Datasets:
ReadingTimeMachine
/
rtm-sgt-ocr-v1

Dataset card Data Studio Files Community
2
source
stringlengths
1
2.05k
⌀
target
stringlengths
1
11.7k
The main problems that one has to face in this type of search are the large distances to the sources (typically a few kpe) and the confusion caused by stellar crowding (OB-type stars form im clusters).
The main problems that one has to face in this type of search are the large distances to the sources (typically a few kpc) and the confusion caused by stellar crowding (OB-type stars form in clusters).
These factors may explain the failure to detect disks in O-type stars. as opposed to the number of detections obtained for B-type stars (Cesaroni et al. 2007)).
These factors may explain the failure to detect disks in O-type stars, as opposed to the number of detections obtained for B-type stars (Cesaroni et al. \cite{ppv}) ).
In association with the most luminous YSOs one finds only huge (<0.1 pe). massive (a few 100 M.) cores. with velocity gradients suggesting rotation.
In association with the most luminous YSOs one finds only huge $\la$ 0.1 pc), massive (a few 100 ) cores, with velocity gradients suggesting rotation.
These objects. named "toroids. are likely non-equilibrium structures. because the ratio between the accretion time scale and the rotation period is shorter than for disks: this implies that the toroid does not have enough time to adjust its structure to the new fresh material falling onto it (Cesaroni et al. 2007::
These objects, named “toroids”, are likely non-equilibrium structures, because the ratio between the accretion time scale and the rotation period is shorter than for disks: this implies that the toroid does not have enough time to adjust its structure to the new fresh material falling onto it (Cesaroni et al. \cite{ppv};
Beltránn et al. 20115).
Beltránn et al. \cite{bel11}) ).
With this 1n. mind. one can see that understanding the formation of the most massive stars (>20M.) may benefit from a detailed investigation of toroids. also because these objects might be hosting true circumstellar disks in. their interiors.
With this in mind, one can see that understanding the formation of the most massive stars $>20~\Msun$ ) may benefit from a detailed investigation of toroids, also because these objects might be hosting true circumstellar disks in their interiors.
Moreover. their mere existence may set tighter constraints on theoretical models.
Moreover, their mere existence may set tighter constraints on theoretical models.
Despite the number of candidates. the existence of rotating toroids is still a matter of debate.
Despite the number of candidates, the existence of rotating toroids is still a matter of debate.
What is questioned is the nature of the velocity gradient. which is sometimes interpreted as expansion instead of rotation (see e.g. Gibb et al.
What is questioned is the nature of the velocity gradient, which is sometimes interpreted as expansion instead of rotation (see e.g. Gibb et al.
2004 and Araya et al.
\cite{gibb} and Araya et al.
2008 and references therein). thus suggesting that one might be seeing a compact bipolar outflow rather than a rotating core.
\cite{araya} and references therein), thus suggesting that one might be seeing a compact bipolar outflow rather than a rotating core.
In an attempt to distinguish between these two possibilities and more in general to shed light on these intriguing objects. we have focused our attention on one of the best examples: the hot molecular core (HMC) G31.41+0.31 (hereafter G31.41).
In an attempt to distinguish between these two possibilities and more in general to shed light on these intriguing objects, we have focused our attention on one of the best examples: the hot molecular core (HMC) G31.41+0.31 (hereafter G31.41).
This prototypical HMC ts located at a kinematic distance of 7.9 kpe and was originally imaged in the high-excitation (4.4) inversion transition of ammonia (Cesaront et al. 1994a))
This prototypical HMC is located at a kinematic distance of 7.9 kpc and was originally imaged in the high-excitation (4,4) inversion transition of ammonia (Cesaroni et al. \cite{cesa94}) )
and in the (6-5) rotational transitions of methyl cyanide(CH3CN:: Cesaront et al. 1994b)).
and in the (6–5) rotational transitions of methyl cyanide; Cesaroni et al. \cite{cesag31}) ).
The latter showed the existence of a striking velocity gradient (centered at an LSR. velocity of ~96.5 )) across the core in the NE-SW direction. already suggested by the distribution and velocities of OH masers (Gaume Mutel 1987)).
The latter showed the existence of a striking velocity gradient (centered at an LSR velocity of $\sim$ 96.5 ) across the core in the NE–SW direction, already suggested by the distribution and velocities of OH masers (Gaume Mutel \cite{gamu}) ).
