nopperl/arxiv-image-text · Datasets at Hugging Face

dedup-isc-ft-v107-score float64 0.32 1	uid stringlengths 32 32	text stringlengths 0 32.5k	paper_id stringlengths 1 14	original_image_filename stringlengths 5 222
0.419077	cd3a3c279d3644648d4666a53e0b3d18	The same as Fig.<ref>, but for Q0405-123.	astro	ph0408105-Q0405-123.jpg
0.45764	d1d9b61cc275406eb210a763d04fe039	The same as Fig.<ref>, but for Q0405-123. We calculated and plotted the line polarization of Hβ+[OIII] region (4731-5032Å) instead of Hβ only, since excluding the telluric absorption wavelengths at around 7600Å leaves a very small wavelength window for Hβ line.	astro	ph0408105-Q0405-123_qu.jpg
0.402824	f407a3e66a31476d8d4df3e8c739750b	The same as Fig.<ref>, but for Q0414-060.	astro	ph0408105-Q0414-060.jpg
0.423725	0ef494fd0aef4411a2b7afc7415c4aa8	The same as Fig.<ref>, but for the Keck observation of Q1004+130.	astro	ph0408105-Q1004+130.jpg
0.558853	66faefbc2910491b8f3666b5992aa76b	The same as Fig.<ref>, but for Q1004+130.	astro	ph0408105-Q1004+130_qu.jpg
0.458486	464a75a934a14380b632fc7707f79e5c	The same as Fig.<ref>, but for Q1912-550.	astro	ph0408105-Q1912-550.jpg
0.423323	c53ed054b36c4d05954868e86037494d	The ISP-corrected polarization of Q1912-550.	astro	ph0408105-Q1912-550_cor010.jpg
0.457019	3431fa9f8ba24f9bb0ea5559dd5846aa	The same as Fig.<ref>, but for Q1912-550.	astro	ph0408105-Q1912-550_qu.jpg
0.370544	ad652668fe8c446f83e957e49abf4a2b	The same as Fig.<ref>, but for Q2115-305. The reference axis of q' and u' is taken along the PA of the polarization integrated over the whole wavelength region, rather than at 4000-4731Å.	astro	ph0408105-Q2115-305.jpg
0.426178	5b086ae59f03456c980b29da443ef65e	The same as Fig.<ref>, but for Q2251+113.	astro	ph0408105-Q2251+113.jpg
0.407383	7b244e5883204d82b83b8cb56ceb79a3	The same as Fig.<ref>, but for Q2349-014.	astro	ph0408105-Q2349-014.jpg
0.453972	fb186e916797477c8115538dec951e51	The same as Fig.<ref>, but for the Keck observation of Ton202.	astro	ph0408105-Ton202_May2003.jpg
0.397882	9c04234e7be644d6b1636e3720bb97de	Comparison of the observations of Ton202 at two epochs. The upper panel show the polarization observed in May 2003 minus that in May 2002. The lower panel show the total flux in May 2003 divided by that in May 2002.	astro	ph0408105-Ton202_dif.jpg
0.471387	ae90fcd98d774ca393ab977768aa3f00	As an illustration, a spectrum (thick smooth line) for a face-on disk model around a Kerr (a/M=0.998) black hole of mass 8×10^9 M_⊙, accreting at 2 M_⊙ yr^-1 is plotted over the May 2002 polarized flux spectrum of Ton202 (corrected for the Galactic reddening). The model spectrum has arbitrary normalization.	astro	ph0408105-Ton202_model.jpg
0.449296	82d84b4808cf47e7ab0b6d1c855c7f76	In the upper panel, the five polarized flux spectra shown in Fig.<ref>, normalized by the mean at 2800-5000Å, are over-plotted with a larger wavelength bin of 20Å. In the lower panel the average of these five spectra with equal weighting is shown. Both panels are in F_λ.	astro	ph0408105-coadd_mobj.jpg
0.428631	14e7de4749434b048f87272a8b661086	The same as Fig.<ref>, but in ν F_ν with both axes in log scale.	astro	ph0408105-coadd_mobj_nFn.jpg
0.4158	f1185b6092354a16818d97681de64d4a	The spectral index α (f_ν∝ν^α) corresponding to the flux ratio at 3050Å and 4200Å is plotted against ΔPA.	astro	ph0408105-dpa-alpha_sbb.jpg
0.470417	d6ed3fb99ab44afb83abfe9834d3202e	The ratio of the polarization at 2891-3600Å to that at 4000-4731Å is plotted against ΔPA.	astro	ph0408105-dpa-pratio.jpg
0.506688	9516f2dad0ce4b44bafd0136b855aab0	The histogram for the difference of the radio axis PA and optical polarization PA.	astro	ph0408105-hist.jpg
0.492364	50d69f17ce324d899e3cc5e6cb6d5fd8	The profiles of the equilibrium quantities plotted against dimensionless x/L_x coordinate: density (panel A), pressure (panel B), the magnetic field (panel C) and plasma beta (panel D)	astro	ph0408114-figure2.jpg
0.457113	60e3a41a1ad24719871b8f070ea62f81	Sample solutions of the standing fast magnetosonic modes. Panel A: zeroth-order harmonic n=0, period T=5.6069 min., α=0.015 ; Panel B: second-order harmonic n=2, period T=1.8514 min., α=0.03	astro	ph0408114-figure3.jpg
0.442963	6f757d6de32842508faff0b35e95f7c2	Results of numerical simulations for the standing fast magnetosonic mode, shown on panel A in Figure <ref>. Panel A_1: the curves of F_1 (Eq. (<ref>)) (solid line) and F_2 (Eq. (<ref>)) (dotted line) vs. fractional distance x/L_x for the zeroth-order standing fast mode n=0 and the standing Alfvén mode with m=3; Panel B_1: The numerical solution of the Alfvén mode m=3 amplitude at location x/L_x≈ x_0,4/L_x=0.0767; Panel C_1: The numerical solution for mode m=3 at x/L_x≈ x_0,3/L_x=0.7967; Panel A_2: As in panel A_1 for mode m=4; Panel B_2: As in panel B_1 for mode m=4; Panel C_2: As in panel C_1 for mode m=4.	astro	ph0408114-figure4.jpg
0.435974	feb80409654a414a9bb850a42c30beaf	As in Figure <ref> for the second-order harmonic of the fast magnetosonic mode shown on Panel B in Figure <ref> (n=2) and the standing Alfvén with m=8 and m=10. The numerical solutions respectively are given at distances x_1/L_x ≈ x_2,10/L_x = 0.6000 (panels B_1 and B_2), x_2/L_x ≈ x_2,8/L_x=0.9933 (panels C_1 and C_2). Dashed line in panel A_1 is curve of F_1 (<ref>) corresponding to the mode m=9.	astro	ph0408114-figure5.jpg
0.469018	101690c961d84b0ca859b141c6b4c16a	As in Figure <ref> for the standing Alfvén modes with m=11 and m=12 at respective resonant points x_1/L_x ≈ x_2,12/L_x = 0.1033 (panels B_1 and B_2), x_2/L_x ≈ x_2,11/L_x=0.3633 (panels C_1 and C_2).	astro	ph0408114-figure6.jpg
0.421282	cbbffa93e4924e358a821e4b4751c8df	Histogram of the rank correlation coefficients calculated from 10^5 random samples drawn from the observed stellar metallicity and orbital period distributions. The observed values discussed in Section 2 (arrows of different styles) have between 6.8% and 0.3% chances of being consistent with no correlation	astro	ph0408116-fig2.jpg
0.470255	f45835c5a2e04578a3bd50f61f3ceb1c	The lightcurve brightness distribution for the simulated dataset.	astro	ph0408118-LCMagDistNonCm.jpg
0.463042	59ab71c0d7134b31b69893aea1b34063	The final iteration of the ROTSE data map (shown in Fig. <ref>) when processed with the U-Matrix algorithm. The colour scale runs from dark blue (similar reference vectors) to red (dissimilar reference vectors).	astro	ph0408118-ROTSENetUMat34.jpg
0.532738	76a9bf01bcbb4e82ba93f22abfe62dc2	A map of the number of training set members assigned to each neuron for the network trained with the ROTSE data. The raw map has been filtered with a 3x3 top-hat kernel. The colour scale runs from black (low occupancy) to red (high occupancy).	astro	ph0408118-SmoothedDistCOL34.jpg
0.478964	15b98d5a0f534616b91613ce6b8d0f43	The average Euclidean distance moved by a training set member in map space from one iteration to the next, . Panel order is the same as Figure <ref>.	astro	ph0408118-dbar2.jpg
0.446258	158732597c534a479089a27d170b13bb	Samples of the four classes of simulated lightcurves, representing type-c RR Lyrae, δ Scuti and Cepheid types (top left), type-ab RR Lyrae (top right), short-period (bottom left) and long-period eclipsing binary (bottom right)	astro	ph0408118-fakelcs.jpg
0.458682	fdc1570674ff40a4b804a70974ea43b3	Evolution of a network trained with our synthetic data and with α(t)=α_0=0.01. Panels show (from left to right, top to bottom), iterations 1, 4, 7, 10, 13, 16, 19, 22, 25, 28, 31, 34. The four classes of synthetic lightcurve (described in Section <ref>) are denoted by (i) squares, (ii) crosses, (iii) diamonds and (iv) stars. Note that, for the purposes of visualisation, the positions of all lightcurves have been randomised within their host neurons.	astro	ph0408118-mapevol2.jpg
0.440647	6a57dcb52ee844fba53cf0fd8a9618ae	The number of training set members that move from one iteration to the next, . Panel order is the same as Figure <ref>.	astro	ph0408118-nmove3.jpg
0.500251	eb0a15f60d61433e82955482957b739c	The final structure of a network trained using 1206 pre-classified ROTSE lightcurves.	astro	ph0408118-rotse_map3.jpg
0.483808	5a60991d545e449b9b7ebbace26792b8	The ROTSE map of Fig. <ref> collapsed in one dimension and expanded with the period of the variable star. Here the δ Scuti, Cepheid and type-c RR Lyrae variables, which formed an overlapping group in Fig. <ref>, have been separated in period. This illustrates how additional information can be used to resolve ambiguity in our maps, which are based on shape alone.	astro	ph0408118-rotse_map3_period.jpg
0.424123	7b8bc887f9af4e6a875df7f21558ac69	The average “goodness of fit”, . In the leftmost column the learning parameter α is held constant throughout the learning phase. In the central column α falls linearly, and in the rightmost column exponentially. The initial value of α assumes the value 0.9 (top row), 0.5 (second row), 0.1 (third row) and 0.01 (bottom row).	astro	ph0408118-sbar2.jpg
0.449691	a4555a57c80a4ec9b71e126154bd27aa	The dependence of the apsidal precession rate ϖ̇ on satellite's eccentricity, for a near-planar prograde satellite (m=0.08, i=5^∘) moving about a low-eccentricity pseudo-Jupiter (P'=11.86 yr, but e'=0.01). Crosses plot the result of a symplectic integration, while lines plot the predictions of secular theories. The precession rates for the latter were obtained by advancing the secular equations of motion using a Burlisch-Stoer numerical integrator. The short-dashed line plots the predictions of Kozai's theory, the long-dashed curve also includes the evection terms, while the solid line depicts the results of the mixed model derived in this paper (Eqs. <ref> and <ref>).	astro	ph0408119-Cuk.fig1.jpg
