astro-ph0006007/ms.tex
1: \documentstyle[epsbox,subfigure]{mn}
2: \def\lsim{\mathrel{\mathpalette\Oversim<}}
3: \def\gsim{\mathrel{\mathpalette\Oversim>}}
4: \def\Oversim#1#2{\lower0.5ex\vbox{\baselineskip0pt\lineskip0pt%
5:             \lineskiplimit0pt\ialign{%
6:           $\mathsurround0pt #1\hfil##\hfil$\crcr#2\crcr\sim\crcr}}}
7: \newcommand{\re}{\par\hangindent=0.5cm\hangafter=1\noindent}
8: \newcommand{\se}{\par\hangindent=12mm\hangafter=1\indent}
9: \newcommand{\noi}{\noindent}
10: \newcommand{\bm}[1]{\mbox{\boldmath $#1$}}
11: \newcommand{\apj}{ApJ}
12: \newcommand{\apjs}{ApJS}
13: \newcommand{\mnras}{MNRAS}
14: \newcommand{\aap}{A\&A}
15: \def\goodgap{
16: \hspace{\subfigtopskip}
17: }
18: %%%%%%%%%%%%%%%%%
19: % (1)TITLE PAGE %
20: %%%%%%%%%%%%%%%%%
21: \title{UV Background-Induced Bifurcation of the Galactic Morphology}
22: \author[Susa \& Umemura]{Hajime~Susa \thanks{e-mail:susa@rccp.tsukuba.ac.jp}
23: and Masayuki~Umemura \thanks{e-mail:umemura@rccp.tsukuba.ac.jp}\\
24: Center for Computational Physics, University of
25:   Tsukuba, Tsukuba 305, Japan }
26: \date{Accepted 2000 May}
27: %\pagerange{\pageref{firstpage}--\pageref{lastpage}}
28: \pubyear{2000}
29: \begin{document}
30: \maketitle
31: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
32: % (2)Abstract  & Subject Headings %
33: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
34: 
35: \begin{abstract}
36: Based upon a novel paradigm of the galaxy formation under UV background,
37: the evolutionary bifurcation of pregalactic clouds is 
38: confronted with observations on elliptical and spiral galaxies.
39: The theory predicts that the dichotomy between 
40: the dissipational and dissipationless galaxy formation
41: stems from the degree of self-shielding from the UV background 
42: radiation. This is demonstrated on a bifurcation diagram of
43: collapse epochs versus masses of pregalactic clouds.
44: Using the observed properties, the collapse epochs are 
45: assessed for each type of galaxies with attentive mass estimation. 
46: By the direct comparison of the theory with the observations, 
47: it turns out that  
48: the theoretical bifurcation branch successfully discriminates
49: between elliptical and spiral galaxies.
50: This suggests that the UV background radiation 
51: could play a profound role for the differentiation 
52: of the galactic morphology into the Hubble sequence.
53: \end{abstract}
54: \begin{keywords}
55: galaxies: formation --- radiative transfer --- --- molecular processes
56: \end{keywords}
57: 
58: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%
59: % (3)TEXT & Acknowledgments %
60: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%
61: \section{Introduction}
62: \label{intro}
63: A substantial basis of the galaxy formation theory has been founded 
64: by several pioneering works in 1970s (\cite{RO77}; \cite{Silk77}). 
65: The theory predicts that the galactic scales are basically
66: determined by atomic cooling of hydrogen and partially helium.
67: The pregalactic evolution is elegantly summarized on the {\it cooling diagram} 
68: for virialized objects.    
69: %%%%%
70: Moreover, the origin of the Hubble type has been attributed to
71: the dissipativeness of the collapse, which is regulated by the efficiency
72: of star formation. If the star formation
73: proceeds after most of the gravitational energy 
74: is dissipated, the pregalactic clouds will evolve into spiral galaxies. 
75: This is a paradigm of the so-called {\it dissipational galaxy formation}
76: (e.g. \cite{Lar76}; \cite{Car85}; \cite{KG91}).