Follow-up interferometric observations with better angular resolution and in high-energy tracers have confirmed this result and revealed the presence of deeply embedded YSOs. which in all likelihood explains the temperature increase toward the core center (Beltránn et al. 2004.. 2005..
Follow-up interferometric observations with better angular resolution and in high-energy tracers have confirmed this result and revealed the presence of deeply embedded YSOs, which in all likelihood explains the temperature increase toward the core center (Beltránn et al. \cite{bel04}, , \cite{bel05}, ,
hereafter BELO4 and BELOS: Cesaront et al. 2010)).
hereafter BEL04 and BEL05; Cesaroni et al. \cite{cesa10}) ).
The G31.41 HMC is separated by ~5”
The G31.41 HMC is separated by $\sim$
As mentioned in Sect. 3.1,
As mentioned in Sect. \ref{sec_evo},
the efficiency of rotational mixing relative to atomic diffusion is found to be greater in the central layers of a solar-type star than in its external layers.
the efficiency of rotational mixing relative to atomic diffusion is found to be greater in the central layers of a solar-type star than in its external layers.
In order to find out how this difference can be seen in the asteroseismic properties of a star, we now compare four different stellar models with the same age but computed with/without atomic diffusion and rotation.
In order to find out how this difference can be seen in the asteroseismic properties of a star, we now compare four different stellar models with the same age but computed with/without atomic diffusion and rotation.
The mean values of the large separation and of the ratio ro» are given in Table 1 for models with an age of GGyr.
The mean values of the large separation and of the ratio $r_{02}$ are given in Table \ref{tab:6gyr} for models with an age of Gyr.
Table 1. shows that models computed without atomic diffusion exhibit the higher values of the mean large separation, while rotating models exhibit the higher values of the mean ratio roy.
Table \ref{tab:6gyr} shows that models computed without atomic diffusion exhibit the higher values of the mean large separation, while rotating models exhibit the higher values of the mean ratio $r_{02}$.
We thus see that rotational effects have a more substantial impact on the ratio (102) than on the mean large separation (Av).
We thus see that rotational effects have a more substantial impact on the ratio $\langle r_{02} \rangle $ than on the mean large separation $\langle \Delta \nu \rangle $.
Rotational mixing is found to only partially inhibit the effects of atomic diffusion on the large separation, while the impact of rotation on the ratio ro? is much more important than the effects of atomic diffusion.
Rotational mixing is found to only partially inhibit the effects of atomic diffusion on the large separation, while the impact of rotation on the ratio $r_{02}$ is much more important than the effects of atomic diffusion.
This can be seen by comparing the rotating model R-D to the non-rotating models NR-D and NR-ND.
This can be seen by comparing the rotating model R-D to the non-rotating models NR-D and NR-ND.
Model D indeed exhibits a higher value of (Av) than model NR-D, but a slightly lower value than model NR-ND, which is computed without atomic diffusion.
Model R-D indeed exhibits a higher value of $\langle \Delta \nu \rangle $ than model NR-D, but a slightly lower value than model NR-ND, which is computed without atomic diffusion.
The situation is quite different for the mean ratio (ro?), because this value is significantly higher for model R-D than for both non-rotating models.
The situation is quite different for the mean ratio $\langle r_{02} \rangle $, because this value is significantly higher for model R-D than for both non-rotating models.
This can also clearly be seen in Fig.
This can also clearly be seen in Fig.
8 which shows the variation of ro? with frequency for the four models listed in Table 1..
\ref{r02_rotdiff_t6} which shows the variation of $r_{02}$ with frequency for the four models listed in Table \ref{tab:6gyr}.
Recalling that the large separation is principally sensitive to the mean stellar density and the ratio 102 to the central properties of the star, we conclude that the greater efficiency of rotational mixing relative to atomic diffusion in the central stellar layers is clearly visible in the asteroseismic properties of the models.
Recalling that the large separation is principally sensitive to the mean stellar density and the ratio $r_{02}$ to the central properties of the star, we conclude that the greater efficiency of rotational mixing relative to atomic diffusion in the central stellar layers is clearly visible in the asteroseismic properties of the models.
The increase of r9? for rotating models shown in Fig.
The increase of $r_{02}$ for rotating models shown in Fig.