0.389291	ec5e7699cfd540899399207d43a877c9	The dependence of eccentricity variation parameter η (defined in Eq. <ref>) on the precession rate of a particle's ϖ for 440 test particles integrated numerically over 30,000 yrs (see text for details). Secular and the Great Inequality resonances are prominent as regions of high η; the onset of Kozai resonance is also visible, with all the particles to the left of the discontinuity being Kozai librators.	astro	ph0408119-Cuk.fig10.jpg
0.5318	f6d6922839094caeb99a844c344074b1	The position of the Great-Inequality resonance in m-i_min space. The top three continuous lines plot (in descending order) the Great Inequality resonance for maximum eccentricities of 0.2, 0.4 and 0.6, respectively. The fourth line plots the position of the Great Inequality resonance for e_max=0.6, in a planetary configuration that has 5 λ_S - 2 λ_J=0.002 yr^-1 (i.e., the Great Inequality has a period of 500 yrs). The orbital elements of the known prograde Saturnian satellites are also shown as individual points.	astro	ph0408119-Cuk.fig11.jpg
0.432715	e453b67df2cb41d1a767f4646abaae02	Same as Fig. 5, only with all satellite clusters and isolated satellites labeled. The groups whose names are followed by an asterisk contain objects with R > 100 km. The clustering of groups around the secular resonance is clearly visible. See text for the full explanation of the abbrevations used in this figure.	astro	ph0408119-Cuk.fig12.jpg
0.582624	2fcdaa9833ab4d81896ffadbd14debcb		astro	ph0408119-Cuk.fig13a.jpg
0.418477	f7bde3e21369474180b74d8c662f1adf	a) The evolution of Sinope's e and its resonant argument Ψ=ϖ-ϖ_J during the numerical integration shown in Fig. 6b, plotted in h=e cosΨ and k=e sinΨ coordinates. The two rings represent two modes of circulation that are characteristic of the regions on either side of the secular resonance, while the librations show as crescent-shaped figures that include portions of each ring. b) Evolution of e over time in the same simulation. Note that the episodes of high-e in this plot are correlated with the instances of retrograde circulation of Ψ in Fig 6b.	astro	ph0408119-Cuk.fig13b.jpg
0.450757	74a6e4aa07ef4240a530f9204a4a927e		astro	ph0408119-Cuk.fig2a.jpg
0.485085	4bd83712c2b045959ef2ba6ff422fe05		astro	ph0408119-Cuk.fig2b.jpg
0.468098	3e445d8b9f5e4be982867ea762819d62	Similar to Fig. 1, for various eccentricities, comparing the motions of both the line of apsides and the line of nodes predicted by secular theories with direct numerical integrations. In all three panels, the precession rate ϖ̇ for the line of apsides predicted by Eq. <ref> is plotted as a solid line; likewise, the predicted precession rate Ω̇ for the line of nodes (Eq. <ref>) is plotted by a dotted line. The same two quantities, predicted on the basis of the Kozai theory (Eqs. 4 and 5) are plotted by dashed and dashed-dotted lines, respectively. All orbits shown have low inclination (i=5^∘) and scaled orbital periods of a) m=0.08 (prograde), b) m=0.08 (retrograde), and c) m=0.16 (retrograde). The planet's period is taken to be that of Jupiter (11.86 yrs).	astro	ph0408119-Cuk.fig2c.jpg
0.456421	171de67bf5a3421d9c5f84ba145f8f9a		astro	ph0408119-Cuk.fig3a.jpg
0.507532	2a51a31fdf874253a132fb3601e34651		astro	ph0408119-Cuk.fig3b.jpg
0.454028	f289101d5b73474f83ca2e516dd956e2	Similar to Fig. 2, but versus inclination, comparing the motion of both the line of apsides and the line of nodes predicted by secular theories with the numerical integrations. The continuous lines and the discrete points have the same meanings as in Figs. 2a-c. All orbits shown are low eccentricity (e_max=0.2) and have scaled orbital periods of a) m=0.08 (prograde), b) m=0.08 (retrograde), and c) m=0.16 (retrograde). The minor irregularities on some of the curves are artifacts of sampling.	astro	ph0408119-Cuk.fig3c.jpg
0.463477	eee55420a40d46ae80b7919833c84253	The critical inclination (as a function of m) at which Kozai resonance occurs for nearly circular orbits, obtained by twenty iterations of Eq. <ref> (solid line). The individual points are the known irregular satellites: Jovians (pluses), Saturnians (crosses), Uranians (asterisks) and Neptunians (boxes). The large triangles mark the known Kozai librators, according to numerical simulation, whose names are also labeled.	astro	ph0408119-Cuk.fig4a.jpg
0.502078	80c1d176db8f40afadb87af184288c90	Critical inclination for the Kozai resonance as determined by numerical advancement of the secular equations of motion, for particles with a maximum eccentricity of 0.2 (solid), 0.4 (long dash) and 0.6 (short dash). The symbols for the known individual satellites have the same meaning as in panel a).	astro	ph0408119-Cuk.fig4b.jpg
0.451029	bb28daa0d81148319c27583d127bcbee	The location of the secular resonance determined by numerical advancement of the secular equations of motion, for particles with a maximum eccentricity of 0.2 (solid), 0.4 (long dash) and 0.6 (short dash). The individual points are the known irregular satellites: Jovians (pluses), Saturnians (crosses), Uranians (asterisks) and Neptunians (boxes). The large triangles mark the objects known to be in, or close to, secular resonance, with their names also labeled.	astro	ph0408119-Cuk.fig5.jpg
0.394787	4d34116a175d4ad2b87e3c2aa5814c89		astro	ph0408119-Cuk.fig6a.jpg
0.449824	ff7c4a3111824b969c5def7ef5e458d0		astro	ph0408119-Cuk.fig6b.jpg
0.457047	8618e1e5ee79451c958072143d364ba8		astro	ph0408119-Cuk.fig6c.jpg
0.410045	36da44d1d80b4625923ac86c3048d18f	Evolution of the resonant argument Ψ=ϖ-ϖ_planet (in radians) during a 300,000-yr numerical integration for four resonant or near-resonant satellites: a) Jupiter's Pasiphae, b) Jupiter's Sinope, c) Saturn's Siarnaq and d) Uranus's Stephano.	astro	ph0408119-Cuk.fig6d.jpg
0.44134	d092f87f7d704d358a9e17476d79403f		astro	ph0408119-Cuk.fig7a.jpg
0.490189	b19512c6b68241e8a3db3f993e5ecb8e	Correlation between the resonant argument Ψ and the parent planet's eccentricity for a) Jupiter's Sinope and b) Saturn's Siarnaq. Note that Sinope's Ψ avoids passing through Ψ=0 when Jupiter's eccentricity is in its upper range, whereas Siarnaq's Ψ passes through Ψ=π only when Saturn's eccentricity is at minimum.	astro	ph0408119-Cuk.fig7b.jpg
0.464134	f821db1cdeea4a58a71b3de7c2351e48	The dependence of S (cos i) as a function of the orbital inclination (defined here as being between 0^∘ and 90^∘). The values of S(cos i) and i for four resonant and near-resonant objects are also labeled. S (cos i) passes through zero at an average inclination of about 41^∘ (139^∘ for retrograde orbits).	astro	ph0408119-Cuk.fig8.jpg
0.447617	8fbf776df4284ffbb10bc4fc2e0ba790		astro	ph0408119-Cuk.fig9a.jpg
0.460164	5410413951fc40b382b315b6382c78e2	a) Evolution of eccentricity over a 300,000-yr numerical integration for a particle orbiting Saturn whose orbit is affected by the Great-Inequality resonance. b) Evolution of the resonant argument ξ=5 λ_S - 2 λ_J - 2 ϖ - ϖ_S during the same integration.	astro	ph0408119-Cuk.fig9b.jpg
0.484636	196a4279eb414b24bd95a13e856be020	The evolution of density in AA and the pressure model (linear approximation and full-order). In the pressure model, the nonlinear effect suppresses the evolution of density fluctuation. Therefore,the linear approximation becomes worse in a strongly nonlinear regime. If γ > 4/3, although we calculate a full-order model, we cannot avoid the formation of the caustic. (a) The case where γ=4/3, the density of the densest point. (b) The case where γ=4/3, the density of the sparsest point. (c) The case where γ=5/3, the density of the densest point. (d) The case where γ=5/3, the density of the sparsest point.	astro	ph0408120-1D-Pressure.jpg
0.425124	60735a68f4f34992826035b4aaa635c7	The evolution of density in linear approximations (ZA, AA, and the pressure model). The viscosity parameter in AA is given by ν=(π/512)^2. (a) The density of the densest point. In AA, the density remains finite forever. On the other hand, in the pressure model, if γ > 4/3, the density becomes infinity. Therefore, we cannot avoid the formation of the caustic. (b) The density of the sparsest point. The growth of the void with the pressure model is fast in comparison with AA.	astro	ph0408120-1D-linear.jpg
0.436408	e361c9bafd434f8cb577138e4a6dc2e4	The evolution of the spherical-symmetric (the Mexican-hat type) density fluctuation at r=0 in dust models. (a) The evolution of a density fluctuation in ZA, PZA, PPZA, the exact model, and AA in the case where ε >0. The viscosity parameter in AA is set as ν = (1/512)^2. The approximation is improved by higher order perturbation. In the exact model, the caustic appears at a ≃ 0.55. On the other hand, the density fluctuation evolves gently in AA. (b) The same as (a) but the case where ε <0. In PZA, the fluctuation becomes positive during evolution. Later (a > 0.6), PPZA deviates from an exact solution greatly more than ZA does.	astro	ph0408120-Mex-dust.jpg