77: On the other hand, an early star formation episode 
78: leads to the {\it dissipationless galaxy formation}
79: (e.g. \cite{AB78}; \cite{AM90}),
80: ending up with the formation of elliptical galaxies.
81: However, the key mechanism which physically controls the star formation
82: efficiency has been hitherto unsolved. 
83: 
84: Very recently, Susa \& Umemura (2000) 
85: (hereafter SU) propose that
86: the self-shielding against UV background radiation regulates the
87: star formation in pregalactic clouds.
88: The star formation processes in primordial gas
89: have been explored by many authors with {\it ab initio} 
90: calculations (e.g. \cite{MST69}; \cite{Hut76}; \cite{Car81};
91: \cite{PSS83}; \cite{SUN96}; \cite{Uehara96}; \cite{ON98} 
92: ; \cite{Nishi98}; \cite{NU99}). 
93: The key physics is the radiative cooling by H$_2$ line emission, 
94: because H$_2$ is the only coolant for primordial gas 
95: at $T\lsim 10^4$ K. 
96: SU have studied the efficiency of H$_2$ cooling 
97: in collapsing clouds exposed to UV background radiation,
98: because it is significant after the reionization of the universe, 
99: probably at $z\lsim 10$ (\cite{NUS99}, and references therein).
100: SU have found that if a cloud undergoes the first sheet collapse 
101: at higher redshifts ($z\gsim 4$),
102: then the cloud is quickly shielded against the UV background and
103: consequently cools down due to the efficient formation of H$_2$.
104: Resultantly, it leads to an early burst of star formation.
105: On the other hand, the shielding is retarded for a later collapsing cloud
106: at $z\lsim 4$, resulting in the dissipational galaxy formation. 
107: As a result, the bifurcation branch of the self-shielding 
108: is corresponding to the
109: boundary between the dissipationless and dissipational galaxy formation. 
110: In this {\it Letter}, this novel bifurcation theory 
111: is confronted with the observations of elliptical and spiral galaxies to
112: elucidate whether the theory is practically successful or not.
113:  
114: \section{Bifurcation Theory}
115: \label{BITH}
116: In Figure 1, we show the bifurcation theory. 
117: Here, the cosmological parameters are assumed to be $\Omega=0.3$,
118: $h=0.7$, and $\Omega_{\rm b} h^2=0.02$
119: with usual meanings. 
120: Originally, 
121: the sheet collapse was pursued with two initial parameters, i.e., 
122: the mean density ($\bar{n}_{\rm ini}$) and the thickness ($\lambda$).
123: Here, the initial parameter space is translated into the baryonic mass 
124: [$M_{\rm b}\equiv (4\pi/3) \bar{n}_{\rm ini}(\lambda/2)^3$]
125: and the collapse epoch ($z_{\rm c}$) by 
126: assuming the initial stage is close to the maximum expansion
127: of a density fluctuation.
128: In the region (a) in Figure 1, a pregalactic cloud is self-shielded
129: against the external UV in the course of the sheet collapse, so that
130: the cloud cools down below $10^3$K and undergoes efficient star formation.
131: Hence, it is expected to evolve into an early type galaxy with 
132: a large bulge-to-disk ratio (B/D) due to the dissipationless virialization.
133: In the region (b), the cloud is not self-shielded during the sheet 
134: collapse, but will be self-shielded through the shrink to the rotation
135: barrier. This leads to the retarded star formation, and thus
136: the virialization would proceed in a fairly dissipative fashion. 
137: As a result, a late type
138: (small B/D) galaxy would be preferentially born. The region (c)
139: represents the forbidden region of the collapse due to the Jeans 
140: stability.
141: 
142: SU have considered only baryonic component, because, 
143: in the later collapsing phase, the cooling sheet 
144: is dominated by baryons, not by the
145: diffuse dark matter. Then, if the UV background is constant,
146: the bifurcation mass scale is given as
147: \begin{eqnarray}
148: M_{\rm SB}^{\rm max}=2.2\times 10^{11} M_\odot 
149: \left(\frac{1+z_{\rm c}}{5}\right)^{-4.2}
150: \left(\frac{I_{21}}{0.5}\right)^{0.6} \label{eq:msb},
151: \end{eqnarray}
152: where $I_{21}$ is the UV background intensity in units of
153: $10^{-21}{\rm erg~ s^{-1} cm^{-2} str^{-1} Hz^{-1}}$.
