion flow and jet are still empirically uncertain (the Atoll sources, the Intermediate/Very High state for BHCs), it seems that all XRBs except the high-field X-ray pulsars will, under the right conditions, produce a synchrotron-emitting jet. It is therefore an important question to address the significance of this jet, energetically and dynamically, for the process of accretion onto compact objects as a whole. One thing obvious from inspection of Figs 1(a),(b) is the apparent anti-correlation between the mass accretion rate, $\\dot{m}$, traditionally estimated from X-ray studies alone, and the presence of a jet, in both NS and BHC systems (see also Belloni, Migliari \\& Fender 2000 and Homan et al. 2001 for further discussion). ",
        "conclusions": "Two different (but not entirely independent) approaches have revealed that a large fraction of the entire luminosity (radiative and mechanical) of some {\\em classes} (this is important -- we are no longer talking about individual `oddballs') of X-ray binaries may be carried by the jet.  This fraction is likely to be $> 0.1$ and may be $>0.5$, ie. the jet may even be the {\\em dominant} power output channel. Since it is clear from studies of e.g. GRS 1915+105 that the disc--jet coupling is occurring in the inner few 100 km of the accretion flow, the extraction of such a large amount of energy (and presumably angular momentum) must have a dramatic effect on the accretion flow close to the compact object.  Based on this, it seems that models for accretion in X-ray binaries which at least parameterise such an outflow are needed. They are not entirely non-existent -- for GRS 1915+105 (again!) both Nayakshin, Rappaport \\& Melia (2000) and Janiuk, Czerny \\& Siemiginowska (2000) have begun to consider the effect of a moderately powerful outflow on the accretion process. A more radical step has been taken by Markoff, Falcke \\& Fender (2000; see also these proceedings) in which almost the entire broadband spectrum of the BHC low state (excepting thermal emission in the optical -- UV range) is explained by jet emission, and in which it is suggested that the X-ray power-law may even be {\\em synchrotron} emission. This would be contrary to the generally accepted view that the X-ray power law in the low state arises in Comptonisation of `seed' photons in a hot, thermal, corona. It is further worth noting that once mass-loss via jets is considered, and the empirical connections between X-ray `state' and jet formation, as outlined in Figs 1(a),(b) recognised, then it is no longer clear that the global mass transfer rate can be inferred from X-ray observations alone."
    },
    "0010/astro-ph0010496_arXiv.txt": {
        "abstract": "We present an observational and theoretical study of the optical emission from nonradiative shocks in three supernova remnants: the Cygnus Loop, RCW 86 and Tycho.  The spectra of these shocks are dominated by collisionally excited hydrogen Balmer lines which have both a broad component caused by proton-neutral charge exchange and a narrow component caused by excitation of cold neutrals entering the shock.  In each remnant we have obtained the broad to narrow flux ratios of the H$\\alpha$ and H$\\beta$ lines and measured the H$\\alpha$ broad component width.  A new numerical shock code computes the broad and narrow Balmer line emission from nonradiative shocks in partially neutral gas. The Balmer line fluxes are sensitive to Lyman line trapping and the degree of electron-proton temperature equilibration.  The code calculates the density, temperature and size of the postshock ionization layer and uses a Monte Carlo simulation to compute narrow Balmer line enhancement from Lyman line trapping.  The initial fraction of the shock energy allocated to the electrons and protons (the equilibration) is a free parameter. Our models show that variations in electron-proton temperature equilibration and Lyman line trapping can reproduce the observed range of broad to narrow ratios. The results give 80\\%$-$100\\% equilibration in nonradiative portions of the NE Cygnus Loop ($v_{S}\\,\\sim\\,$300 \\kms), 40\\%$-$50\\% equilibration in nonradiative portions of RCW 86 ($v_{S}\\,\\sim\\,$600 \\kms) and $\\lesssim\\,$20\\% equilibration in Tycho ($v_{S}\\,\\sim\\,$2000 \\kms).  Our results suggest an inverse correlation between magnetosonic Mach number and equilibration in the observed remnants. ",
        "introduction": "Supernova explosions produce some of the strongest shocks in nature.  During the event, the outer layers of a star are ejected at speeds as high as 30,000 \\kms. The dense ejecta behave like a highly supersonic piston, producing a strong shock wave in the ISM ahead of the piston (commonly known as the forward shock). When the mass swept up by the forward shock begins to exceed the ejecta mass, the supernova remnant (SNR) enters the blast wave (Sedov-Taylor) phase of evolution (Hamilton \\& Sarazin 1984, Truelove \\& McKee 1999). From the genesis of the SNR through the blast wave stage the forward shock is nonradiative, meaning that it loses a negligible fraction of its internal energy to radiative cooling.  Due to the low density (n\\,$\\lesssim\\,$1 cm$^{-3}$) and high Mach number ($M\\,\\gtrsim\\,$200) of the forward shock, the heated interstellar gas behaves like a collisionless plasma (Draine \\& McKee 1993). According to the strong shock jump conditions, electrons and protons are heated to temperatures in a minimum ratio $T_{e}/T_{p}\\,=\\,m_{e}/m_{p}$\\,($\\sim\\,$1/2000).  However, plasma waves and MHD turbulence at the shock front may transfer energy from protons to electrons (Tidman \\& Krall 1971, Kennel 1985, Cargill \\& Papadapoulos 1988), further equilibrating the two particle temperatures. Since the electrons and protons are affected by a different range of plasma waves and the Coulomb equilibration time downstream often exceeds the age of the remnant, the two particle temperatures can remain unequal throughout the shock. The types of plasma waves excited at the shock transition (and hence the amount of collisionless heating) depend strongly on parameters such as the magnetosonic Mach number $M_{S}$ and magnetic field orientation.  Due to this complicated dependence and the extreme difficulty of creating high Mach number collisionless shocks in the laboratory, the properties of collisionless shocks remain poorly understood.  In a cold ($\\lesssim\\,$10$^{4}$ K), partially ionized medium, the charged and neutral distributions respond very differently to the passage of a collisionless shock.  The cold neutrals are initially unaffected by the shock transition, while charged particles are compressed by a factor of four and strongly heated.  Some of the cold neutrals entering the shock are collisionally excited before being destroyed by collisional ionization or charge transfer.  Radiative decay by these excited neutrals produces narrow component Balmer emission with a line width given by the preshock temperature.  In contrast, charge exchange between cold neutrals and protons produces fast neutrals having the velocity distribution of the postshock protons.  