0.425624	c7f6fe3849024c22b4f4921a9ae48db4	The evolution of the spherical-symmetric (Mexican-hat type) density fluctuation at r=0 in the pressure models with the linear approximation. (a) The evolution of a density fluctuation in the case where ε >0. Until a quasi-nonlinear regime develops, the pressure models show behavior resembling AA. However, the fluctuation oscillates in the nonlinear region. In the case where γ=1, the fluctuation oscillates little by little. Finally the fluctuation decays and disappears. In the case where γ=4/3, the fluctuation oscillates and the caustic appears at a ≃ 1.44. In the case where γ=5/3, the oscillation of the fluctuation in the intermediate state grows very large. Then the caustic appears at a ≃ 1.28. (b) The same as (a) but here ε <0. In the case where γ=1, finally the fluctuation decays and disappears. In the case where γ=4/3, although the fluctuation oscillates, the density asymptotically decreases. In the case where γ=5/3, the density fluctuation becomes positive during evolution because the oscillation of the fluctuation grows very large.	astro	ph0408120-Mex-linear.jpg
0.393045	ac509fa1c3b24675a0bd5ddf1d49be90	The evolution of a density fluctuation at r=0 in the spherical-symmetric case. These figures show the evolution in AA and the pressure models (linear approximation and the full-order calculation). These figures show the case where ε >0. (a) In the case where ε >0 and γ=1. In the pressure model, linear approximation seems valid until δ≃ 1. After that, in linear approximation, the fluctuation oscillates violently. In a strongly nonlinear region (δ >10), even if we consider full-order calculation in the pressure model, the evolution of a fluctuation similar to AA cannot be reproduced. (b) The same as (a), but here ε < 0. As in the case where ε >0, when the fluctuation evolve fully (δ <-0.5), the fluctuation begins to oscillate. Finally the fluctuation decays and approaches to 0. (c) The same as (a), but here γ=4/3. In the pressure model with linear approximation, the fluctuation oscillates, and the caustic appears at a ≃ 1.44. When we consider a full-order equation, although we can delay the density divergence, we cannot avoid the formation of the caustic. (d) The same as (b), but here γ=4/3. In the linear approximation in the pressure model, although the fluctuation oscillates, the density asymptotically decreases. In the pressure model, when we consider a full-order calculation, we can realize the evolution of a fluctuation similar to that of AA. (e) The same as (a), but here γ=5/3. In the pressure model with the linear approximation, the oscillation of the fluctuation in the intermediate state grows very large. Then the caustic appears at a ≃ 1.28. When we consider a full-order equation, the density diverges a little to the outside at a ≃ 1.14, and the model fails. (f) The same as (b), but here γ=5/3. In the linear approximation in the pressure model, the density fluctuation becomes positive during evolution because the oscillation of the fluctuation grows very large. On the other hand, when we consider a full-order equation, it is different from linear approximation, the density fluctuation always remaining negative.	astro	ph0408120-MexFull.jpg
0.434431	8515daa41fb34aa78e5230cb6ebba3e0	Mexican-hat type model. The average of density fluctuation over the whole space becomes zero.	astro	ph0408120-Mexican1.jpg
0.489841	a818e8b7219b4d5a978aa32dd74d673b	(a) – R_ max/h_ R versus disc scalelength. The arrows indicate lower limits for galaxies for which no R_ max could be determined. The cross-hatched region shows the prediction of the collapse model for disc galaxy formation (see text). The dashed lines show the predicted threshold radii for star formation <cit.> for disc masses of, from bottom to top, 7.5 × 10^9, 5 × 7.5 10^9 and 25 × 7.5 10^9 . (b) – R_ max/h_ R versus face-on central surface brightness in the I-band. The arrows and the cross-hatched region is as in (a). The dashed line shows the constant star-formation threshold prediction. The shifted dotted line is obtained when allowing for an underestimation of the surface brightnesses by 0.7 mag (see text).	astro	ph0408125-fig1.jpg
0.434268	0fd2b56cfb1641bca1e0d80758b763b4	R_ max/h_ R versus face-on central surface brightness for the V and R-band results of Pohlen for the sharply truncated disc model (). The face-on surface brightnesses were calculated assuming a exponential vertical luminosity distribution. As in Fig. 1b, the cross-hatched region shows the prediction of the collapse model and the dotted line shows the prediction of the star-formation treshold model, with the except that here the (M/L)_⋆ for the threshold model is arbitrary.	astro	ph0408125-fig2.jpg
0.422923	d544df08f6504eb19db69ef1f3aaf94f	Likelihood contour plot in the n0-n'0 plane showing the 1,2 and 3σ contours from the WMAP data (Top Panel), WMAP+SLOAN (Middle Panel) and WMAP+SLOAN+Ly-α (Bottom Panel).	astro	ph0408129-clmo_run_tot.jpg
0.507566	9f076bf8281941c494a8cc0a7ec046cd	Likelihood contour plot in the plane c-(s+c/2) showing the 1,2 and 3σ contours from the WMAP+SLOAN+Ly-α data. We recall that these parameters are related to the physical inflaton potential parameters by c = -β_m(lnϕ_0)/(3 H_I^2) and s+c/2 = m^2(lnϕ_0)/(3 H_I^2).	astro	ph0408129-clmo_suo_lya.jpg
0.4734	660590c131114f54b8022aace9132de7	The theoretically expected region for the parameters s and c for a value of N0 = 50; the solid(red)-line-hatched region is strongly excluded, while the dashed(green)-line-hatched region is excluded only if the mass is supposed to run up to the end of inflation. The dotted (blue) line gives the prediction for the case when the end of inflation is triggered by η = 1. The circles correspond to the values in the explicit models discussed in section II.C: the upper ones (magenta) refer to the case of gauge coupling dominance, while the (blue) one at the origin to the case of Yukawa coupling dominance.	astro	ph0408129-clmo_th_sc.jpg
0.476068	ff3e7b07dc404e78abb3fb5cdebffb6a	Likelihood contour plot in the c-s plane showing the 1,2 and 3σ contours from the WMAP data (Top Panel), WMAP+SLOAN (Middle Panel) and WMAP+SLOAN+Ly-α (Bottom Panel).	astro	ph0408129-clmo_tot.jpg
0.474887	1bd535dc084c432ca1b45f0838ddb3cc	Four spectra of SN 2002cx, from L03, are compared. Epochs are in days with respect to the date of maximum brightness in the B band, 2002 May 21. The flux is per unit wavelength interval and the vertical displacement is arbitrary. Spectra have been corrected for the redshift of the parent galaxy, cz = 7184 . No correction for interstellar reddening has been applied.	astro	ph0408130-f1.jpg
0.402912	a14f8492553c40c1aa17050dc71e2983	The day -4 and day -1 spectra of SN 2002cx, and the day -4 spectrum of SN 1997br, are compared. The SN 1997br spectrum has been redshifted by 3000 to align the Fe III absorptions. The flux is per unit frequency interval (so that the spectra are roughly flat rather than steeply sloped) and the vertical displacement is arbitrary.	astro	ph0408130-f2.jpg
0.3937	a05c230d78fb41fdb1eac6923d0731d4	The day -1 spectrum of SN 2002cx (solid line) is compared with a synthetic spectrum (dashed line) that has v_phot=7000 , T_bb=14,000 K, and contains lines of six ions. The flux is per unit frequency interval. In this and subsequent figures the flux is linear rather than logarithmic.	astro	ph0408130-f3.jpg
0.446351	8040b7fd975b4550b939c50235d8089c	The day -4 spectrum of SN 1997br (solid line) is compared with a synthetic spectrum (dashed line) like that of Figure 3, for SN 2002cx, except that has been increased from 7000 to 12,000 .	astro	ph0408130-f4.jpg
0.403379	e94035517f0441fba21e8965f3df25a3	The day 12 spectrum of SN 2002cx (solid line) is compared with a synthetic spectrum (dashed line) that has v_phot=7000 , T_bb=9000 K, and contains lines of five ions. The flux is per unit wavelength interval.	astro	ph0408130-f5.jpg
0.48886	8e5944852fcc4ad69852f29d2879cc7f	The day 25 spectrum of SN 2002cx (solid line) is compared with a synthetic spectrum (dashed line) that has v_phot=5000 , T_bb=5500 K, and contains lines of five ions. The flux is per unit wavelength interval.	astro	ph0408130-f6.jpg
0.438254	5c114aef034341d8901979ddd24d0672	The day 56 spectrum of SN 2002cx (solid line) is compared with a synthetic spectrum (dashed line) that has v_phot=2000 , T_bb=7000 K, and contains lines of five ions. The flux is per unit wavelength interval.	astro	ph0408130-f7.jpg
0.421693	243c45ff6de04d42adeb91819ef2f87d	The spatial distribution (equatorial coordinates, Aitoff projection) of the 914 HRI pointings used for the BMW–HRI cluster survey. The origin of right ascension is in the center of the plot and the grid is in steps of 3^h in RA and 30^∘ in DEC. The empty area corresponds to the \|b_II\| < 20^∘ zone of avoidance, while the boxes mark the Magellanic clouds and Virgo cluster areas which were also excluded from the survey. Different symbols indicate different ranges of exposure time, as explained.	astro	ph0408131-1326f1.jpg
0.432291	bf9873699348488392d85263f2e54d60	The expected number of spurious detections in the BMW–HRI cluster catalogue as a function of the source significance, expressed in units of standard deviations σ (see text for details). The curve includes the contamination produced both by background fluctuations detected as true (extended) sources and by real point sources mistaken as extended by the algorithm. On the basis of this curve, the BMW–HRI p_s>4 master sample is expected to have a 27% contamination, while selecting at p_s>5 one obtains a sample with virtually null contamination.	astro	ph0408131-1326f11.jpg