8 is of course due to the higher value of X; at a given age when rotational effects are taken into account (see Fig. 3)),
\ref{r02_rotdiff_t6} is of course due to the higher value of $X_{\rm c}$ at a given age when rotational effects are taken into account (see Fig. \ref{xct_rotdiff}) ),
but also to the changes of the chemical profiles in the central stellar layers due to rotational mixing (see Fig. 9)).
but also to the changes of the chemical profiles in the central stellar layers due to rotational mixing (see Fig. \ref{profx}) ).
The effects of rotational mixing on the central stellar layers have been studied by discussing the changes of the small separation between modes with €=0 and £=2 and of the corresponding ratio of the small to large separation ro».
The effects of rotational mixing on the central stellar layers have been studied by discussing the changes of the small separation between modes with $\ell=0$ and $\ell=2$ and of the corresponding ratio of the small to large separation $r_{02}$.
The increase of the small separation and of the ratio of the small to large separation due to rotational mixing is also clearly visible for the separation between €=1 and €=3 modes (óvis()=νι— Vn-1,c=3) and between €=0 and ἐξ1 modes (ὀγρι(π)=νπε-ο—(πιει+ Yne=1)/2).
The increase of the small separation and of the ratio of the small to large separation due to rotational mixing is also clearly visible for the separation between $\ell=1$ and $\ell=3$ modes $\delta \nu_{13}(n) \equiv \nu_{n,\ell=1}- \nu_{n-1,\ell=3}$ ) and between $\ell=0$ and $\ell=1$ modes $\delta \nu_{01}(n) \equiv \nu_{n,\ell=0}-(\nu_{n-1,\ell=1} + \nu_{n,\ell=1})/2$ ).
This can be seen in Fig. 10,,
This can be seen in Fig. \ref{r01r02r13_t6},
which shows the ratio {οι=6vo1(1)/Ave=i(1) and 713=óvis(n)/Avizo(n+1) for rotating and non-rotating models with the same age of GGyr.
which shows the ratio $r_{01} \equiv \delta \nu_{01}(n) / \Delta \nu_{\ell=1}(n)$ and $r_{13} \equiv \delta \nu_{13}(n) / \Delta \nu_{\ell=0}(n+1)$ for rotating and non-rotating models with the same age of Gyr.
Figure 10 also shows that the changes of these asteroseismic ratios increase with the initial velocity of the model.
Figure \ref{r01r02r13_t6} also shows that the changes of these asteroseismic ratios increase with the initial velocity of the model.
This illustrates that the effects of rotational mixing on the stellar structure and hence on the asteroseismic properties of the models depend on the initial velocity, since a higher initial velocity leads to steeper gradients of the internal angular velocity, resulting in a more effective mixing by shear instability.
This illustrates that the effects of rotational mixing on the stellar structure and hence on the asteroseismic properties of the models depend on the initial velocity, since a higher initial velocity leads to steeper gradients of the internal angular velocity, resulting in a more effective mixing by shear instability.
In Sect. 3.2.1,,
In Sect. \ref{sec_modini},
the effects of rotational mixing on asteroseismic properties of solar-type stars have been studied by comparing rotating and non-rotating models computed with the same initial parameters.
the effects of rotational mixing on asteroseismic properties of solar-type stars have been studied by comparing rotating and non-rotating models computed with the same initial parameters.
We are now interested in comparing rotating and non-rotating models sharing the same luminosity, effective
We are now interested in comparing rotating and non-rotating models sharing the same luminosity, effective
observations through the ultra-violet. optical. ancl infrared Es»ectral regions (Ciareli et al.
observations through the ultra-violet, optical, and infrared spectral regions (Ciardi et al.
1998: Mason 2001). (2) looking for signatures of the secondary. star (Steeghs et 2001. Littlelair et al.
1998; Mason 2001), (2) looking for signatures of the secondary star (Steeghs et 2001, Littlefair et al.
2000. Dhillon Marsh 1905). (3) "weighting 10 secondary star in systems where the superhump ancl bital periods are known (the primary mass is usually wsumed. see for instance Patterson 2001). and (4) by Es»ectroscopie diagnostic of the stellar masses in. systenis where the white dwarf is revealed by their optical absorption wines (Moennickent ct al.