154: %%%%%%%%%%%%%%%%%%%
155: However, the dark matter potential may affect the
156: evolution in the early phase of collapse. 
157: If we take this effect into account maximally, 
158: the Jeans length is reduced by a factor of 
159: $\sqrt{\Omega_{\rm b}/\Omega}$. 
160: With this Jeans scale,
161: the bifurcation mass is changed to be
162: $M_{\rm SB}^{\rm min}=\left(\Omega_{\rm b}/\Omega\right)^{1.2}
163: M_{\rm SB}^{\rm max}$. Thus, the practical bifurcation mass would be 
164: between $M_{\rm SB}^{\rm max}$ and $M_{\rm SB}^{\rm min}$.
165: 
166: Also, SU have assumed the UV intensity to be independent of time. 
167: Practically, the intensity seems to evolve.
168: Here we include the effect of the evolution of 
169: UV background radiation.
170: We assume $I_{21}=0.5\left[\left(1+z\right)/3 \right]^3$ for $z\le 2$ 
171: and $I_{21} =0.5$ for $2 < z \le 4$.
172: This dependence is consistent with the UV intensity
173: in the present epoch (\cite{Mal93}; \cite{DS94}), and the value inferred from 
174: the QSO proximity effects at high redshifts 
175: (\cite{BDO88}; 
176: Giallongo et al. 1996).
177: As for $z > 4$, two extreme models are employed. 
178: The first one is (A) the exponentially damping model,
179: $I_{21}=0.5\exp\left[3\left(4-z\right)\right]$
180: (\cite{Ume00}), and 
181: the second one is (B) the constant extrapolation model,
182: $I_{21}=0.5$. 
183: In Figure \ref{fig1}, these two extreme models are shown.
184: The difference emerges especially at $z_{\rm c} \gsim 7$. 
185: 
186: \section{Estimation for Galactic Masses and Collapse Epochs}
187: \label{INOBDA}
188: Here, based upon the observational data, we attempt to 
189: assess the total baryonic masses of ellipticals as well 
190: as spirals and their collapse epochs. Then, the observed galaxies
191: are compared with the theory in the bifurcation diagram.
192: 
193: To begin with, we evaluate the baryonic masses from B-band luminosities 
194: for ellipticals and from I-band luminosities for spirals
195: with the mass-to-light ratio; 
196: $
197: M_{\rm b}=\left[M_*/L\right]L/f_*, 
198: $
199: where $M_*$ is the total stellar mass and 
200: $f_*$ is the mass fraction of stellar component 
201: in the total baryonic mass.
202: The mass-to-light ratios $M_*/L$ are obtained theoretically as well as 
203: observationally.  
204: Based on the chemical evolution theory 
205: (\cite{KA97}),
206: the $M_*/L$ at B-band for ellipticals is $4-9$.
207: This value is consistent with the observational estimates 
208: of $4-7$
209: (\cite{Ber93}; \cite{Piz97}). 
210: In this paper, we adopt the luminosity-dependent values given 
211: by Kodama \& Arimoto (1997).
212: The $M_*/L$ at I-band for spirals is 
213: 3.37(Sa), 2.91(Sb), 1.79(Sc), and 1.33(Sd) 
214: based on the code developed by Kodama
215: \& Arimoto (1997) with the S1 model in Arimoto, Yoshii,
216: \& Takahara (1992).
217: These are also consistent with the observed data 
218: in Rubin et al.(1985).
219: $f_*$ is theoretically calculated 
220: by the population synthesis model to be 
221: 0.963(Sa), 0.908(Sb), 0.462(Sc), and 0.125(Sd), while
222: basically $f_*=1$ for ellipticals.
223: With the estimated baryonic masses, we can assess the total masses as
224: $M_{\rm tot}=M_{\rm b}(\Omega/\Omega_{\rm b})$.