0.555489	97e5ad1507994e6b9dc4e8899a04311a	The solid angle covered by the BMW–HRI survey (sky–coverage) as a function of flux and source extension, obtained by integrating the survey selection function over the observed fields.	astro	ph0408131-1326f12.jpg
0.551474	1ceccb2c6a764a11b481969d5b9c4c6e	Comparison of the BMW–HRI size-weighted sky coverage to those of the EMSS (Gioia et al. 1990), 160 Square Degrees (Vikhlinin et al. 1998a) and RDCS (Rosati et al. 1998). Since the sky coverage is a function of both flux and source extension, the BMW–HRI curve has been computed combining the sky coverages pertaining to different source extensions, according to the observed distribution of angular sizes in the p_s>4 sample.	astro	ph0408131-1326f13.jpg
0.40897	373b082f135f455ab88c9803f908e9d1	The classical Log N – Log S plot — i.e. the mean cumulative surface density of clusters as a function of X–ray flux — from the BMW–HRI survey (filled circles), compared to the RDCS (Rosati et al. 1998) and 160SD survey (Vikhlinin et al. 1998b). The BMW–HRI points are computed from the high-significance p_s>5 σ sub-sample, for which the expected contamination is negligible. The curves give the expected counts in the same Λ-cosmology adopted in this paper, for the cases of no-evolution and evolution of the XLF. The BMW-HRI and PSPC data are all consistent with a negligible to moderate evolution of the XLF.	astro	ph0408131-1326f15.jpg
0.466731	7ed41e9b890043b0a0f692c93fcacf2f	Optical/X–ray overlays for a selection of new BMW–HRI groups/clusters at different redshifts. From top left to bottom right: BMW134654.9-301328, z_spec=0.358; BMW212415.7-334754, z_spec=0.92; BMW122842.6-391612, z_phot≃ 0.8; BMW112059.0+130450, z_spec=0.615. The redshift for BMW212415.7-334754, a very compact cluster, awaits further confirmation, being based on only 3 redshifts and displaying a second system at z=0.62. The X–ray images have been smoothed with a Gaussian filter (σ = 40) and the isophotes correspond to 0.7,1,2,3 standard deviations over the background (after smoothing). Optical images are combined Gunn-r+i ∼ 60min total exposures with EFOSC2 at the ESO 3.6 m telescope. For all but the first object the sides of the figures are 2.5.	astro	ph0408131-1326f16.jpg
0.468062	3f79fc3b2b1444a38c9407dd2d21acf5	Distribution of exposure times for the 914 pointings used for the cluster survey.	astro	ph0408131-1326f2.jpg
0.448271	c34d8cfe803e41e98dfe9a0af40b05c2	Source extensions as measured by the BMW detection algorithm (W_clu, see text), plotted against the off-axis angle, showing our empirical definition of extended sources. The continuous line gives the mean extension of a point–like source at different off-axis angles, and the dashed lines identifies the 3 σ locus. Sources selected as cluster candidates have to lie, with their error bars, above the dashed line and are indicated by filled circles.	astro	ph0408131-1326f3.jpg
0.464011	033aec91140c41568602e4e512d1560e	Left panel: 24× 24 portion of an HRI image (Gaussian smoothed with σ=5) showing the detection of cluster candidate BMW044743.4-202042, marked by the solid 90 radius circle. The dashed circle (3 radius) indicates the central area containing the original HRI target (here barely visible), excluded from the cluster search. Right panel: Integral, background-subtracted flux growth curve of the candidate cluster (filled circles), plotted together with its 1σ error corridor (dotted lines) and the best-fitting β-model (Grey dashed curve). The total flux of the source is computed as the asymptotic value of the growth curve (horizontal dashed line). From the total flux, the core radius is then derived by comparison to a grid of β profiles convolved with the instrumental PSF at the specific off-axis angle. This is a typical source in our catalog, detected at ∼ 7 off–axis with 75 net counts in a 10 ks exposure, corresponding to a flux of ∼ 3×10^-13 erg s^-1 cm^-2.	astro	ph0408131-1326f4.jpg
0.494937	7d06684b8c5f478dbb1da4d822668eac	Comparison of the global pixel statistics of a real (black line) and a simulated image (Grey line), for a typical exposure time of 20,000 sec. X–ray photons have been binned in pixels of 5 size. The histograms show how the background statistical properties of the simulated image accurately match those of the real one. The simulated background is built as a Poissonian realization of the background map, and the differences that can be noticed for pixels with more than five counts are due to the effect of the simulated (extended) sources, which are overabundant in the simulated image.	astro	ph0408131-1326f8.jpg
0.469463	b56cfa08b06c4fa3837afb2bab0fc7b4	Top panel: visual comparison of the HRI image of BMW000141.3-154042 (left), the cluster detected with the largest number of net counts in the BMW–HRI survey, with a simulated X–ray source with the same counts and core radius. The bottom panel plots the average of the central 10 columns of the two images, with the real cluster described by the black line.	astro	ph0408131-1326f9.jpg
0.473244	86dd93a13e874e51bd4e801dca98e8a5	A comparison of the H41α (solid line) and H92α (dotted line, multiplied by a factor of 15) spectrum, one at the Sgr A^⋆ position (right) and at an adjacent position 7 to the east (left).	astro	ph0408136-f10.jpg
0.410014	b3c2ce99e33f42b1ab8076b8a6d5babf	Filter transmission curves for this survey. Solid line: curve for KP1590, used for Fields 1 and 2. Dashed line: curve for KP1467, used for Field 3. Both curves have been shifted 13 Åblueward to account for the f/3.5 beam of the Schmidt and the observing temperature. The histogram is a Gaussian distribution with mean –300 and σ = 200 km/s representing the expected M31 halo velocity distribution. The vertical lines show the location of velocities ± 400 km/s from M31's systemic velocity.	astro	ph0408137-f1.jpg
0.516722	94966ecf28e4480d8da6dc8ee4ea2586	The distribution of PN along the minor axis. PN in the complete subsample within 3 kpc of the minor axis are assigned an isophotal distance and binned into 2 kpc bins. The solid line is the number of PN expected based on the minor axis surface brightness profile from , normalized to the depth of our survey. The first three bins are reduced by our exclusion of PN near the center of the galaxy.	astro	ph0408137-f10.jpg
0.465167	0641e2dc721940d89886f0504e43e2ed	Upper panel: One realization of a bulge model with = 80 km/s and = 100 km/s. Lower panel: The probability distribution for the mean and for measured from the positions at \|Y\|>6 and \|X\|>10 for 1000 realizations of the same bulge model as in the upper panel. The solid lines represent the values measured from the data, = 116 km/s and = 79 km/s. All velocities are in km/s.	astro	ph0408137-f11.jpg
0.484035	9476f898ed424bc7bf417d256ed4fca3	The probability distribution for measured from the positions at \|Y\|>6 and \|X\|>10 for 1000 realizations of various bulge models. The solid lines represent the value from the data, = 79 km/s, and the dashed lines are the 10% confidence limits from the models. Upper panel: a model bulge with = 160 km/s and = 90 km/s. Middle panel: a model bulge with = 160 km/s and = 70 km/s. Lower panel: a model bulge with = 160 km/s and = 50 km/s.	astro	ph0408137-f12.jpg
0.466302	b331f0a74d004ab4b49cb4f3c659329e	The probability distribution for the mean and for measured from the positions at \|Y\|>6 and \|X\|>10 for 1000 realizations of various bulge models. The solid lines represent the values from the data, = 116 km/s and = 79 km/s, and the dashed lines are the 10% confidence limits from the models. Upper panel: a model bulge with = 170 km/s and = 70 km/s. Middle panel: a model bulge with = 150 km/s and = 70 km/s. Lower panel: a model bulge with = 130 km/s and = 70 km/s.	astro	ph0408137-f13.jpg
0.446898	e61486fe8d034b928d54e1e98bc32d01	Figure 2 from , showing M31's halo and tidal streams, with the positions of our potential halo objects marked (red in the electronic edition). All 5 fall on or near possible tidal streams, and three are counter-rotating with respect to the disk at their position. The two halo candidates near the center have velocities within 40 km/s of each other, and have velocities relative to M31 which are at least 3 times the central bulge dispersion. Orbits suggested by <cit.> and <cit.> for the Southern Stream <cit.> project near the position of these PN.	astro	ph0408137-f14.jpg
0.431194	48b3edc125784f1996a59b8d718d8cb6	Positions of our Schmidt fields and PN in M31. The R-band image was kindly provided by Rene Walterbos. North is up, East is left. M31's position angle is 37.3 <cit.>. The SW end of the galaxy is approaching, and the far side is to the SE of the major axis. Squares: PN in M31. Pluses: PN in M32. Xs: HII regions in And IV.	astro	ph0408137-f2.jpg
0.493979	5e9e9aba9c2f41fa9518851cccc16d4f	The estimated PN luminosity function for Fields 1 (panel a) and 3 (panel b). The X axis is the instrumental distribution shifted so that the bright edge of the distribution is at m_5007 = –4.5.	astro	ph0408137-f3.jpg
0.405172	b24f7c58d0e6428f93a1b4fa85ce32e6	Three typical spectra for our sample, with a range of signal-to-noise, showing the strongest PN emission lines, [OIII]and , and Hα.	astro	ph0408137-f4.jpg
0.448636	3f92597f229840bb877dbab1e7ed5a4c	Plot showing the positions of PN in M31. The major axis has been rotated through the position angle so that it is horizontal in the figure. Symbol size corresponds to velocity. Circles are PN velocities from this study, and squares are those from . Ellipses are bulge isophotes with b/a=0.55.	astro	ph0408137-f5.jpg