2000, Dhillon Marsh 1995), (3) “weighting" the secondary star in systems where the superhump and orbital periods are known (the primary mass is usually assumed, see for instance Patterson 2001), and (4) by spectroscopic diagnostic of the stellar masses in systems where the white dwarf is revealed by their optical absorption wings (Mennickent et al.
2001).
2001).
Concerning the first. two methods. one should mention that the determination of basic properties of secondaries in CVs by comparison. of their spectra with field stars calibrations is intrinsically uncertain.
Concerning the first two methods, one should mention that the determination of basic properties of secondaries in CVs by comparison of their spectra with field stars calibrations is intrinsically uncertain.
These results are prone to illumination and heating of the companion photosphere.
These results are prone to illumination and heating of the companion photosphere.
In. addition. the absorption spectrum may be allected by the filling of some lines with emission components.
In addition, the absorption spectrum may be affected by the filling of some lines with emission components.
While the Ht spectra. of. CVs has been measured and modeled in the past vears there are only à. few spectrophotometric observations of low-Luminosity systems in J. ancl Ix. band.
While the IR spectra of CVs has been measured and modeled in the past years there are only a few spectrophotometric observations of low-luminosity systems in J,H and K band.
In this paper we describe the spectra of Y CVs with orbital period below the period. gap.
In this paper we describe the spectra of 7 CVs with orbital period below the period gap.
These candidates [or systems at late evolutionary stages were selected for observation using the following criteria: i. their short orbital period. and/or ii.
These candidates for systems at late evolutionary stages were selected for observation using the following criteria: i. their short orbital period and/or ii.
a low mass transfer rate. inferred. from. their. high-aniplituce chwearlnova outbursts with long recurrence time.
a low mass transfer rate, inferred from their high-amplitude dwarf-nova outbursts with long recurrence time.
In the next section. we describe the LR spectroscopic observations while in Section. 3 a description of each spectra is given.
In the next section we describe the IR spectroscopic observations while in Section 3 a description of each spectra is given.
A brief discussion of the observational results is niade in Section 4.
A brief discussion of the observational results is made in Section 4.
A few conclusions and. perspectives are outlined in Section 5.
A few conclusions and perspectives are outlined in Section 5.
The infrared. spectroscopic observations reported. in this paper were obtained at ISO with VLT-AXntu using ISAAC spectrograph in service mode.
The infrared spectroscopic observations reported in this paper were obtained at ESO with VLT-Antu using ISAAC spectrograph in service mode.
Data were taken under photometric atmospheric conditions.
Data were taken under photometric atmospheric conditions.
An observing log is given in Table 1.
An observing log is given in Table 1.
Spectra in the J. HE and. Ix. bands: were obtained.
Spectra in the J, H and K bands were obtained.
Sullicient overlapping was assured for composing a single spectra ranging from 1.09 to 2.57 microns with a EWIIM resolution of 12 (J) to 27 (Ix) angstroms.
Sufficient overlapping was assured for composing a single spectra ranging from 1.09 to 2.57 microns with a FWHM resolution of 12 (J) to 27 (K) angstroms.
The data were reduced using by first. applying the combined dark and fatfield images as supplied by the ESO service mode operation group.
The data were reduced using by first applying the combined dark and flatfield images as supplied by the ESO service mode operation group.
Median sky frames were then combined for each object ancl spectral window.
Median sky frames were then combined for each object and spectral window.
This process made use of jittered images where the object is located at. different. positions along the slit.
This process made use of jittered images where the object is located at different positions along the slit.
Wavelength calibration was achieved by measuring the location of atmospheric OLL emission lines (Rousselot et.
Wavelength calibration was achieved by measuring the location of atmospheric OH emission lines (Rousselot et.
22000) in the sky. background.
2000) in the sky background.
Ehe flux calibration was performed using observations of the AO standard LD216009. mace with the same instrumental setup (but with a wider slit) by the ESO operation team as part of the service mode program.
The flux calibration was performed using observations of the A0 standard HD216009, made with the same instrumental setup (but with a wider slit) by the ESO operation team as part of the service mode program.
Finally. the tcllurie absorption features were corrected with the aid of the absorption template constructed by dividing the spectrum of LID216009 with a low order Xxvnomial fit. excluding a few stellar absorption lines.
Finally, the telluric absorption features were corrected with the aid of the absorption template constructed by dividing the spectrum of HD216009 with a low order polynomial fit, excluding a few stellar absorption lines.