Collisional excitation of the fast neutrals produces broad Balmer line emission (Chevalier \\& Raymond 1978, Chevalier, Kirshner \\& Raymond 1980 (hereafter CKR80)). While the low temperatures in radiative shocks favor the emission of strong forbidden lines such as [O~II], [O~III], [N~II] and [S~II], the high temperatures behind nonradiative shocks produce hydrogen Balmer line emission far more efficiently.   Balmer-dominated shocks have been detected in Tycho (Kirshner, Winkler \\& Chevalier 1987 (hereafter KWC87), Smith \\etal\\, 1991, Ghavamian \\etal\\, 2000), SN 1006 (KWC87, Smith \\etal\\, 1991, Winkler \\& Long 1997), RCW 86 (Long \\& Blair 1990, Smith 1997), Kepler (Blair, Long \\& Vancura 1991), portions of the Cygnus Loop (Raymond \\etal\\, 1983 (hereafter RBFG83), Fesen \\& Itoh 1985, Hester, Raymond \\& Blair 1994 (hereafter HRB94)), four remnants in the LMC (Tuohy \\etal\\, 1982, Smith \\etal\\, 1991, 1994) and the bow shock surrounding the pulsar PSR 1957+20 (Aldcroft, Romani \\& Cordes 1992).  The optically emitting layer behind a nonradiative shock is extremely thin ($\\lesssim\\,$10$^{-3}$ pc), so the proton temperature very close to the shock front can be measured from the width of a broad Balmer line. The proton temperature in turn depends on the shock velocity through the jump conditions, making the broad line profile a powerful diagnostic for probing the global kinematics of young SNRs.  The broad to narrow flux ratio is another important diagnostic because it also depends on the shock velocity (CKR80, Smith \\etal\\, 1991).  Proper motion measurements of Balmer-dominated filaments have been combined with shock velocity estimates from the broad component to estimate distances to several young SNRs (for example see Long, Blair \\& van den Bergh 1988 for SN 1006, and Hesser \\& van den Bergh 1981 for Tycho).  Balmer-dominated spectra can be difficult to interpret because the broad component width and broad to narrow ratio yield different shock velocities depending on the assumed electron-proton equilibration.  Since the equilibration is not known a priori, spectroscopic observations in the past have only yielded a range of shock velocities $v_{S}$ from the Balmer line profiles: a minimum $v_{S}(min)$ for no equilibration and a maximum $v_{S}(max)$ for full equilibration.  In addition, collisional excitation behind the shock generates both Lyman line photons and Balmer line photons.  Lyman line trapping by slow neutrals partially converts Ly $\\beta$ and higher Lyman photons into Balmer line photons; this enhances the Balmer line flux in the narrow component and complicates the diagnostic interpretation of the broad to narrow ratio (CKR80, Smith \\etal\\, 1991, Ghavamian 1999)).  Moreover, since the Ly $\\beta$ optical depth is larger than that of Ly $\\gamma$, the H$\\alpha$ broad to narrow ratio is more strongly affected than the H$\\beta$ broad to narrow ratio.  The Lyman line optical depth behind the shock typically lies between 0 and 1, so that neither Case A nor Case B conditions apply.  It is clear that disentangling the combined effects of Lyman line trapping and equilibration on a Balmer-dominated spectrum requires (1) the acquisition of high S/N line profiles in both H$\\alpha$ and H$\\beta$, and (2) careful modeling of the atomic physics and radiative transfer.  In \\S 2 of this paper, we present high S/N spectra of three nonradiative shocks which resolve the broad and narrow components of both H$\\alpha$ and H$\\beta$.  In \\S 3, we present measurements of the H$\\alpha$ broad component width and broad to narrow ratios in H$\\alpha$ and H$\\beta$.  We describe numerical models of nonradiative shocks in \\S 4, including calculations of ionization structures for a range of equilibrations and Monte Carlo simulations of Ly $\\beta$ and Ly $\\gamma$ trapping.  In \\S 5 we compare the observed broad to narrow ratios with model predictions and simultaneously determine both the shock velocity and equilibration for the observed SNRs.  In \\S 6 we discuss our results and in \\S 7 we present our conclusions. ",
        "conclusions": "To investigate the properties of nonradiative shocks, we have obtained high S/N spectra of the Cygnus Loop, RCW 86 and Tycho covering a factor of 10 in Mach number.  In each remnant, we have measured broad to narrow ratios of both H$\\alpha$ and H$\\beta$.   We find that the broad to narrow ratios show considerable variation from one remnant to the next, with the H$\\beta$ ratio systematically larger than that of H$\\alpha$.  The difference between the two ratios is evidence of Lyman line trapping in the narrow component.  We have devised a numerical code which predicts the ionization structure behind a nonradiative shock and uses a Monte Carlo simulation to calculate the influence of Ly $\\beta$ and Ly $\\gamma$ trapping on the H$\\alpha$ and H$\\beta$ emission lines.  We have modeled the H$\\alpha$ and H$\\beta$ broad to narrow ratios in three of the observed shocks: Cygnus P7 ($v_{S}\\,=\\,$300$-$400 \\kms), SW RCW 86 ($v_{S}\\,=\\,$580$-$660 \\kms) and Tycho Knot g ($v_{S}\\,=\\,$1940$-$2300 \\kms).  Overall, the shock code matches the observations in the first two remnants, but yields only marginal agreement for Tycho Knot g.  The models indicate nearly complete equilibration for Cygnus P7 and half equilibration for RCW 86 SW, evidence for substantial collisionless heating in these nonradiative shocks.  In Knot g, the predicted H$\\alpha$ broad to narrow ratios are systematically larger than the observed values for all equilibrations. The difficulty in modeling the Knot g broad to narrow ratio may be due to the large uncertainty in proton excitation cross sections near threshold.   However, the models do successfully reproduce the observed H$\\beta$ broad to narrow ratios. From the H$\\beta$ results, we infer a low equilibration for Tycho, $\\lesssim\\,$20\\%.  This value is consistent with the findings of Laming \\etal\\, (1996), who used ultraviolet observations with HUT to determine the equilibration of shocks of similar strength in SN 1006.  For a given shock velocity, we find that the thickness of the ionization layer depends on the electron-proton temperature equilibration.  At shock velocities $\\lesssim\\,$1000 \\kms, the collisional ionization of hydrogen is dominated by the electrons; the greater the equilibration, the thinner the ionization layer.  This trend is reversed at high shock velocities.  Above 2500 \\kms, proton ionization dominates electron ionization; therefore the more equilibrated the shock, the thicker the ionization layer.  P. G. would like to the thank J. Weisheit, J. M. Laming and R. Bandiera for several helpful discussions. We would also like to thank the anonymous referee for valuable suggestions in improving the presentation of this paper. The work of P. G. was supported by grant D70832 from Rice University, STScI grant GO-07515-02.96A, NSF atomic physics theory grant PHY-9772634 and student travel support from CTIO.  P.G. also acknowledges the hospitality of the Harvard-Smithsonian Center for Astrophysics, where some of this paper was completed.  The work of J. R. was supported by NASA grant NAG 5-2845.  \\clearpage"
    },