0.419077

cd3a3c279d3644648d4666a53e0b3d18

The same as Fig.<ref>, but for Q0405-123.

astro

ph0408105-Q0405-123.jpg

0.45764

d1d9b61cc275406eb210a763d04fe039

The same as Fig.<ref>, but for Q0405-123. We calculated and plotted the line polarization of Hβ+[OIII] region (4731-5032Å) instead of Hβ only, since excluding the telluric absorption wavelengths at around 7600Å leaves a very small wavelength window for Hβ line.

astro

ph0408105-Q0405-123_qu.jpg

0.402824

f407a3e66a31476d8d4df3e8c739750b

The same as Fig.<ref>, but for Q0414-060.

astro

ph0408105-Q0414-060.jpg

0.423725

0ef494fd0aef4411a2b7afc7415c4aa8

The same as Fig.<ref>, but for the Keck observation of Q1004+130.

astro

ph0408105-Q1004+130.jpg

0.558853

66faefbc2910491b8f3666b5992aa76b

The same as Fig.<ref>, but for Q1004+130.

astro

ph0408105-Q1004+130_qu.jpg

0.458486

464a75a934a14380b632fc7707f79e5c

The same as Fig.<ref>, but for Q1912-550.

astro

ph0408105-Q1912-550.jpg

0.423323

c53ed054b36c4d05954868e86037494d

The ISP-corrected polarization of Q1912-550.

astro

ph0408105-Q1912-550_cor010.jpg

0.457019

3431fa9f8ba24f9bb0ea5559dd5846aa

The same as Fig.<ref>, but for Q1912-550.

astro

ph0408105-Q1912-550_qu.jpg

0.370544

ad652668fe8c446f83e957e49abf4a2b

The same as Fig.<ref>, but for Q2115-305. The reference axis of q' and u' is taken along the PA of the polarization integrated over the whole wavelength region, rather than at 4000-4731Å.

astro

ph0408105-Q2115-305.jpg

0.426178

5b086ae59f03456c980b29da443ef65e

The same as Fig.<ref>, but for Q2251+113.

astro

ph0408105-Q2251+113.jpg

0.407383

7b244e5883204d82b83b8cb56ceb79a3

The same as Fig.<ref>, but for Q2349-014.

astro

ph0408105-Q2349-014.jpg

0.453972

fb186e916797477c8115538dec951e51

The same as Fig.<ref>, but for the Keck observation of Ton202.

astro

ph0408105-Ton202_May2003.jpg

0.397882

9c04234e7be644d6b1636e3720bb97de

Comparison of the observations of Ton202 at two epochs. The upper panel show the polarization observed in May 2003 minus that in May 2002. The lower panel show the total flux in May 2003 divided by that in May 2002.

astro

ph0408105-Ton202_dif.jpg

0.471387

ae90fcd98d774ca393ab977768aa3f00

As an illustration, a spectrum (thick smooth line) for a face-on disk model around a Kerr (a/M=0.998) black hole of mass 8×10^9 M_⊙, accreting at 2 M_⊙ yr^-1 is plotted over the May 2002 polarized flux spectrum of Ton202 (corrected for the Galactic reddening). The model spectrum has arbitrary normalization.

astro

ph0408105-Ton202_model.jpg

0.449296

82d84b4808cf47e7ab0b6d1c855c7f76

In the upper panel, the five polarized flux spectra shown in Fig.<ref>, normalized by the mean at 2800-5000Å, are over-plotted with a larger wavelength bin of 20Å. In the lower panel the average of these five spectra with equal weighting is shown. Both panels are in F_λ.

astro

ph0408105-coadd_mobj.jpg

0.428631

14e7de4749434b048f87272a8b661086

The same as Fig.<ref>, but in ν F_ν with both axes in log scale.

astro

ph0408105-coadd_mobj_nFn.jpg

0.4158

f1185b6092354a16818d97681de64d4a

The spectral index α (f_ν∝ν^α) corresponding to the flux ratio at 3050Å and 4200Å is plotted against ΔPA.

astro

ph0408105-dpa-alpha_sbb.jpg

0.470417

d6ed3fb99ab44afb83abfe9834d3202e

The ratio of the polarization at 2891-3600Å to that at 4000-4731Å is plotted against ΔPA.

astro

ph0408105-dpa-pratio.jpg

0.506688

9516f2dad0ce4b44bafd0136b855aab0

The histogram for the difference of the radio axis PA and optical polarization PA.

astro

ph0408105-hist.jpg

0.492364

50d69f17ce324d899e3cc5e6cb6d5fd8

The profiles of the equilibrium quantities plotted against dimensionless x/L_x coordinate: density (panel A), pressure (panel B), the magnetic field (panel C) and plasma beta (panel D)

astro

ph0408114-figure2.jpg

0.457113

60e3a41a1ad24719871b8f070ea62f81

Sample solutions of the standing fast magnetosonic modes. Panel A: zeroth-order harmonic n=0, period T=5.6069 min., α=0.015 ; Panel B: second-order harmonic n=2, period T=1.8514 min., α=0.03

astro

ph0408114-figure3.jpg

0.442963

6f757d6de32842508faff0b35e95f7c2

Results of numerical simulations for the standing fast magnetosonic mode, shown on panel A in Figure <ref>. Panel A_1: the curves of F_1 (Eq. (<ref>)) (solid line) and F_2 (Eq. (<ref>)) (dotted line) vs. fractional distance x/L_x for the zeroth-order standing fast mode n=0 and the standing Alfvén mode with m=3; Panel B_1: The numerical solution of the Alfvén mode m=3 amplitude at location x/L_x≈ x_0,4/L_x=0.0767; Panel C_1: The numerical solution for mode m=3 at x/L_x≈ x_0,3/L_x=0.7967; Panel A_2: As in panel A_1 for mode m=4; Panel B_2: As in panel B_1 for mode m=4; Panel C_2: As in panel C_1 for mode m=4.

astro

ph0408114-figure4.jpg

0.435974

feb80409654a414a9bb850a42c30beaf

As in Figure <ref> for the second-order harmonic of the fast magnetosonic mode shown on Panel B in Figure <ref> (n=2) and the standing Alfvén with m=8 and m=10. The numerical solutions respectively are given at distances x_1/L_x ≈ x_2,10/L_x = 0.6000 (panels B_1 and B_2), x_2/L_x ≈ x_2,8/L_x=0.9933 (panels C_1 and C_2). Dashed line in panel A_1 is curve of F_1 (<ref>) corresponding to the mode m=9.

astro

ph0408114-figure5.jpg

0.469018

101690c961d84b0ca859b141c6b4c16a

As in Figure <ref> for the standing Alfvén modes with m=11 and m=12 at respective resonant points x_1/L_x ≈ x_2,12/L_x = 0.1033 (panels B_1 and B_2), x_2/L_x ≈ x_2,11/L_x=0.3633 (panels C_1 and C_2).

astro

ph0408114-figure6.jpg

0.421282

cbbffa93e4924e358a821e4b4751c8df

Histogram of the rank correlation coefficients calculated from 10^5 random samples drawn from the observed stellar metallicity and orbital period distributions. The observed values discussed in Section 2 (arrows of different styles) have between 6.8% and 0.3% chances of being consistent with no correlation

astro

ph0408116-fig2.jpg

0.470255

f45835c5a2e04578a3bd50f61f3ceb1c

The lightcurve brightness distribution for the simulated dataset.

astro

ph0408118-LCMagDistNonCm.jpg

0.463042

59ab71c0d7134b31b69893aea1b34063

The final iteration of the ROTSE data map (shown in Fig. <ref>) when processed with the U-Matrix algorithm. The colour scale runs from dark blue (similar reference vectors) to red (dissimilar reference vectors).

astro

ph0408118-ROTSENetUMat34.jpg

0.532738

76a9bf01bcbb4e82ba93f22abfe62dc2

A map of the number of training set members assigned to each neuron for the network trained with the ROTSE data. The raw map has been filtered with a 3x3 top-hat kernel. The colour scale runs from black (low occupancy) to red (high occupancy).

astro

ph0408118-SmoothedDistCOL34.jpg

0.478964

15b98d5a0f534616b91613ce6b8d0f43

The average Euclidean distance moved by a training set member in map space from one iteration to the next, . Panel order is the same as Figure <ref>.

astro

ph0408118-dbar2.jpg

0.446258

158732597c534a479089a27d170b13bb

Samples of the four classes of simulated lightcurves, representing type-c RR Lyrae, δ Scuti and Cepheid types (top left), type-ab RR Lyrae (top right), short-period (bottom left) and long-period eclipsing binary (bottom right)

astro

ph0408118-fakelcs.jpg

0.458682

fdc1570674ff40a4b804a70974ea43b3

Evolution of a network trained with our synthetic data and with α(t)=α_0=0.01. Panels show (from left to right, top to bottom), iterations 1, 4, 7, 10, 13, 16, 19, 22, 25, 28, 31, 34. The four classes of synthetic lightcurve (described in Section <ref>) are denoted by (i) squares, (ii) crosses, (iii) diamonds and (iv) stars. Note that, for the purposes of visualisation, the positions of all lightcurves have been randomised within their host neurons.

astro

ph0408118-mapevol2.jpg

0.440647

6a57dcb52ee844fba53cf0fd8a9618ae

The number of training set members that move from one iteration to the next, . Panel order is the same as Figure <ref>.

astro

ph0408118-nmove3.jpg

0.500251

eb0a15f60d61433e82955482957b739c

The final structure of a network trained using 1206 pre-classified ROTSE lightcurves.

astro

ph0408118-rotse_map3.jpg

0.483808

5a60991d545e449b9b7ebbace26792b8

The ROTSE map of Fig. <ref> collapsed in one dimension and expanded with the period of the variable star. Here the δ Scuti, Cepheid and type-c RR Lyrae variables, which formed an overlapping group in Fig. <ref>, have been separated in period. This illustrates how additional information can be used to resolve ambiguity in our maps, which are based on shape alone.