The LIAE task "telluric was used to find the best scale ancl shift. factors which. when applied to the normalized elluric template. provided: a reasonable correction of the clluric absorptions in H1D2160009. and. science exposures.
The IRAF task “telluric" was used to find the best scale and shift factors which, when applied to the normalized telluric template, provided a reasonable correction of the telluric absorptions in HD216009 and science exposures.
This procedure worked well. except for the regions between 1.35-1.44 microns and 1.80-1.04 microns. characterized by wavy telluric absorption.
This procedure worked well, except for the regions between 1.35-1.44 microns and 1.80-1.94 microns, characterized by heavy telluric absorption.
These regions. corresponding to he ends of the J. HE ancl Ix spectra. were excluded. [rom he analysis and are not shown in this paper.
These regions, corresponding to the ends of the J, H and K spectra, were excluded from the analysis and are not shown in this paper.
Synthetic J. Hl. and. Ix. photometry of our. calibrated. standard: star observations were compared to broadband: photometry by Carter Aleadows (1995) showing cdillerences below 0.08 magnitudes.
Synthetic J, H, and K photometry of our calibrated standard star observations were compared to broadband photometry by Carter Meadows (1995) showing differences below 0.08 magnitudes.
Slit losses for our objects were estimated: by assuming a Gaussian seeing profile centered. in the slit aperture.
Slit losses for our objects were estimated by assuming a Gaussian seeing profile centered in the slit aperture.
For that we used the seeing value. included: in the report of the observing block and neglected any [actor due to telescope guiding.
For that we used the seeing value included in the report of the observing block and neglected any factor due to telescope guiding.
Fhese corrections were included in the J. IE and Ix magnitudes and in the final spectrum as well.
These corrections were included in the J, H and K magnitudes and in the final spectrum as well.
We are confident that this method worked. well. since our spectrophotomoetric magnitudes compare well with previously published photometry.
We are confident that this method worked well, since our spectrophotometric magnitudes compare well with previously published photometry.
In addition. we obtain a good match in the overlapping region between JL ancl Ix spectra.
In addition, we obtain a good match in the overlapping region between J,H and K spectra.
An attempt to quantify the properties of the LV spectra was made by cmploving a numerical fitting procedure.
An attempt to quantify the properties of the IR spectra was made by employing a numerical fitting procedure.
The spectral energy distribution (SED) in the LR was tentatively parameterized by adding the contribution from a lIate-tvpe template spectrum and power-law component.
The spectral energy distribution (SED) in the IR was tentatively parameterized by adding the contribution from a late-type template spectrum and power-law component.
Although the emission of the disk should differ substantially from a power-law in the ER. we introduced. this component as a. first approximation to the aceretion disc continuum.
Although the emission of the disk should differ substantially from a power-law in the IR, we introduced this component as a first approximation to the accretion disc continuum.
While the power law is smooth in our wavelength domain ane basically alfects the slope of the Ht continuum. the detailed. shape of the synthetic continuum in the J. IL and [dx is strongly dependent on the stellar template contribution.
While the power law is smooth in our wavelength domain and basically affects the slope of the IR continuum, the detailed shape of the synthetic continuum in the J, H and K is strongly dependent on the stellar template contribution.
A nonlinear least. squares fitting. procedure was calculated using the following TO)|λ' where $ is the observed. spectrum. T the red. dwarf template spectrum. A the wavelength in microns and. a.b.e parameters to be found.
A nonlinear least squares fitting procedure was calculated using the following $ S(\lambda) = a \times T(\lambda) + b \times \lambda^{c}$ where S is the observed spectrum, T the red dwarf template spectrum, $\lambda$ the wavelength in microns and $\it{a}$ $\it{b}$ $\it{c}$ parameters to be found.
Phe parameters e. b and e were adjusted to minimize the reduced. Chi-square between the observed. spectrum. and. a model fit.
The parameters $a$, $b$ and $c$ were adjusted to minimize the reduced Chi-square between the observed spectrum and a model fit.
OL course. a and ὁ are constrained to the positive domain.
Of course, $\it{a}$ and $\it{b}$ are constrained to the positive domain.