    "0010/astro-ph0010343_arXiv.txt": {
        "abstract": "We present an analysis of interstellar and intergalactic absorption lines in the FUSE spectrum of the low-redshift quasar PG\\,0804\\,+761 ($z_{\\rm em}=0.100$) at intermediate resolution (FWHM$\\sim 25$ km\\,s$^{-1}$) in the direction $l=138\\fdg3$, $b=31\\fdg0$. With a good signal-to-noise ratio (S/N) and the presence of several interesting Galactic and extragalactic absorption components along the sight line, this spectrum provides a good opportunity to demonstrate the ability of FUSE to do both interstellar and extragalactic science. Although the spectrum of PG\\,0804\\,+761 is dominated by strong absorption from local Galactic gas at 0 km\\,s$^{-1}$, we concentrate our study on absorption by molecular hydrogen and neutral and ionized metals related to an intermediate-velocity cloud (IVC) in the lower Galactic halo at $-55$ km\\,s$^{-1}$, and on absorption from O\\,{\\sc vi} extended to negative velocities. In the IVC, weak molecular hydrogen absorption is found in 5 lines for rotational levels $0$ and $1$, leading to a total H$_2$ column density of log $N = 14.71 \\pm 0.30$. We derive an O\\,{\\sc i} gas-phase abundance for the IVC of $1.03^{+0.71}_{-0.42}$ solar. Lower abundances of other elements (Fe, Si) imply depletion onto dust grains or the presence of higher, undetected ionization states. The presence of N\\,{\\sc ii} and Fe\\,{\\sc iii} absorption at $-55$ km\\,s$^{-1}$ indicates that a fraction of the hydrogen is ionized. From the relative abundances of O\\,{\\sc i} and P\\,{\\sc ii} we estimate a degree of ionization H$^+$/(H$^0 +$H$^+$) of $\\sim 19$ percent. Absorption by O\\,{\\sc vi} is found at velocities as negative as $-110$ km\\,s$^{-1}$, but no absorption from any species is found at velocities of $\\sim -180$ km\\,s$^{-1}$ where absorption from the nearby high-velocity cloud Complex A would be expected. We suggest that the extended O\\,{\\sc vi} absorption traces hot gas situated above the Perseus spiral arm. Finally, we find intergalactic absorption by an intervening H\\,{\\sc i} Ly\\,$\\beta$ absorber at $z_{\\rm abs}=0.019$ and absorption by H\\,{\\sc i}, C\\,{\\sc iii} and O\\,{\\sc vi} in an associated system at $z_{\\rm abs}=0.102$. No intervening O\\,{\\sc vi} absorbers are seen in the spectrum of PG\\,0804\\,+761. ",
        "introduction": "The succesful launch of the {\\it Far Ultraviolet Spectroscopic Explorer} (FUSE) in June 1999 (Moos et al.\\,2000) has provided a major new facility for obtaining absorption line spectra at intermediate resolution (FWHM$\\sim 25$ km\\,s$^{-1}$) in the far ultraviolet (FUV) regime from 905 to 1187 \\AA. In this wavelength range, many spectral diagnostic lines provide information over a wide range of physical conditions found in the diffuse and translucent interstellar medium. The Lyman and Werner bands of molecular hydrogen (H$_2$), as well as absorption bands of carbon monoxide (CO), give important information about the cold molecular gas. The complete Lyman series of neutral hydrogen and deuterium, in combination with a large number of transitions of neutral and weakly ionized metals, can be used to study the conditions of the neutral and diffuse ionized gas in great detail. Finally, absorption lines from highly ionized species, in particular the O\\,{\\sc vi} doublet near 1030\\,\\AA, are excellent tracers of the hot component of the ISM, as found in halos of galaxies or within expanding shells and superbubbles. Absorption spectroscopy in the FUV range is an extremely powerful technique for the study of the multi-phase structure of the ISM of the Milky Way and gives additional information about the structure and physics of the intergalactic medium (IGM), when applied to quasars or Active Galactic Nuclei (AGNs).  FUV absorption spectroscopy with the {\\it Copernicus} satellite (1972-1981) was performed to intensively study the interstellar medium (for a review see Spitzer \\& Jenkins 1975), but was limited in sensitivity to nearby bright stars as background sources. Later satellites, such as the {\\it International Ultraviolet Explorer} (IUE) and the {\\it Hubble Space Telescope} (HST), did not cover wavelengths below 1150 \\AA. The {\\it Orbiting and Retrievable Far and Extreme Ultraviolet Spectrometer} (ORFEUS), launched in 1996 for a 14-day {\\it Space Shuttle} mission, was able to obtain intermediate resolution FUV spectra toward extragalactic background sources. Observations with ORFEUS led to many new insights concerning the ISM in the Galactic halo and the Magellanic Clouds (Widmann et al.\\,1998; de\\,Boer et al.\\,1998; Richter et al.\\,1998, 1999), but the number of observed objects was limited due to the short mission duration. With FUSE, the number of extragalactic sight-line spectra in the FUV range is growing rapidly and the very first results from FUSE (e.g., Oegerle et al.\\,2000, Savage et al.\\,2000; Sembach et al.\\,2000; Shull et al.\\,2000; Snow et al.\\,2000) already imply that the satellite will make important contributions to studies of the ISM and IGM. With its high sensitivity and its moderate resolution, FUSE is able to obtain spectra toward fainter (and thus more distant) background sources and so gives access to regions in space which could not be studied before in the FUV range.  In this paper we present measurements of absorption lines in the spectrum of the bright low-redshift quasar PG\\,0804\\,+761 ($V=15.20, z_{\\rm em} =0.100$, Schmidt \\& Green 1983) obtained with FUSE in January 2000. The line of sight to PG\\,0804\\,+761 in the Galactic direction $l=138\\fdg3, b=+31\\fdg0$ passes through local Galactic gas and through the Galactic halo above nearby and distant spiral arms. The FUSE spectrum of PG\\,0804\\,+761 thus provides an opportunity to investigate the ISM along a path extending over the outer regions of the Milky Way. The goal of this paper is to analyze the sight-line structure in this interesting direction and to provide a reference spectrum for further studies.  The outline of this paper is as follows: a review of the sight-line structure is given in \\S2. In \\S3 we present the FUSE observations and the data reduction. Absorption by local Galactic gas is briefly discussed in \\S4. Absorption by molecules, neutral and weakly ionized species at intermediate negative velocities is presented in \\S5, while absorption by moderately and highly ionized gas is described in \\S6. In \\S7 we shortly discuss the lack of absorption by high-velocity gas from Complex~A. \\S8 presents the results for the intergalactic gas and the associated system in front of PG\\,0804\\,+761. Finally, a summary of our study is given in \\S9. ",