astro

ph0408118-rotse_map3_period.jpg

0.424123

7b8bc887f9af4e6a875df7f21558ac69

The average “goodness of fit”, . In the leftmost column the learning parameter α is held constant throughout the learning phase. In the central column α falls linearly, and in the rightmost column exponentially. The initial value of α assumes the value 0.9 (top row), 0.5 (second row), 0.1 (third row) and 0.01 (bottom row).

astro

ph0408118-sbar2.jpg

0.449691

a4555a57c80a4ec9b71e126154bd27aa

The dependence of the apsidal precession rate ϖ̇ on satellite's eccentricity, for a near-planar prograde satellite (m=0.08, i=5^∘) moving about a low-eccentricity pseudo-Jupiter (P'=11.86 yr, but e'=0.01). Crosses plot the result of a symplectic integration, while lines plot the predictions of secular theories. The precession rates for the latter were obtained by advancing the secular equations of motion using a Burlisch-Stoer numerical integrator. The short-dashed line plots the predictions of Kozai's theory, the long-dashed curve also includes the evection terms, while the solid line depicts the results of the mixed model derived in this paper (Eqs. <ref> and <ref>).

astro

ph0408119-Cuk.fig1.jpg

0.389291

ec5e7699cfd540899399207d43a877c9

The dependence of eccentricity variation parameter η (defined in Eq. <ref>) on the precession rate of a particle's ϖ for 440 test particles integrated numerically over 30,000 yrs (see text for details). Secular and the Great Inequality resonances are prominent as regions of high η; the onset of Kozai resonance is also visible, with all the particles to the left of the discontinuity being Kozai librators.

astro

ph0408119-Cuk.fig10.jpg

0.5318

f6d6922839094caeb99a844c344074b1

The position of the Great-Inequality resonance in m-i_min space. The top three continuous lines plot (in descending order) the Great Inequality resonance for maximum eccentricities of 0.2, 0.4 and 0.6, respectively. The fourth line plots the position of the Great Inequality resonance for e_max=0.6, in a planetary configuration that has 5 λ_S - 2 λ_J=0.002 yr^-1 (i.e., the Great Inequality has a period of 500 yrs). The orbital elements of the known prograde Saturnian satellites are also shown as individual points.

astro

ph0408119-Cuk.fig11.jpg

0.432715

e453b67df2cb41d1a767f4646abaae02

Same as Fig. 5, only with all satellite clusters and isolated satellites labeled. The groups whose names are followed by an asterisk contain objects with R > 100 km. The clustering of groups around the secular resonance is clearly visible. See text for the full explanation of the abbrevations used in this figure.

astro

ph0408119-Cuk.fig12.jpg

0.582624

2fcdaa9833ab4d81896ffadbd14debcb

astro

ph0408119-Cuk.fig13a.jpg

0.418477

f7bde3e21369474180b74d8c662f1adf

a) The evolution of Sinope's e and its resonant argument Ψ=ϖ-ϖ_J during the numerical integration shown in Fig. 6b, plotted in h=e cosΨ and k=e sinΨ coordinates. The two rings represent two modes of circulation that are characteristic of the regions on either side of the secular resonance, while the librations show as crescent-shaped figures that include portions of each ring. b) Evolution of e over time in the same simulation. Note that the episodes of high-e in this plot are correlated with the instances of retrograde circulation of Ψ in Fig 6b.

astro

ph0408119-Cuk.fig13b.jpg

0.450757

74a6e4aa07ef4240a530f9204a4a927e

astro

ph0408119-Cuk.fig2a.jpg

0.485085

4bd83712c2b045959ef2ba6ff422fe05

astro

ph0408119-Cuk.fig2b.jpg

0.468098

3e445d8b9f5e4be982867ea762819d62

Similar to Fig. 1, for various eccentricities, comparing the motions of both the line of apsides and the line of nodes predicted by secular theories with direct numerical integrations. In all three panels, the precession rate ϖ̇ for the line of apsides predicted by Eq. <ref> is plotted as a solid line; likewise, the predicted precession rate Ω̇ for the line of nodes (Eq. <ref>) is plotted by a dotted line. The same two quantities, predicted on the basis of the Kozai theory (Eqs. 4 and 5) are plotted by dashed and dashed-dotted lines, respectively. All orbits shown have low inclination (i=5^∘) and scaled orbital periods of a) m=0.08 (prograde), b) m=0.08 (retrograde), and c) m=0.16 (retrograde). The planet's period is taken to be that of Jupiter (11.86 yrs).

astro

ph0408119-Cuk.fig2c.jpg

0.456421

171de67bf5a3421d9c5f84ba145f8f9a

astro

ph0408119-Cuk.fig3a.jpg

0.507532

2a51a31fdf874253a132fb3601e34651

astro

ph0408119-Cuk.fig3b.jpg

0.454028

f289101d5b73474f83ca2e516dd956e2

Similar to Fig. 2, but versus inclination, comparing the motion of both the line of apsides and the line of nodes predicted by secular theories with the numerical integrations. The continuous lines and the discrete points have the same meanings as in Figs. 2a-c. All orbits shown are low eccentricity (e_max=0.2) and have scaled orbital periods of a) m=0.08 (prograde), b) m=0.08 (retrograde), and c) m=0.16 (retrograde). The minor irregularities on some of the curves are artifacts of sampling.

astro

ph0408119-Cuk.fig3c.jpg

0.463477

eee55420a40d46ae80b7919833c84253

The critical inclination (as a function of m) at which Kozai resonance occurs for nearly circular orbits, obtained by twenty iterations of Eq. <ref> (solid line). The individual points are the known irregular satellites: Jovians (pluses), Saturnians (crosses), Uranians (asterisks) and Neptunians (boxes). The large triangles mark the known Kozai librators, according to numerical simulation, whose names are also labeled.

astro

ph0408119-Cuk.fig4a.jpg

0.502078

80c1d176db8f40afadb87af184288c90

Critical inclination for the Kozai resonance as determined by numerical advancement of the secular equations of motion, for particles with a maximum eccentricity of 0.2 (solid), 0.4 (long dash) and 0.6 (short dash). The symbols for the known individual satellites have the same meaning as in panel a).

astro

ph0408119-Cuk.fig4b.jpg

0.451029

bb28daa0d81148319c27583d127bcbee

The location of the secular resonance determined by numerical advancement of the secular equations of motion, for particles with a maximum eccentricity of 0.2 (solid), 0.4 (long dash) and 0.6 (short dash). The individual points are the known irregular satellites: Jovians (pluses), Saturnians (crosses), Uranians (asterisks) and Neptunians (boxes). The large triangles mark the objects known to be in, or close to, secular resonance, with their names also labeled.

astro

ph0408119-Cuk.fig5.jpg

0.394787

4d34116a175d4ad2b87e3c2aa5814c89

astro

ph0408119-Cuk.fig6a.jpg

0.449824

ff7c4a3111824b969c5def7ef5e458d0

astro

ph0408119-Cuk.fig6b.jpg

0.457047

8618e1e5ee79451c958072143d364ba8

astro

ph0408119-Cuk.fig6c.jpg

0.410045

36da44d1d80b4625923ac86c3048d18f

Evolution of the resonant argument Ψ=ϖ-ϖ_planet (in radians) during a 300,000-yr numerical integration for four resonant or near-resonant satellites: a) Jupiter's Pasiphae, b) Jupiter's Sinope, c) Saturn's Siarnaq and d) Uranus's Stephano.

astro

ph0408119-Cuk.fig6d.jpg

0.44134

d092f87f7d704d358a9e17476d79403f

astro

ph0408119-Cuk.fig7a.jpg

0.490189

b19512c6b68241e8a3db3f993e5ecb8e

Correlation between the resonant argument Ψ and the parent planet's eccentricity for a) Jupiter's Sinope and b) Saturn's Siarnaq. Note that Sinope's Ψ avoids passing through Ψ=0 when Jupiter's eccentricity is in its upper range, whereas Siarnaq's Ψ passes through Ψ=π only when Saturn's eccentricity is at minimum.

astro

ph0408119-Cuk.fig7b.jpg

0.464134

f821db1cdeea4a58a71b3de7c2351e48

The dependence of S (cos i) as a function of the orbital inclination (defined here as being between 0^∘ and 90^∘). The values of S(cos i) and i for four resonant and near-resonant objects are also labeled. S (cos i) passes through zero at an average inclination of about 41^∘ (139^∘ for retrograde orbits).

astro

ph0408119-Cuk.fig8.jpg

0.447617

8fbf776df4284ffbb10bc4fc2e0ba790

astro

ph0408119-Cuk.fig9a.jpg

0.460164

5410413951fc40b382b315b6382c78e2

a) Evolution of eccentricity over a 300,000-yr numerical integration for a particle orbiting Saturn whose orbit is affected by the Great-Inequality resonance. b) Evolution of the resonant argument ξ=5 λ_S - 2 λ_J - 2 ϖ - ϖ_S during the same integration.

astro

ph0408119-Cuk.fig9b.jpg

0.484636

196a4279eb414b24bd95a13e856be020

The evolution of density in AA and the pressure model (linear approximation and full-order). In the pressure model, the nonlinear effect suppresses the evolution of density fluctuation. Therefore,the linear approximation becomes worse in a strongly nonlinear regime. If γ > 4/3, although we calculate a full-order model, we cannot avoid the formation of the caustic. (a) The case where γ=4/3, the density of the densest point. (b) The case where γ=4/3, the density of the sparsest point. (c) The case where γ=5/3, the density of the densest point. (d) The case where γ=5/3, the density of the sparsest point.

astro

ph0408120-1D-Pressure.jpg

0.425124

60735a68f4f34992826035b4aaa635c7

The evolution of density in linear approximations (ZA, AA, and the pressure model). The viscosity parameter in AA is given by ν=(π/512)^2. (a) The density of the densest point. In AA, the density remains finite forever. On the other hand, in the pressure model, if γ > 4/3, the density becomes infinity. Therefore, we cannot avoid the formation of the caustic. (b) The density of the sparsest point. The growth of the void with the pressure model is fast in comparison with AA.