225: 
226: Next, we estimate the collapse epochs with the help of virial theorem.
227: For the purpose, we use the observed 1-D velocity dispersion. 
228: We assume that the system is spherical and the dark halo has
229: isothermal distributions after the virialization. 
230: We suppose a density perturbation as 
231: $\delta_i (r) = \bar{\delta_i} g(r)$, where
232: $r$ is the comoving radial coordinate and
233: $g(r)$ is a function which satisfies the normalization 
234: as $\frac{3}{R^3}\int_0^R g(r)r^2dr = 1$,
235: with $R$ being the comoving radius of the perturbed region.
236: $\bar{\delta_i}$ is the spatially averaged $\delta_i(r)$
237: in the volume $r \le R$.
238: Summing up the initial kinetic energy and gravitational energy,
239: we have the initial total energy of the perturbed region as
240: \begin{eqnarray}
241: E_{\rm ini}=-\frac{3GM_{\rm tot}^2}{5R} 
242: \left(1+z_i\right)\bar{\delta_i}\left(1+\frac{2}{3}\phi \right)F,
243: \label{eqn:Eini}
244: \end{eqnarray}
245: where $z_i$ is the initial redshift, $\phi$ is the contribution from 
246: the peculiar velocities, 
247: and $M_{\rm tot}$ is the total mass enclosed 
248: within the comoving radius $R$. 
249: The factor $F$ is defined as 
250: \begin{eqnarray}
251: F &\equiv& \frac{5}{R^5}\int_0^R r^4 \bar{g}(r)dr, \label{eqn:defG2}
252: \end{eqnarray}
253: where $\bar{g}(r)\equiv \frac{3}{r^3}\int_0^r r'^2 g(r')dr'$.
254: In the linear regime, $g(r)\propto r^{1/(n+3)}$ around a density peak
255: (\cite{HS85}), where $n$ denotes the index of the CDM power
256: spectrum. Then  equation (\ref{eqn:defG2})
257: is readily integrated to give $F = 5/(2-n)$. 
258: We assume $n=-1$ throughout this paper, because it is the case
259: for galactic scales in CDM cosmologies.
260: The total energy of an observed galaxy is derived
261: in terms of the virial theorem;
262: $
263: E_{\rm obs}=-\frac{3}{2}M_{\rm tot}\sigma_{\rm 1D}^2,
264: $
265: where $\sigma_{\rm 1D}$ is the 1-D internal velocity dispersion 
266: of the galaxy. 
267: Assuming the energy conservation, i.e., $E_{\rm int}=E_{\rm obs}$,
268: we have
269: \begin{eqnarray}
270: \frac{\left(1+z_i\right)\delta_i\left(1+2\phi/3 \right)}{R} 
271: = \frac{5\sigma_{\rm 1D}^2}{2GM_{\rm tot}} F^{-1}.
272: \label{eqn:Econ}
273: \end{eqnarray}
274: 
275: On the other hand, the time when the outer boundary 
276: of a perturbation collapses is analytically obtained (\cite{Pee93})
277: as
278: \begin{eqnarray}
279: t=\frac{\pi}{\left(GM_{\rm tot}\right)^{1/2}}
280: \left(
281: \frac{R}{2 \left(1+z_i\right) \delta_i \left(1+2\phi/3 \right) 
282: }\right)^{3/2}.
283: \label{eqn:tff}
284: \end{eqnarray}
285: Finally, using equations (\ref{eqn:Econ}) and (\ref{eqn:tff}), 
286: we have the collapse epoch as
287: $
288: t=\pi GM_{\rm tot} F^{3/2}/5^{3/2} \sigma_{\rm 1D}^3. 
289: $
290: By equating this time with the Hubble time $t_{\rm H}(z_{\rm c})$,
291: $t$ can be translated into the collapse redshift, $z_{\rm c}$. 
292: Thus, it turns out that the velocity dispersion
293: is a key quantity to determine $z_{\rm c}$.
294: We use the observed stellar velocity dispersion 
295: as $\sigma_{\rm 1D}$ for elliptical galaxies. 