        "conclusions": "We present FUSE far ultraviolet observations at intermediate resolution of interstellar and intergalactic absorption lines along the line of sight toward the quasar PG\\,0804\\,+761 at $z_{\\rm abs}=0.102$. Along this sight line we find absorption from molecules and weakly ionized species in local Galactic gas at $0$ km\\,s$^{-1}$, and in gas associated with the Low Latitude Intermediate Velocity Arch (LLIV Arch) in the lower Galactic halo at $-55$ km\\,s$^{-1}$. We also detect absorption from highly ionized gas in the Galactic halo extended to velocities as negative as $-110$ km\\,s$^{-1}$, and from intergalactic gas at redshifts of $z=0.019$ and $0.102$, the latter most likely being associated with the quasar itself.  1. We find H$_2$ absorption in gas of the LLIV Arch with a total column density of log $N=14.71 \\pm 0.3$ from the lowest 2 rotational states ($J=0$ and $1$). An excitation temperature of $193^{+322}_{-75}$ K is derived for the H$_2$ gas from a Boltzmann fit to the relative populations of the rotational states $J=0,1$. From the presence of H$_2$ in the LLIV Arch we conclude that dust is present in this cloud in the lower Galactic halo.  2. An analysis of metal abundances in the LLIV Arch is performed, including absorption lines from C\\,{\\sc ii}, N\\,{\\sc i}, N\\,{\\sc ii}, O\\,{\\sc i}, Si\\,{\\sc ii}, P\\,{\\sc ii}, Ar\\,{\\sc i}, Fe\\,{\\sc ii} and Fe\\,{\\sc iii}. For O\\,{\\sc i}, we find a gas phase abundance of $1.03^{+0.71}_{-0.42}$ solar. The presence of N\\,{\\sc ii} and Fe\\,{\\sc iii} indicates that a substantial fraction of the gas is ionized. We estimate a degree of ionization of H$^+/$(H$^0+$H$^+$) $\\approx 0.19$. The presence of dust, molecules and metals at nearly solar abundances lead us to conclude that the LLIV Arch is part of the return flow of a Galactic fountain.  3. We investigate velocity profiles of the moderately and highly ionized species C\\,{\\sc iii} and O\\,{\\sc vi}. While C\\,{\\sc iii} absorption is blended by an O\\,{\\sc i} line, we find O\\,{\\sc vi} absorption at $\\lambda = 1031.926$ \\AA\\, extended to negative velocities as far as $-110$ km\\,s$^{-1}$.  We suggest that the O\\,{\\sc vi} absorption at these negative velocities is related to highly ionized gas situated above the Perseus spiral arm. Another possibility is that the O\\,{\\sc vi} absorption is due to a highly ionized envelope of the LLIV Arch.  4. Six intergalactic absorption lines of H\\,{\\sc i}, C\\,{\\sc iii} and O\\,{\\sc vi} are found at redshifts of $z=0.019$ and $0.102$. The latter absorber is probably associated with the quasar PG\\,0804+761 itself. The non-detection of intervening O\\,{\\sc vi} absorbers adds information about the number density per unit redshift of intergalactic O\\,{\\sc vi} absorbers at low redshift."
    },
    "0010/astro-ph0010519_arXiv.txt": {
        "abstract": "The ($J,K$) = (1,1), (2,2), (3,3), and (4,4) inversion lines of ammonia (NH$_3$) have been detected toward the nuclear 40$''$ sized bar of the nearby spiral galaxy Maffei 2. The relative intensities of the ammonia lines are characterized by a rotational temperature of 85\\,K. This is higher than rotational temperatures measured toward IC\\,342 and most Galactic Center clouds, implying kinetic temperatures $\\ga$100\\,K. Since the kinetic temperature of the gas is larger than that of the dust, NH$_{3}$ is tracing a particularly dense warm gas component that is heated by young massive stars, cloud-cloud collisions, or ion-slip heating in the nuclear starburst. The gas north of the nucleus ($V_{\\rm LSR}$ = --80\\,\\kms) is more highly excited than the gas further south (+6\\,\\kms). This asymmetry might be related to pronounced morphological distortions that are observed in the north-eastern part of the galaxy.  \\ \\ \\ \\ \\\\ ",
        "introduction": "For the past decade, it has been possible to study the excitation of the molecular gas in external galaxies (e.g. Mauersberger \\& Henkel 1989; Mauersberger et al. 1990; Jackson et al. 1995; H{\\\"u}ttemeister et al. 1997; Paglione et al. 1997; Heikkil{\\\"a} et al. 1999; Mao et al. 2000). While the interpretation of CO emission often requires the application of complex models that incorporate temperature gradients at the surfaces of clouds, species like CS, HCN, or HC$_3$N trace higher density gas closer to the cloud cores and may thus be properly modeled with the standard `Large Velocity Gradient' approach. Observed line intensity ratios of such high density tracers depend mostly on the number density \\numd, but also on the kinetic temperature $T_{\\rm kin}$ of the gas. It is virtually impossible to disentangle these parameters, making observations of a molecule tracing exclusively temperature highly desirable.  Ammonia (NH$_3$) is such a molecule. With a single telescope-receiver configuration at 18--26\\,GHz, a large number of NH$_3$ inversion transitions can be measured, covering an enormous range of molecular excitation levels. These include the ($J,K$) = (1,1) to (4,4) inversion lines at 23, 65, 123, and 199\\,K above the ground state. Extragalactic ammonia was first detected in the ($J,K$) = (1,1) line toward the nearby face-on spiral IC\\,342 (Martin \\& Ho 1979). This was followed by detections of the (2,2), (3,3), and (4,4) lines (Ho et al. 1982, 1990; Martin \\& Ho 1986) that permitted a first estimate of the kinetic temperature of the dense nuclear gas in an external galaxy.  In view of the four ammonia lines observed toward IC\\,342 (the (6,6) line was also tentatively detected) it is surprising how little is known about ammonia in other external galaxies. Motivated by our unpublished detections of ($J,K$) = (1,1) and (2,2) emission in 1994 and by recent improvements in receiver sensitivity, baseline stability, and accessible bandwidth, we present a multilevel ammonia study of Maffei 2, an optically obscured barred spiral galaxy that is located behind the plane of the Milky Way at a distance of $\\sim$2.5\\,Mpc (e.g. Karachentsev et al. 1997). High lower limits to the kinetic temperature of its dense nuclear gas are derived. ",
        "conclusions": "\\subsection{Interpretation of rotational temperatures}  The $J = K$ ammonia inversion doublets are called metastable because they decay radiatively via slow $\\Delta K$ = $\\pm$3 transitions. As a consequence, the relative population of these states is strongly affected by collisions and thus reflects (e.g. Walmsley \\& Ungerechts 1983) the kinetic temperature of the gas ($T_{\\rm rot}$ $\\leq$ $T_{\\rm kin}$).  One way to explain rising rotational temperatures with increasing energy above the ground state is the presence of kinetic temperature gradients. Alternatively, optically thick lines or subthermal excitation can produce the same effect (Goldsmith \\& Langer 1999). Accounting for these latter two effects, radiative transfer calculations (e.g. Walmsley \\& Ungerechts 1983; Schilke 1989) predict that rotational temperatures approach the kinetic temperature with rising energy of the analysed metastable inversion doublets. This yields rising $T_{\\rm rot}$ with rising $E$/k. It is therefore likely, based on the corrections of the models, that the true kinetic temperature is $>$100\\,K. $T_{\\rm kin}$ should be larger for the --80 than for the +6\\,\\kms\\ velocity component.  \\subsection{A comparison with IC\\,342 and Galactic Center clouds}  NH$_3$ integrated line temperatures are {\\it larger} towards Maffei 2 than towards IC\\,342, also located at $D$$\\sim$2.5\\,Mpc (Karachentsev et al. 1997). Emission from the inner 500\\,pc of the Milky Way would show similar intensities if observed from this distance (e.g. H{\\\"u}ttemeister et al. 1993a,b).  