The data fit in this way were selected carefully avoiding emission lines and
The data fit in this way were selected carefully avoiding emission lines and
The impact of ACN outflows is well above the noise for all sources. even if Εν=2’ and 7.x = (10 p IK)? per arcuiu?.
The impact of AGN outflows is well above the noise for all sources, even if $\theta_{\rm FWHM} = 2'$ and $\sigma_{\rm pix}^2$ = (10 $\mu$ $^2$ per $^2$.
However. iu most cases. the profile of the SZ sienal above the noise is indistinguishable frou a point source (Which would appear on this plot as an inverted parabola that drops by a factor of two at a distance of Opwiuar/2).
However, in most cases, the profile of the SZ signal above the noise is indistinguishable from a point source (which would appear on this plot as an inverted parabola that drops by a factor of two at a distance of $\theta_{\rm FWHM}/2$ ).
At the same time quasars themselves are often significant microwave point sources.
At the same time quasars themselves are often significant microwave point sources.
To estimate this intrinsic contribution we convert the flux per unit frequency. Fy. to CMD temperature units as AT= =1Landy ji, where B, is the Plauck Function aud. as in ((2.2)). psLmhwefkfoxp (Scott White 1999).
To estimate this intrinsic contribution we convert the flux per unit frequency, $F_\nu$, to CMB temperature units as T = = )^2, where $B_\nu$ is the Planck Function and, as in \ref{eq:DT}) ), $x \equiv h \nu/ k T_{\rm CMB}$ (Scott White 1999).
The ratio of UL, at microwave waveleneths to the quasars bolometric huninosity ~1/1000 for "radio loud” objects which ike up about 10:4 of the population. and several orders of magnitude less for radio-quiet objects (Elvis 199D).
The ratio of $\nu L_\nu$ at microwave wavelengths to the quasars bolometric luminosity $\sim 1/1000$ for “radio loud” objects which make up about $10\%$ of the population, and several orders of magnitude less for radio-quiet objects (Elvis 1994).
Estimating a typical luminosity distance to a quasar as z 3000 Mpe ho? and cousideriug a typical observing frequency of 100 GITz this gives AT Ba. which is much larecr than the thermal SZ signal.
Estimating a typical luminosity distance to a quasar as $\approx$ 3000 Mpc $h^{-1}$ and considering a typical observing frequency of 100 GHz this gives T )^2, which is much larger than the thermal SZ signal.
While this can be reduced substantially bv selecting radio-quiet quasars. a better approach is to work with objects observed at later ties after the merger.
While this can be reduced substantially by selecting radio-quiet quasars, a better approach is to work with objects observed at later times after the merger.
In Figure 9 we show the azuuuthallv averaged excess thermal SZ contribution around post-starburst ealaxics. identified as mergers observed after the quasar phase. but within 200 Mrs after coalescence.
In Figure \ref{fig:200} we show the azimuthally averaged excess thermal SZ contribution around post-starburst galaxies, identified as mergers observed after the quasar phase, but within 200 Myrs after coalescence.
Here we have converted cach black hole mass to its corresponding stellar bulge mass using a factor of zLOO as obtained from the analysis in Marcoui Tut (2003).
Here we have converted each black hole mass to its corresponding stellar bulge mass using a factor of $\approx 400$ as obtained from the analysis in Marconi Hunt (2003).
Workiug with these sources has two main advantages.
Working with these sources has two main advantages.
Firstly. as they are much more common. one can consider VAightly larger bulges anc hence larger auc more spatially extended SZ cistortious. while at the same iue nuprovius the umuber of sources that can be 'oadded. im a fixed region of the sky.
Firstly, as they are much more common, one can consider slightly larger bulges and hence larger and more spatially extended SZ distortions, while at the same time improving the number of sources that can be coadded in a fixed region of the sky.
Furthermore. as ie nuclear LDuuinosifies are typically more than 100 iues less durius this phase Hopkins 2007). oomtf source subtraction is much easier. although some exclusion of racdio-loud sources may still be necessary.
Furthermore, as the nuclear luminosities are typically more than 100 times less during this phase Hopkins 2007), point source subtraction is much easier, although some exclusion of radio-loud sources may still be necessary.
Tn Figure 10. we show the SZ coutribution obtained w coadding all bulges observed at any time after the initial quasar pliase.
In Figure \ref{fig:1000} we show the SZ contribution obtained by coadding all bulges observed at any time after the initial quasar phase.