296: Also, for some luminous ellipticals, $\sigma_{\rm 1D}$ is estimated
297: by the X-ray temperature $T_X$ as
298: $\sigma_{\rm 1D} = \sqrt{kT_X/\mu m_p}$. The estimation by X-ray
299: gives typically twice the optical estimation, being probably 
300: the maximal assessment of $\sigma_{\rm 1D}^2$.
301: For spiral galaxies, we interpret the asymptotic rotational velocity
302: into the velocity dispersion by the relation 
303: $\sigma_{\rm 1D}=v_{\rm rot}/\sqrt{2}$.
304: 
305: The basic data on the luminosities and the velocity dispersions
306: are taken from 
307: the table of Faber et al. (1989) 
308: for 332 ellipticals. The galaxies are selected from the original
309: sample on the condition that the surface brightness, angular size, and
310: velocity dispersion are measured. 
311: The X-ray data for 12 ellipticals are adopted from the table 
312: in Matsumoto et al. (1997).
313: Further, for 468 spiral galaxies, we combine the
314: I-band luminosities from Mathewson et al. (1996) 
315: and the asymptotic rotation velocities 
316: from Persic \& Salucci (1995).
317: We pick up the galaxies whose asymptotic velocities [$V_{as}$ as in
318: Persic \& Salucci (1995)] are given. Thus the
319: number of galaxies is smaller than the original one.
320: 
321:  
322: \section{Theory versus Observations}
323: \label{THOB}
324: %Coincidence
325: In Figure \ref{fig2}, the bifurcation theory is confronted with 
326: observations.
327: In this figure, cosmological parameters are assumed as $\Omega=0.3$,
328: $\Omega_\Lambda=0$, 
329: $h=0.7$, and $\Omega_{\rm b} h^2=0.02$. All of these
330: values are plausible in the light of the recent observations,
331: although the value of $\Lambda$ parameter is still controversial. 
332: The small filled triangles and open circles represent respectively 
333: the observed elliptical and spiral galaxies. 
334: The open stars are the X-ray luminous elliptical galaxies.
335: We find that the two types of galaxies are successfully divided 
336: by the theoretical bifurcation mass scale.
337: This implies that the self-shielding against 
338: UV background radiation is practically related to 
339: the origin of the galactic morphology.
340: The difference between model A and B is not so significant, although
341: model A predicts a bit more ellipticals than observed.
342: Also, it is worth noting that few spiral galaxies reside above
343: the bifurcation mass scale, 
344: whereas there are a noticeable number of elliptical galaxies
345: which seem to have formed at lower redshift epochs 
346: than predicted by the present theory.
347: Since a condition for constant velocity dispersion gives a relation
348: $M \propto (1+z)^{-3/2}$ in this diagram,
349: most of discrepant low-redshift ellipticals turn out to be 
350: relatively luminous galaxies with higher stellar velocity dispersions.
351: Intriguingly, Gonzalez (1993) has reported,
352: using strengths of $H_\beta$ and [MgFe] indices, the evidence
353: for intermediate-age populations in elliptical galaxies. 
354: In addition, it has been argued that the galaxy merger with
355: burst-like star formation can lead to the formation of
356: elliptical galaxies (\cite{LT78}; Barnes \& Hernquist 1991, 1996; 
357: \cite{KC98}). 
358: Also, Shier \& Fischer (1998) suggest, by studying stellar kinematics of
359: starbursting infrared-luminous galaxies with obvious morphological
360: signatures of merger, that they can be progenitors of elliptical galaxies 
361: with high stellar velocity dispersions. 
362: Thus, the discrepant low-redshift ellipticals 
363: might not be of primordial collapse, but could be a category of
364: merger remnants.
365: 
366: In Figure \ref{fig2}, $1\sigma$, $2\sigma$, and $3\sigma$
367: CDM density perturbations normalized by COBE data 
368: (\cite{HuS96}; \cite{BW97}) are also plotted.
369: We find that elliptical galaxies form typically 
370: from $\sim 4\sigma$ perturbations 
371: and spirals do from $\sim 2\sigma$ peaks. 