From the ($J,K$) = (1,1) to (4,4) lines, $T_{\\rm rot}$$\\sim$85\\,K in Maffei 2, but only 50$\\pm$10\\,K in IC\\,342 (Martin \\& Ho 1986). Toward peak positions of clouds near the center of the Galaxy, (1,1) lines tend to be optically thick, while more highly excited inversion lines show no significant optical depth effects (H{\\\"u}ttemeister et al. 1993b). There are two main temperature components, one at $T_{\\rm kin}$ $\\sim$ 25 and one at $\\sim$200\\,K. These, if present in Maffei 2 as well, could be responsible for the increasing rotational temperatures with $E$/k (Fig.\\,\\ref{fig:trot}). Such gradients are also seen in IC\\,342, if the (6,6) line is as strong as suggested by the spectrum presented by Martin \\& Ho (1986).  \\subsection{The nuclear region of Maffei 2}  What kind of molecular gas is seen in ammonia? Critical densities, where collisional de-excitation matches spontaneous emission rates, are \\numd\\ $\\sim$ 3$\\cdot$10$^{3}$\\percc\\ (e.g. Green 1980; Ho \\& Townes 1983). Our beam size (40$''$) matches the central bar with its size of 43$''$$\\times$9$''$ in $^{12}$CO {\\hbox{$J$ = 1--0}} (Ishiguro et al. 1989) and 40$''$$\\times$12$''$ in $^{13}$CO $J$ = 1--0 (Hurt \\& Turner 1991). The presence of bars provides a viable mechanism to transfer interstellar gas into the nuclear region of a galaxy, leading to the formation of dense molecular clouds and triggering bursts of star formation. As long as the gas is not yet fully ionized or swept away by stellar winds, it will coexist with the newly formed stars and will be heated by their radiation, by an enhanced flux of cosmic rays, by cloud-cloud collisions, and by the dissipation of turbulent energy. Both Maffei 2 and IC\\,342 show prominent nuclear bars and strong CO, infrared, and radio continuum emission (e.g. Lo et al. 1984; Ishiguro et al. 1989; Ishizuki et al. 1990; Hurt \\& Turner 1991; Turner \\& Ho 1994; Meier et al. 2000).  Assuming a grain emissivity law of $\\lambda^{-1}$, dust temperatures deduced from the Kuiper Airborne Observatory and, for a larger region, from IRAS data (Rickard \\& Harvey 1983; IRAS 1989) are $T_{\\rm d}$ = 33--47\\,K. Ignoring non-metastable inversion doublets, Table 1 implies that $N$(NH$_3$) is on the order of 10$^{14}$\\,\\cmsq. For H$_2$, with $D$=2.5\\,Mpc, $N$(H$_2$)$\\sim$10$^{22}$\\,\\cmsq\\ (Hurt \\& Turner 1991). This yields a relative abundance of $X$(NH$_3$)$\\sim$10$^{-8}$. With $T_{\\rm rot}$(NH$_3$) $\\gg$ $T_{\\rm d}$ and an unusually small NH$_3$ abundance (e.g. Benson \\& Myers 1983; Brown et al. 1988), ammonia should trace a gas component that is different from that seen in CO (for an analysis of CO emission in a nuclear starburst environment, see Mao et al. 2000). Cloud-cloud collisions, young massive stars, or ion-slip heating (H{\\\"u}ttemeister et al. 1993a,b) may cause the high rotational temperatures. SiO, an exclusive tracer of high temperature gas, was detected in Maffei 2 (Sage \\& Ziurys 1995) and is likely arising from the same gas component. Detailed studies of NH$_3$ and SiO in the Galactic Center region (e.g. H{\\\"u}ttemeister et al. 1993a,b, 1998) indicate that the warm gas not directly associated with massive stars is only moderately dense (\\numd $\\sim$5$\\cdot$10$^{3}$\\percc).  From a comparison with the interferometric maps of Ishiguro et al. (1989) and Hurt \\& Turner (1991), the ammonia component at +6\\,kms\\ must arise SW, the {\\hbox{--80\\,\\kms}} component NE of the center of the galaxy. Why is the rotation diagram (Fig.\\,\\ref{fig:trot}) indicating higher excitation NE of the nucleus? The bar in Maffei 2 may be caused by a merger with a satellite galaxy in the NE that severely disrupts this side of the galaxy (Hurt et al. 1996). This might have led to differences in inflow rates, star forming rates, and gas heating at the north-eastern and south-western sides of the nucleus. While our observed small scale ($\\sim$500\\,pc) nuclear asymmetry in rotational temperature may be consistent with the large scale asymmetry observed in H{\\sc i}, the proposed connection between these phenomena remains speculative. More data on the putative interacting dwarf galaxy, detailed simulations of the interaction, and models of the evolution of the nuclear bar are needed to verify or to reject the proposed scenario and to explain the recent past of Maffei 2.  \\begin{acknowledgement}  We wish to thank J.L. Turner for critically reading the manuscript.  \\end{acknowledgement}"
    },
    "0010/astro-ph0010033_arXiv.txt": {
        "abstract": "We use a combination  of three-dimensional Monte Carlo radiative transfer techniques and  multi-wavelength ($BRHK$, H$\\alpha$) imaging data to investigate the nature of the interstellar medium (ISM) in the edge-on, low surface brightness (LSB) galaxy UGC~7321. Using realistic models that incorporate multiple scattering effects and clumping of the stars and the interstellar material, we explore the distribution and  opacity of the interstellar material (gas+dust), and its effects on the observed stellar disk luminosity profiles, color gradients, and rotation curve shape.  We find that UGC~7321 contains a small but non-negligible dusty component to its ISM, yielding a $B$-band optical depth $\\bar\\tau_{e,B}\\sim$4.0 from disk edge to center. A significant fraction ($\\sim50\\pm10$\\%) of the interstellar material in the innermost regions of UGC~7321 appears to be contained in a clumpy medium, indicating that LSB galaxies can support a modest, multi-phase ISM structure.  In spite of the clear presence of dust, we conclude that the large radial optical color gradients observed in UGC~7321 and other similar LSB spiral galaxies cannot be accounted for by dust and must result primarily from significant stellar population and/or metallicity gradients. We show that realistic optical depth effects will have little impact on the observed rotation curves of edge-on disk galaxies and cannot explain the linear, slowly rising rotation curves seen in some edge-on LSB spirals. Projection effects create a far larger uncertainty in recovering the true underlying rotation curve shape of galaxies viewed at inclinations $i\\ga85^{\\circ}$. ",