astro

ph0408120-1D-linear.jpg

0.436408

e361c9bafd434f8cb577138e4a6dc2e4

The evolution of the spherical-symmetric (the Mexican-hat type) density fluctuation at r=0 in dust models. (a) The evolution of a density fluctuation in ZA, PZA, PPZA, the exact model, and AA in the case where ε >0. The viscosity parameter in AA is set as ν = (1/512)^2. The approximation is improved by higher order perturbation. In the exact model, the caustic appears at a ≃ 0.55. On the other hand, the density fluctuation evolves gently in AA. (b) The same as (a) but the case where ε <0. In PZA, the fluctuation becomes positive during evolution. Later (a > 0.6), PPZA deviates from an exact solution greatly more than ZA does.

astro

ph0408120-Mex-dust.jpg

0.425624

c7f6fe3849024c22b4f4921a9ae48db4

The evolution of the spherical-symmetric (Mexican-hat type) density fluctuation at r=0 in the pressure models with the linear approximation. (a) The evolution of a density fluctuation in the case where ε >0. Until a quasi-nonlinear regime develops, the pressure models show behavior resembling AA. However, the fluctuation oscillates in the nonlinear region. In the case where γ=1, the fluctuation oscillates little by little. Finally the fluctuation decays and disappears. In the case where γ=4/3, the fluctuation oscillates and the caustic appears at a ≃ 1.44. In the case where γ=5/3, the oscillation of the fluctuation in the intermediate state grows very large. Then the caustic appears at a ≃ 1.28. (b) The same as (a) but here ε <0. In the case where γ=1, finally the fluctuation decays and disappears. In the case where γ=4/3, although the fluctuation oscillates, the density asymptotically decreases. In the case where γ=5/3, the density fluctuation becomes positive during evolution because the oscillation of the fluctuation grows very large.

astro

ph0408120-Mex-linear.jpg

0.393045

ac509fa1c3b24675a0bd5ddf1d49be90

The evolution of a density fluctuation at r=0 in the spherical-symmetric case. These figures show the evolution in AA and the pressure models (linear approximation and the full-order calculation). These figures show the case where ε >0. (a) In the case where ε >0 and γ=1. In the pressure model, linear approximation seems valid until δ≃ 1. After that, in linear approximation, the fluctuation oscillates violently. In a strongly nonlinear region (δ >10), even if we consider full-order calculation in the pressure model, the evolution of a fluctuation similar to AA cannot be reproduced. (b) The same as (a), but here ε < 0. As in the case where ε >0, when the fluctuation evolve fully (δ <-0.5), the fluctuation begins to oscillate. Finally the fluctuation decays and approaches to 0. (c) The same as (a), but here γ=4/3. In the pressure model with linear approximation, the fluctuation oscillates, and the caustic appears at a ≃ 1.44. When we consider a full-order equation, although we can delay the density divergence, we cannot avoid the formation of the caustic. (d) The same as (b), but here γ=4/3. In the linear approximation in the pressure model, although the fluctuation oscillates, the density asymptotically decreases. In the pressure model, when we consider a full-order calculation, we can realize the evolution of a fluctuation similar to that of AA. (e) The same as (a), but here γ=5/3. In the pressure model with the linear approximation, the oscillation of the fluctuation in the intermediate state grows very large. Then the caustic appears at a ≃ 1.28. When we consider a full-order equation, the density diverges a little to the outside at a ≃ 1.14, and the model fails. (f) The same as (b), but here γ=5/3. In the linear approximation in the pressure model, the density fluctuation becomes positive during evolution because the oscillation of the fluctuation grows very large. On the other hand, when we consider a full-order equation, it is different from linear approximation, the density fluctuation always remaining negative.

astro

ph0408120-MexFull.jpg

0.434431

8515daa41fb34aa78e5230cb6ebba3e0

Mexican-hat type model. The average of density fluctuation over the whole space becomes zero.

astro

ph0408120-Mexican1.jpg

0.489841

a818e8b7219b4d5a978aa32dd74d673b

(a) – R_ max/h_ R versus disc scalelength. The arrows indicate lower limits for galaxies for which no R_ max could be determined. The cross-hatched region shows the prediction of the collapse model for disc galaxy formation (see text). The dashed lines show the predicted threshold radii for star formation <cit.> for disc masses of, from bottom to top, 7.5 × 10^9, 5 × 7.5 10^9 and 25 × 7.5 10^9 . (b) – R_ max/h_ R versus face-on central surface brightness in the I-band. The arrows and the cross-hatched region is as in (a). The dashed line shows the constant star-formation threshold prediction. The shifted dotted line is obtained when allowing for an underestimation of the surface brightnesses by 0.7 mag (see text).

astro

ph0408125-fig1.jpg

0.434268

0fd2b56cfb1641bca1e0d80758b763b4

R_ max/h_ R versus face-on central surface brightness for the V and R-band results of Pohlen for the sharply truncated disc model (). The face-on surface brightnesses were calculated assuming a exponential vertical luminosity distribution. As in Fig. 1b, the cross-hatched region shows the prediction of the collapse model and the dotted line shows the prediction of the star-formation treshold model, with the except that here the (M/L)_⋆ for the threshold model is arbitrary.

astro

ph0408125-fig2.jpg

0.422923

d544df08f6504eb19db69ef1f3aaf94f

Likelihood contour plot in the n0-n'0 plane showing the 1,2 and 3σ contours from the WMAP data (Top Panel), WMAP+SLOAN (Middle Panel) and WMAP+SLOAN+Ly-α (Bottom Panel).

astro

ph0408129-clmo_run_tot.jpg

0.507566

9f076bf8281941c494a8cc0a7ec046cd

Likelihood contour plot in the plane c-(s+c/2) showing the 1,2 and 3σ contours from the WMAP+SLOAN+Ly-α data. We recall that these parameters are related to the physical inflaton potential parameters by c = -β_m(lnϕ_0)/(3 H_I^2) and s+c/2 = m^2(lnϕ_0)/(3 H_I^2).

astro

ph0408129-clmo_suo_lya.jpg

0.4734

660590c131114f54b8022aace9132de7

The theoretically expected region for the parameters s and c for a value of N0 = 50; the solid(red)-line-hatched region is strongly excluded, while the dashed(green)-line-hatched region is excluded only if the mass is supposed to run up to the end of inflation. The dotted (blue) line gives the prediction for the case when the end of inflation is triggered by η = 1. The circles correspond to the values in the explicit models discussed in section II.C: the upper ones (magenta) refer to the case of gauge coupling dominance, while the (blue) one at the origin to the case of Yukawa coupling dominance.

astro

ph0408129-clmo_th_sc.jpg

0.476068

ff3e7b07dc404e78abb3fb5cdebffb6a

Likelihood contour plot in the c-s plane showing the 1,2 and 3σ contours from the WMAP data (Top Panel), WMAP+SLOAN (Middle Panel) and WMAP+SLOAN+Ly-α (Bottom Panel).

astro

ph0408129-clmo_tot.jpg

0.474887

1bd535dc084c432ca1b45f0838ddb3cc

Four spectra of SN 2002cx, from L03, are compared. Epochs are in days with respect to the date of maximum brightness in the B band, 2002 May 21. The flux is per unit wavelength interval and the vertical displacement is arbitrary. Spectra have been corrected for the redshift of the parent galaxy, cz = 7184 . No correction for interstellar reddening has been applied.

astro

ph0408130-f1.jpg

0.402912

a14f8492553c40c1aa17050dc71e2983

The day -4 and day -1 spectra of SN 2002cx, and the day -4 spectrum of SN 1997br, are compared. The SN 1997br spectrum has been redshifted by 3000 to align the Fe III absorptions. The flux is per unit frequency interval (so that the spectra are roughly flat rather than steeply sloped) and the vertical displacement is arbitrary.

astro

ph0408130-f2.jpg

0.3937

a05c230d78fb41fdb1eac6923d0731d4

The day -1 spectrum of SN 2002cx (solid line) is compared with a synthetic spectrum (dashed line) that has v_phot=7000 , T_bb=14,000 K, and contains lines of six ions. The flux is per unit frequency interval. In this and subsequent figures the flux is linear rather than logarithmic.

astro

ph0408130-f3.jpg

0.446351

8040b7fd975b4550b939c50235d8089c

The day -4 spectrum of SN 1997br (solid line) is compared with a synthetic spectrum (dashed line) like that of Figure 3, for SN 2002cx, except that has been increased from 7000 to 12,000 .

astro

ph0408130-f4.jpg

0.403379

e94035517f0441fba21e8965f3df25a3

The day 12 spectrum of SN 2002cx (solid line) is compared with a synthetic spectrum (dashed line) that has v_phot=7000 , T_bb=9000 K, and contains lines of five ions. The flux is per unit wavelength interval.

astro

ph0408130-f5.jpg

0.48886

8e5944852fcc4ad69852f29d2879cc7f

The day 25 spectrum of SN 2002cx (solid line) is compared with a synthetic spectrum (dashed line) that has v_phot=5000 , T_bb=5500 K, and contains lines of five ions. The flux is per unit wavelength interval.

astro

ph0408130-f6.jpg

0.438254

5c114aef034341d8901979ddd24d0672

The day 56 spectrum of SN 2002cx (solid line) is compared with a synthetic spectrum (dashed line) that has v_phot=2000 , T_bb=7000 K, and contains lines of five ions. The flux is per unit wavelength interval.

astro

ph0408130-f7.jpg

0.421693

243c45ff6de04d42adeb91819ef2f87d

The spatial distribution (equatorial coordinates, Aitoff projection) of the 914 HRI pointings used for the BMW–HRI cluster survey. The origin of right ascension is in the center of the plot and the grid is in steps of 3^h in RA and 30^∘ in DEC. The empty area corresponds to the |b_II| < 20^∘ zone of avoidance, while the boxes mark the Magellanic clouds and Virgo cluster areas which were also excluded from the survey. Different symbols indicate different ranges of exposure time, as explained.