372: If we change the cosmology, the results alter to some degree, 
373: because the bifurcation mass, the CDM spectra and the data points
374: of elliptical, and spiral galaxies are relatively changed. 
375: The main change of Figure \ref{fig2} comes from changing $\Omega$.
376: The collapse epoch $z_c$ for the observational data is dependent upon 
377: $\Omega$ as roughly $(1+z_c) \propto \Omega^{-0.74}$  
378: for $\Omega_\Lambda=0$ universe.
379: This dependence is mainly due to the evaluation of dark mass in a
380: galaxy. For instance, if we employ $\Omega=1$ (although it is unlikely), 
381: $1+z_{\rm c}$ becomes almost 2.5 times smaller than 
382: the estimates in Figure \ref{fig2}.
383: In this case, the theory and the observational data seem perceptibly 
384: discrepant with each other.
385: On the other hand, Figure \ref{fig2} is insensitive to 
386: the cosmological constant parameter $\Omega_\Lambda$.
387: In fact, even if we employ $\Omega=0.3$ and $\Omega_\Lambda=0.7$,
388: the relative position between the bifurcation mass and
389: observations remains almost unchanged, although
390: the correspondence between CDM fluctuations and observations
391: shifts from $\sim 4\sigma$ to $\sim 3\sigma$ for ellipticals.
392: In this case, the theory is in a good agreement with the observations.
393: Further details of 
394: the dependences on the cosmological model will be discussed in a
395: forthcoming paper.
396: 
397: %density-morphology relation 
398: The present results are also intriguing from a view point of
399: the statistics of galaxies.
400: Bardeen et al. (1986) have shown that 
401: higher $\sigma$ peaks reside preferentially in denser regions 
402: rather than in low-dense regions. 
403: As a result, the so-called density morphology relation 
404: can be explained as a natural consequence of the bifurcation theory.
405: Furthermore, it is known that the specific angular momenta $J/M$ of spirals 
406: are systematically greater by a factor of $\gsim 3$ than $J/M$ of 
407: ellipticals in the same mass range (\cite{Fal83}).
408: Based upon the tidal origin of the angular momentum, 
409: we expect $J/M \propto (1+z_c)^{-1/2}\nu^{-1}$, 
410: where $\nu \equiv \delta/\sigma$ (\cite{HP88}).
411: If ellipticals form from $\sim 4\sigma$ and
412: spirals do from $\sim 2\sigma$, then we anticipate
413: $(J/M)_{\rm spiral}\approx 3 (J/M)_{\rm elliptical}$.
414: \section*{Acknowledgments}
415: We appreciate the comments by the anonymous referee 
416: which helped us to improve the paper.
417: We thank M. Nagashima who kindly provided the data of mass-to-light
418: ratio. We also thank Y. Kanya and R. Nishi for useful information 
419: and comments. The analysis has been made with computational facilities 
420: at the Center for Computational Physics in University of Tsukuba. 
421: This work is supported in part by Research Fellowships of the Japan 
422: Society for the Promotion of Science for Young Scientists, No. 2370 (HS)
423: and by the Grants-in Aid of the Ministry of Education, Science, 
424: Culture, and Sport, 09874055 (MU). 
425: 
426: %%%%%%%%%%%%%%%%%%
427: % (4) Appendices %
428: %%%%%%%%%%%%%%%%%%
429: % NO APPENDICES
430: 
431: %%%%%%%%%%%%%%%%%
432: % (5)References %
433: %%%%%%%%%%%%%%%%%
434: \begin{thebibliography}{}
435: \bibitem[Aarseth \& Binney 1978]{AB78}
436: Aarseth, S. J., \& Binney, J. 1978, \mnras, 185, 227
437: 
438: \bibitem[Aguilar \& Merritt 1990]{AM90}
439: Aguilar, L. A., \& Merritt, D. 1990, \apj, 354, 33
440: 
441: \bibitem[Arimoto, Yoshii, \& Takahara 1992]{AYT92}
442: Arimoto, N., Yoshii, Y., \& Takahara, F. 1992, \aap, 253, 21
443: 
444: \bibitem[Bajtlik, Duncan, \& Ostriker 1988]{BDO88}
445: Bajtlik, S. Duncan, R. C., \& Ostriker, J. P. 1988, ApJ, 327, 570
446:  
447: \bibitem[Bardeen et al. 1986]{BBKS}
448: Bardeen, J. M., Bond, J. R., Kaiser, N., \& Szalay, A. S. 