        "introduction": "\\subsection{Background} Low surface brightness (LSB) disk galaxies can be broadly defined as rotationally-dominated galaxies with extrapolated face-on central surface brightness at least $\\sim$1 magnitude fainter than the canonical Freeman (1970) value of $\\mu_{B,0}=$21.65~mag arcsec$^{-2}$. Although LSB galaxies span the full range of Hubble types (e.g., Schombert \\etal\\ 1992), here we use the term primarily in reference to the latest type spirals (Scd-Sdm).  Numerous observational studies have now shown that LSB galaxies are not ``abnormal'' objects, but rather are a common product of disk galaxy formation and evolution. More extensive summaries of the properties of LSB galaxies can be found in Bothun, Impey, \\& McGaugh (1997) and Impey \\& Bothun (1997) and references therein.  In spite of the vast body of multiwavelength observational data now accumulated on LSB galaxies, several aspects of these systems still remain enigmatic. One important example is the  detailed nature of their interstellar medium (ISM). Although the bulk of LSB galaxies are known to be rich in neutral hydrogen gas (\\HI), they are generally presumed to be dust and molecule-poor systems. Numerous observations directly and indirectly support this picture, including: low metal abundances (e.g., McGaugh 1994; R\\\"onnback \\& Bergvall 1995), low statistically inferred internal extinctions (e.g., Tully \\etal\\ 1998; Matthews, van Driel, \\& Gallagher 1998), strong similarities between optical and near-infrared (NIR) morphologies (Bergvall \\etal\\ 1999); undetectable CO fluxes (Schombert \\etal\\ 1990; Knezek 1993; de Blok \\& van der Hulst 1998; but see Matthews \\& Gao 2000), and the high transparency of their stellar disks (e.g., O'Neil \\etal\\ 1998; Matthews \\etal\\ 1999; Matthews, Gallagher, \\& van Driel 1999, hereafter MGvD99). Theoretical and numerical models also suggest that the low surface densities and low metallicities of LSB galaxies cannot support significant molecular gas fractions (e.g., Mihos, Spaans, \\& McGaugh 1999; Gerritsen \\& de Blok 1999). Nonetheless, since most presently known LSB galaxies are actively star-forming systems, having blue disk colors and at least modest amounts of H$\\alpha$ emission, it seems likely that at least some molecular material must be present in these systems.  Even if dust and molecular gas contents in LSB galaxies are typically small, extinction and reddening due to interstellar material may still have a non-negligible impact on the observed properties of these galaxies, and molecular gas could still be an important constituent of the ISM (e.g., Spaans 1999).  For example, estimates of dust reddening, even if small, are needed to correctly interpret the stellar populations inferred from broad-band colors and spectroscopic measurements (e.g., Bell \\etal\\ 2000). It has been found that many of the lowest-luminosity LSB galaxies deviate from the standard optical Tully-Fisher relation in the sense that they rotate faster than predicted for their luminosities (e.g., Matthews, van Driel, \\& Gallagher 1998; Stil 1999; McGaugh \\etal\\ 2000), while more luminous LSB galaxies seem to follow the same Tully-Fisher relation as higher surface brightness galaxies (e.g., Zwaan \\etal\\ 1995; Sprayberry \\etal\\ 1995). This appears to have important implications for galaxy formation models (e.g., van den Bosch 2000) and for establishing the possible existence of a ``baryonic'' Tully-Fisher relation (e.g., Matthews, van Driel, \\& Gallagher 1998; McGaugh \\etal\\ 2000); however, the internal extinction corrections appropriate for the LSB galaxies used in these analyses remains a point of contention (e.g., Han 1992; Rhee 1996; Matthews, van Driel, \\& Gallagher 1998; Pierni 1999). Lastly, rotation curves of LSB galaxies offer hope to furthering our insight into the nature of dark matter in galaxies (e.g., de Blok \\& McGaugh 1997; McGaugh \\& de Blok 1998; Kravtsov \\etal\\ 1998; Swaters, Madore, \\& Trewhella 2000). Accurately interpreting high resolution optical rotation curve data, which can play a crucial part in such analyses,  requires an assessment of the role of internal extinction in these systems (see Section~\\ref{curves}). In short, a more extensive and quantitative knowledge of the total amounts and distribution of dust and other molecular material in LSB disks is overall vital to better understanding the global properties as well as the star-forming and evolutionary histories of these systems.  To date, few attempts have been made to model the dust in individual LSB galaxies in a sophisticated manner. One difficulty is that these objects tend to be very weak far-infrared and sub-millimeter sources (e.g., Hoeppe \\etal\\ 1994; Pickering \\& van der Hulst 1999; Bell \\etal\\ 2000), hence direct observations of re-emitted thermal radiation from their dust are lacking. Observing the dust in absorption at optical and near-infrared wavelengths can also be problematic, since for  LSB systems viewed at moderate or low inclinations, the signatures of dust can be difficult to infer against  the patchy and diffuse background light of the stellar disk. The data necessary to test dust models are therefore sparse. For these reasons, {\\it edge-on} examples of LSB galaxies are particularly valuable.  Because the viewing geometries of edge-on LSB galaxies provide maximal path length through the disks, such systems furnish us with the unique opportunity to directly observe the vertical and radial extent of the dust distribution, as well as any associated dust-induced vertical and radial color gradients. With the aid of 3D radiative transfer models, it becomes possible to constrain the amount and distribution of dust and quantify its effects on the observed properties of the galaxy, including: total internal extinction, effects on observed color  gradients, and effects on the disk rotation curve. Properties of the disk viewed at other arbitrary inclinations can also be predicted. Such techniques have been used to explore the effects of dust in more luminous, high surface brightness galaxies seen edge-on (e.g., Kylafis \\& Bahcall 1987; Kuchinski \\& Terndrup 1996; Kuchinski \\etal\\ 1998; Trewhella, Madore, \\& Kuchinski 1999; Xilouris \\etal\\ 1997,1998,1999). Here we present one of the first applications of full 3-D Monte Carlo radiative transfer techniques to studying systems at the low optical depth regime of LSB galaxy disks.  \\subsection{UGC~7321: A ``Superthin'' LSB Galaxy\\protect\\label{7321}}  UGC~7321 is a nearby example of an Sd spiral galaxy with an extraordinarily thin, highly flattened and diffuse stellar disk with no obvious bulge component (MGvD99; Matthews 2000). MGvD99 showed that this ``superthin'' galaxy is actually an example of an LSB spiral galaxy seen near edge-on ($i\\approx88^{\\circ}$). MGvD99 estimated that after correction for projection and internal extinction, UGC~7321 would have a $B$-band central disk surface brightness $\\mu_{B,i}(0)\\sim$23.6~mag arcsec$^{-2}$.  Although the {\\it observed} surface brightness of UGC~7321 is enhanced through projection, several lines of evidence clearly indicate this is an intrinsically LSB galaxy, independent of internal extinction corrections. These include: the extreme transparency of its disk  (e.g., several background galaxies are clearly visible directly through it; MGvD99); the small stellar scale height and very low estimated stellar velocity dispersion (implying a very low disk surface density; Matthews 2000); low emission line intensity ratios (Goad \\& Roberts 1981); weak far-infrared and radio continuum emission (see below); and a rather high $M_{HI}/L_{B}$ ratio (1.1 in solar units).  UGC~7321 is an ideal  LSB galaxy for detailed modelling since it is nearby ($D\\approx10^{+3}_{-3}$~Mpc; Gallagher \\etal\\ 2000) and well-resolved.  