astro

ph0408131-1326f1.jpg

0.432291

bf9873699348488392d85263f2e54d60

The expected number of spurious detections in the BMW–HRI cluster catalogue as a function of the source significance, expressed in units of standard deviations σ (see text for details). The curve includes the contamination produced both by background fluctuations detected as true (extended) sources and by real point sources mistaken as extended by the algorithm. On the basis of this curve, the BMW–HRI p_s>4 master sample is expected to have a 27% contamination, while selecting at p_s>5 one obtains a sample with virtually null contamination.

astro

ph0408131-1326f11.jpg

0.555489

97e5ad1507994e6b9dc4e8899a04311a

The solid angle covered by the BMW–HRI survey (sky–coverage) as a function of flux and source extension, obtained by integrating the survey selection function over the observed fields.

astro

ph0408131-1326f12.jpg

0.551474

1ceccb2c6a764a11b481969d5b9c4c6e

Comparison of the BMW–HRI size-weighted sky coverage to those of the EMSS (Gioia et al. 1990), 160 Square Degrees (Vikhlinin et al. 1998a) and RDCS (Rosati et al. 1998). Since the sky coverage is a function of both flux and source extension, the BMW–HRI curve has been computed combining the sky coverages pertaining to different source extensions, according to the observed distribution of angular sizes in the p_s>4 sample.

astro

ph0408131-1326f13.jpg

0.40897

373b082f135f455ab88c9803f908e9d1

The classical Log N – Log S plot — i.e. the mean cumulative surface density of clusters as a function of X–ray flux — from the BMW–HRI survey (filled circles), compared to the RDCS (Rosati et al. 1998) and 160SD survey (Vikhlinin et al. 1998b). The BMW–HRI points are computed from the high-significance p_s>5 σ sub-sample, for which the expected contamination is negligible. The curves give the expected counts in the same Λ-cosmology adopted in this paper, for the cases of no-evolution and evolution of the XLF. The BMW-HRI and PSPC data are all consistent with a negligible to moderate evolution of the XLF.

astro

ph0408131-1326f15.jpg

0.466731

7ed41e9b890043b0a0f692c93fcacf2f

Optical/X–ray overlays for a selection of new BMW–HRI groups/clusters at different redshifts. From top left to bottom right: BMW134654.9-301328, z_spec=0.358; BMW212415.7-334754, z_spec=0.92; BMW122842.6-391612, z_phot≃ 0.8; BMW112059.0+130450, z_spec=0.615. The redshift for BMW212415.7-334754, a very compact cluster, awaits further confirmation, being based on only 3 redshifts and displaying a second system at z=0.62. The X–ray images have been smoothed with a Gaussian filter (σ = 40) and the isophotes correspond to 0.7,1,2,3 standard deviations over the background (after smoothing). Optical images are combined Gunn-r+i ∼ 60min total exposures with EFOSC2 at the ESO 3.6 m telescope. For all but the first object the sides of the figures are 2.5.

astro

ph0408131-1326f16.jpg

0.468062

3f79fc3b2b1444a38c9407dd2d21acf5

Distribution of exposure times for the 914 pointings used for the cluster survey.

astro

ph0408131-1326f2.jpg

0.448271

c34d8cfe803e41e98dfe9a0af40b05c2

Source extensions as measured by the BMW detection algorithm (W_clu, see text), plotted against the off-axis angle, showing our empirical definition of extended sources. The continuous line gives the mean extension of a point–like source at different off-axis angles, and the dashed lines identifies the 3 σ locus. Sources selected as cluster candidates have to lie, with their error bars, above the dashed line and are indicated by filled circles.

astro

ph0408131-1326f3.jpg

0.464011

033aec91140c41568602e4e512d1560e

Left panel: 24× 24 portion of an HRI image (Gaussian smoothed with σ=5) showing the detection of cluster candidate BMW044743.4-202042, marked by the solid 90 radius circle. The dashed circle (3 radius) indicates the central area containing the original HRI target (here barely visible), excluded from the cluster search. Right panel: Integral, background-subtracted flux growth curve of the candidate cluster (filled circles), plotted together with its 1σ error corridor (dotted lines) and the best-fitting β-model (Grey dashed curve). The total flux of the source is computed as the asymptotic value of the growth curve (horizontal dashed line). From the total flux, the core radius is then derived by comparison to a grid of β profiles convolved with the instrumental PSF at the specific off-axis angle. This is a typical source in our catalog, detected at ∼ 7 off–axis with 75 net counts in a 10 ks exposure, corresponding to a flux of ∼ 3×10^-13 erg s^-1 cm^-2.

astro

ph0408131-1326f4.jpg

0.494937

7d06684b8c5f478dbb1da4d822668eac

Comparison of the global pixel statistics of a real (black line) and a simulated image (Grey line), for a typical exposure time of 20,000 sec. X–ray photons have been binned in pixels of 5 size. The histograms show how the background statistical properties of the simulated image accurately match those of the real one. The simulated background is built as a Poissonian realization of the background map, and the differences that can be noticed for pixels with more than five counts are due to the effect of the simulated (extended) sources, which are overabundant in the simulated image.

astro

ph0408131-1326f8.jpg

0.469463

b56cfa08b06c4fa3837afb2bab0fc7b4

Top panel: visual comparison of the HRI image of BMW000141.3-154042 (left), the cluster detected with the largest number of net counts in the BMW–HRI survey, with a simulated X–ray source with the same counts and core radius. The bottom panel plots the average of the central 10 columns of the two images, with the real cluster described by the black line.

astro

ph0408131-1326f9.jpg

0.473244

86dd93a13e874e51bd4e801dca98e8a5

A comparison of the H41α (solid line) and H92α (dotted line, multiplied by a factor of 15) spectrum, one at the Sgr A^⋆ position (right) and at an adjacent position 7 to the east (left).

astro

ph0408136-f10.jpg

0.410014

b3c2ce99e33f42b1ab8076b8a6d5babf

Filter transmission curves for this survey. Solid line: curve for KP1590, used for Fields 1 and 2. Dashed line: curve for KP1467, used for Field 3. Both curves have been shifted 13 Åblueward to account for the f/3.5 beam of the Schmidt and the observing temperature. The histogram is a Gaussian distribution with mean –300 and σ = 200 km/s representing the expected M31 halo velocity distribution. The vertical lines show the location of velocities ± 400 km/s from M31's systemic velocity.

astro

ph0408137-f1.jpg

0.516722

94966ecf28e4480d8da6dc8ee4ea2586

The distribution of PN along the minor axis. PN in the complete subsample within 3 kpc of the minor axis are assigned an isophotal distance and binned into 2 kpc bins. The solid line is the number of PN expected based on the minor axis surface brightness profile from , normalized to the depth of our survey. The first three bins are reduced by our exclusion of PN near the center of the galaxy.

astro

ph0408137-f10.jpg

0.465167

0641e2dc721940d89886f0504e43e2ed

Upper panel: One realization of a bulge model with = 80 km/s and = 100 km/s. Lower panel: The probability distribution for the mean and for measured from the positions at |Y|>6 and |X|>10 for 1000 realizations of the same bulge model as in the upper panel. The solid lines represent the values measured from the data, = 116 km/s and = 79 km/s. All velocities are in km/s.

astro

ph0408137-f11.jpg

0.484035

9476f898ed424bc7bf417d256ed4fca3

The probability distribution for measured from the positions at |Y|>6 and |X|>10 for 1000 realizations of various bulge models. The solid lines represent the value from the data, = 79 km/s, and the dashed lines are the 10% confidence limits from the models. Upper panel: a model bulge with = 160 km/s and = 90 km/s. Middle panel: a model bulge with = 160 km/s and = 70 km/s. Lower panel: a model bulge with = 160 km/s and = 50 km/s.

astro

ph0408137-f12.jpg

0.466302

b331f0a74d004ab4b49cb4f3c659329e

The probability distribution for the mean and for measured from the positions at |Y|>6 and |X|>10 for 1000 realizations of various bulge models. The solid lines represent the values from the data, = 116 km/s and = 79 km/s, and the dashed lines are the 10% confidence limits from the models. Upper panel: a model bulge with = 170 km/s and = 70 km/s. Middle panel: a model bulge with = 150 km/s and = 70 km/s. Lower panel: a model bulge with = 130 km/s and = 70 km/s.

astro

ph0408137-f13.jpg

0.446898

e61486fe8d034b928d54e1e98bc32d01

Figure 2 from , showing M31's halo and tidal streams, with the positions of our potential halo objects marked (red in the electronic edition). All 5 fall on or near possible tidal streams, and three are counter-rotating with respect to the disk at their position. The two halo candidates near the center have velocities within 40 km/s of each other, and have velocities relative to M31 which are at least 3 times the central bulge dispersion. Orbits suggested by <cit.> and <cit.> for the Southern Stream <cit.> project near the position of these PN.

astro

ph0408137-f14.jpg

0.431194

48b3edc125784f1996a59b8d718d8cb6

Positions of our Schmidt fields and PN in M31. The R-band image was kindly provided by Rene Walterbos. North is up, East is left. M31's position angle is 37.3 <cit.>. The SW end of the galaxy is approaching, and the far side is to the SE of the major axis. Squares: PN in M31. Pluses: PN in M32. Xs: HII regions in And IV.

astro

ph0408137-f2.jpg

0.493979

5e9e9aba9c2f41fa9518851cccc16d4f

The estimated PN luminosity function for Fields 1 (panel a) and 3 (panel b). The X axis is the instrumental distribution shifted so that the bright edge of the distribution is at m_5007 = –4.5.

astro

ph0408137-f3.jpg

0.405172

b24f7c58d0e6428f93a1b4fa85ce32e6

Three typical spectra for our sample, with a range of signal-to-noise, showing the strongest PN emission lines, [OIII]and , and Hα.

astro

ph0408137-f4.jpg

0.448636

3f92597f229840bb877dbab1e7ed5a4c

Plot showing the positions of PN in M31. The major axis has been rotated through the position angle so that it is horizontal in the figure. Symbol size corresponds to velocity. Circles are PN velocities from this study, and squares are those from . Ellipses are bulge isophotes with b/a=0.55.

astro

ph0408137-f5.jpg