Some key properties of UGC~7321 are summarized in Table~1. Although UGC~7321 is physically smaller, less massive, and less luminous than most other well-studied edge-on spirals, UGC~7321 appears to be a prototypical example a relatively common class of late-type LSB galaxies seen on edge (see also Goad \\& Roberts 1981; Karachentsev, Karachentseva, \\& Parnovsky 1993; Bergvall \\& R\\\"onnback 1995; Dalcanton \\& Schectman 1996; Gerritsen \\& de Blok 1999; Matthews \\& van Driel 2000). Possible face-on analogs of UGC~7321 might be such systems as NGC~4395 (see Sandage \\& Bedke 1994) or ESO~305-009 (see Matthews \\& Gallagher 1997).  \\subsubsection{The Data\\protect\\label{data}} Much of the primary observational data we use for our present analysis were described in MGvD99; these include near-infrared (NIR) $H$-band and optical ($B$- \\& $R$-band) imaging data, and narrow-band H$\\alpha$ imaging of UGC~7321. In addition, we supplement these data with observations from several additional sources. {\\it Hubble Space Telescope} ({\\it HST}) Wide Field and Planetary Camera~2 $F702W$ and $F814W$ ($R$- and $I$-band equivalent) imaging observations of the inner disk regions of UGC~7321 (Figure~\\ref{fig:WFPC2}) are an important component of our study and will be discussed further in Gallagher \\etal\\ (2000). We also used the optical longslit rotation curve data of Goad \\& Roberts (1981), and NIR $H$ and $K_{s}$ images of UGC~7321 obtained from the Two Micron All Sky Survey (2MASS) database (http://sirtf.jpl.nasa.gov/2mass/).  \\subsubsection{Color Gradients in UGC~7321\\protect\\label{truegrad}} One particularly intriguing finding of MGvD99 is that UGC~7321 exhibits quite strong radial color gradients: $\\Delta (B-R)\\sim$1.0 magnitude (before correcting for internal reddening), with the outer disk regions being considerably bluer than the galaxy center (Figure~\\ref{fig:BRreal}). These authors interpreted this as evidence of significant stellar age gradients as a function of radius in the disk, and suggested that this is a galaxy that is likely to have evolved relatively slowly, from the inside out. Large radial color gradients [$\\Delta (B-R)\\ga 0.5$] were also reported for a sample of 3 additional superthin galaxies by Matthews, Gallagher, \\& van Driel (2000). The strength of the observed color gradients in UGC~7321 may also indicate that viscous evolution has been minimal in this disk (cf. Firmani, Hern\\'andez, \\& Gallagher 1996), leading MGvD99 to postulate that galaxies like UGC~7321 may be some of the most pristine star-forming disks in the nearby universe, and hence ideal systems in which to explore disk evolutionary processes.  MGvD99 also reported significant {\\it vertical} color gradients in the UGC~7321 disk (as large as $\\Delta (B-R)\\sim$0.45, with the colors reddening as a function of increasing $z$ height). Such gradients are predicted to occur in galaxy disks as a result of dynamical heating processes (e.g., Just, Fuchs, \\& Wielen 1996), but in practice they are difficult to observe and interpret in most edge-on galaxies due to the effects of seeing, contamination from bulge light, and particularly due to contamination from dust. Thus relatively few empirical constraints exist for present models of vertical disk heating.  A critical assumption in interpreting the physical significance of the large color gradients observed in the UGC~7321 disk, as well as the significant gradients observed in the disks of other less inclined LSB spirals  (e.g., de Blok, van der Hulst, \\& Bothun 1995; Bell \\etal\\ 2000) is that reddening due to dust is small. For example, in a sample of ``normal'' edge-on spirals, de Grijs (1998) argued that to within observational errors, the observed color gradients and scale length differences in different wavebands could be explained solely from dust.  Based on {\\it HST} imaging  observations, Matthews (1998) and Gallagher \\etal\\ (2000) have shown that although UGC~7321, as well as another nearby, edge-on LSB system UGC~711 lack the quintessential dust lanes seen in brighter edge-on spirals (cf. Howk \\& Savage 1999), they are not dust-free. Nonetheless, using a simple foreground screen model, MGvD99 suggested that in the case of UGC~7321, dust reddening could not explain more than $\\sim$0.2 magnitudes of the observed $B-R$ color gradient. Using both foreground screen and Triplex dust models (see Disney, Davies, \\& Phillipps 1989), Bell \\etal\\ (2000) argued that the amount of dust reddening needed to produce the observed color gradients in their sample of less inclined LSB galaxies was also too large to be consistent with observations.  To obtain a first estimate of the role of dust in UGC~7321, MGvD99 employed a simple foreground screen model. These authors sought not to derive a realistic dust model for UGC~7321, but rather to test the possibility that a significant fraction of the observed color gradient in UGC~7321 could be caused by dust. To put an upper limit on the dust-induced color gradient, these authors  assumed that the galaxy was effectively optically thin in $H$, and that the entire observed $R-H$ radial color gradient could be due to dust.  A reanalysis of the $R-H$ color profile of UGC~7321 (Figure~\\ref{fig:RHreal}), actually shows a significantly larger color gradient ($\\sim 0.9\\pm0.1$ mag) than that reported by MGvD99. This difference was traced to an error by MGvD99 in extracting the $H$-band radial axis profile. However, because MGvD99 were extremely conservative in their attribution of the full $R-H$ color gradient to possible dust reddening, as we further demonstrate here, the validity of their final conclusions remains unaffected.  Although the arguments presented by MGvD99 and Bell \\etal\\ (2000) regarding the dust reddening in their sample LSB galaxies are expected to be for the most part robust, they nonetheless relied on simplifying assumptions, including the  use of smooth, idealized dust geometries and the neglect of scattering effects. And because the color gradients observed in UGC~7321 and some other LSB galaxies are often found to be large compared with those typical in most normal surface brightness spiral galaxies (cf. de Jong 1996), it is important to establish  (1) whether galaxies like it are in fact a genre of nearby disks with some of the largest intrinsic color gradients, (2) whether these are pathological LSBs that contain more dust than average, and (3) whether our knowledge of the dust in typical LSB galaxies is somehow incomplete. Detailed studies of individual galaxies rather than statistical studies are needed to address these questions. If the large observed color gradients are intrinsic (i.e., due to true stellar population and/or metallicity gradients) rather than a result of dust reddening, this is important not only for understanding  the evolution and stellar contents of these and other LSB disks, but also because such gradients may imply non-negligible changes in the stellar mass-to-light ratio ($\\Upsilon_{\\star}$)  of disks as a function of radius. Such trends could have an important consequence for mass-modelling of LSB galaxies (i.e., for determining their dark matter halo parameters from rotation curves; e.g., Kent 1986) and understanding the underlying physics behind the Tully-Fisher relation (cf. Matthews, van Driel, \\& Gallagher 1998; Stil 1999; McGaugh \\etal\\ 2000). It is of interest therefore to derive sophisticated, realistic dust models of UGC~7321 as a first test case.  We note that an advantage of studying color gradients in edge-on or near edge-on disks is that due to the higher projected surface brightnesses of the outer disk compared to a face-on system, we can better observe the faintest (and perhaps least evolved) outer disk regions. The trade-offs are of course the uncertain effects of dust toward the central disk regions, as well as the possible superposition of many different stellar populations along a given line of sight. This makes the use of full three-dimensional (3D) models critical for unravelling the effects of dust on the observed colors and inferred stellar population distributions. This is the approach we utilize in the present work.  \\subsection{The Present Study} The fundamental questions this paper seeks to address are: \\begin{itemize} \\item What is the integrated optical depth of the dusty medium  in UGC~7321, and what are its total mass and distribution?  \\item What internal extinction corrections are appropriate for edge-on LSB galaxies?  \\item How much of the observed color gradients in UGC~7321 and other similar galaxies can be attributed to dust?  \\item What are the predicted dust-induced color gradients for analogous LSB disks seen at alternate viewing angles?  \\item What is the impact of optical depth effects on the shape of optically-derived rotation curves of LSB galaxies observed near edge-on?  \\end{itemize}  To answer these questions we use a suite of high resolution, 3D Monte Carlo radiation transfer codes that simulate scattered light images in multiple wavebands and also can be used to model the effects of dust on optical rotation curves. As described below, our Monte Carlo radiation transfer code has recently been modified (see Wood \\& Reynolds 1999 and Section~\\ref{radtreat}) to enable the efficient simulation of 3D, optically thin systems, thus making it ideal for our present study of LSB galaxies. In doing this we also lay the groundwork for future studies of the nature of dust in other LSB galaxies. Much of the general framework required for realistically modelling the effects of dust in galaxies has been discussed extensively by others (e.g., Kylafis \\& Bahcall 1987; Bosma \\etal\\ 1992; Xilouris \\etal\\ 1997,1998,1999), including the application of Monte Carlo methods to galactic environments (e.g., de Jong 1996; Witt \\& Gordon 1996,2000; Kuchinski \\etal\\ 1998; Bianchi, Ferrara, \\& Giovanardi 1996; Bianchi \\etal\\ 2000). We do not attempt to repeat the detailed findings of these works here, but rather apply our codes to explore specific questions relating to LSB galaxies. In particular, in this paper we focus on using our Monte Carlo models to reproduce the observed properties of the edge-on LSB galaxy UGC~7321 and to predict the role dust may play on the observed properties of similar LSB galaxies viewed over a range of inclinations. ",
        "conclusions": "We have presented some of the first 3D Monte Carlo simulations of the dusty ISM properties in the low optical depth regimes encountered in low surface brightness (LSB) spiral galaxies. We have demonstrated the power of using such techniques in combination with multiwavelength observational data to constrain the amount and distribution of dust in such systems. Our realistic models fully take into account scattering and the  effects of a clumpy, multi-phase ISM.  Because late-type LSB galaxies are relatively optically thin, edge-on examples of these systems allow a unique opportunity to explore the structure and distribution of their ISM and place important empirical constraints on dust models. We have shown that while dust contents in LSB galaxies are relatively low, these galaxies can contain modest amounts of dust and molecular material. Moreover, in at least some cases, a significant fraction of this dusty material is contained within a clumpy medium, confirming that even LSB galaxies can maintain sufficient pressures to support a modest multi-phase ISM structure.  From detailed models of the nearly edge-on LSB galaxy UGC~7321 ($i=88^{\\circ}$), our Monte Carlo models indicate an edge-on optical depth $\\bar\\tau_{e,B}\\sim$4.0, and hence a total mass of dust is this galaxy of $\\sim 8.8\\times10^{5}$~\\msun. This is $\\sim20\\times$ the warm dust content inferred from {\\it IRAS} 100$\\mu$m measurements.  Total $B$-band extinction toward the center of this galaxy is estimated to be $\\sim0.43\\pm0.2$ mag.  We infer that $\\sim$50\\% of the dusty material in UGC~7321 is contained in a clumpy medium. Based on the detection of CO emission from UGC~7321 by Matthews \\& Gao (2000), it appears that some fraction of the clumped material is almost certainly molecular gas. However, the dark clouds in LSB galaxies may also contain significant fractions of atomic hydrogen gas. Together our models and direct \\HI\\ and CO measurements  suggest that UGC~7321 has a gas-to-dust ratio of at least 850 in the inner regions of its stellar disk, while the global gas-to-dust ratio for the entire galaxy (including \\HI\\ gas extending beyond the stellar disk) is estimated to be $\\sim$1600.  In spite of the modest dust contents of LSB galaxies, we have shown that for the range of optical depths expected for such systems, dust extinction will have appreciable effects on the observed total magnitudes and colors only when these galaxies are observed near edge-on. However, even then, the amount of reddening due to dust will be insufficient to explain the large  radial color gradients now observed in a number of LSB galaxies (de Blok \\etal\\ 1996; MGvD99; Matthews \\etal\\ 2000; Bell \\etal\\ 2000). Thus some LSB galaxies must have large intrinsic stellar population and/or metallicity  gradients in their disks.  We have shown that for realistic optical depths, dust has no appreciable effect on the  rotation curves of edge-on spiral galaxies observed in the H$\\alpha$ emission line.  Possible holes in the central  H$\\alpha$ distributions of such galaxies also cannot fully account for the observed slowly rising rotation curves. Thus the linearly, slowly rising rotation curves seen frequently in many  late-type, edge-on or nearly edge-on  spirals cannot be fully accounted for by dust and must be due to low central matter densities in these systems.  We have demonstrated that in general, projection effects will create a far greater uncertainty than optical depth effects in interpreting the precise intrinsic shapes of the rotation curves of edge-on galaxies."
    },
    "0010/astro-ph0010205_arXiv.txt": {
        "abstract": "We compute the bispectrum for the \\IRAS PSC$z$ catalog and find that the galaxy distribution displays the characteristic signature of gravity. Assuming Gaussian initial conditions, we obtain galaxy biasing parameters $1/b_1=1.20^{+0.18}_{-0.19}$ and $b_2/b_1^2=-0.42\\pm0.19$, with no sign of scale-dependent bias for $k\\leq 0.3$ h/Mpc. These results impose stringent constraints on non-Gaussian initial conditions. For dimensional scaling models with $\\chi^2_N$ statistics, we find $N>49$, which implies a constraint on primordial skewness $B_3<0.35$. ",
        "introduction": " ",
        "conclusions": ""
    }
}