449: 1986, \apj, 304, 15
450: 
451: \bibitem[Barnes \& Hernquist 1991]{BH91}
452: Barnes, J. E. \& Hernquist, L.E. 1991, \apj, 370, L65
453: 
454: \bibitem[Barnes \& Hernquist 1996]{BH96}
455: Barnes, J. E. \& Hernquist, L.E. 1996, \apj, 471, 115
456: 
457: \bibitem[Bertola et al. 1993]{Ber93}
458: Bertola, F., Pizzella, A., Persic, M., \& Salucci, P. 1993, \apj, 416, L45
459: 
460: \bibitem[Bunn \& White 1997]{BW97}
461: Bunn, E. F., \& White, M. 1997, \apj, 480, 6
462: 
463: \bibitem[Carlberg 1981]{Car81}
464: Carlberg, R. G. 1981, \mnras, 197, 1021 
465: 
466: \bibitem[Carlberg 1985]{Car85} 
467: Carlberg, R. G. 1985, \apj, 298, 486
468: 
469: \bibitem[Dove \& Shull 1994]{DS94}
470: Dove, J. B., \& Shull, M. 1994, \apj, 423, 196
471: 
472: \bibitem[Faber et al. 1989]{Fab89}
473: Faber, S. M. et al. 1989, \apjs, 69, 763
474: 
475: \bibitem[Fall 1983]{Fal83}
476: Fall, S. M. 1983, in IAU Symp. 100, 
477: Internal kinematics and dynamics of galaxies
478: (Dordrecht; Reidel), p.391 
479: 
480: \bibitem[Giallongo et al. 1996]{GCD96}
481: Giallongo, E., Cristiani, S., D'Odorico, S., Fontana, A., \& Savaglio, S. 1996,
482: ApJ, 466, 46
483: 
484: \bibitem[Gonzalez 1993]{Gon93}
485: Gonzalez, J.J. 1993, Ph.D. thesis, Univ. California, Santa Cruz
486: 
487: \bibitem[Heavens \& Peacock 1988]{HP88}
488: Heavens, A., \& Peacock, J. 1988, \mnras, 232, 339
489: 
490: \bibitem[Hoffman \& Shaham 1985]{HS85}
491: Hoffman, Y., \& Shaham, J. 1985, \apj, 297, 16
492: 
493: \bibitem[Hu \& Sugiyama 1996]{HuS96}
494: Hu, W., \& Sugiyama, N. 1996, \apj, 471, 542
495: 
496: \bibitem[Hutchins 1976]{Hut76}
497: Hutchins, J. B. 1976, \apj, 205, 103
498: 
499: \bibitem[Kauffmann \& Charlot 1998]{KC98} 
500: Kauffmann, G. \& Charlot, S. 1998, \mnras, 294, 705
501: 
502: \bibitem[Katz \& Gunn 1991]{KG91} 
503: Katz, N., \& Gunn, J. E. 1991, \apj, 377, 365
504: 
505: \bibitem[Kodama \& Arimoto 1997]{KA97}
506: Kodama, T., \& Arimoto, N. 1997, \aap, 320, 41
507: 
508: \bibitem[Larson 1976]{Lar76} 
509: Larson, R. B. 1976, \mnras, 176, 31
510: 
511: \bibitem[Larson \& Tinsley 1976]{LT78} 
512: Larson, R. B., \& Tinsley, B. M. 1978, \apj, 219, 46
513: 
514: \bibitem[Maloney 1993]{Mal93}
515: Maloney, P. 1993, ApJ, 414, 41
516: 
517: \bibitem[Mathewson et al. 1996]{Math96}
518: Mathewson, D. S., \& Ford, V. L., 1996, \apjs, 107, 97
519: 
520: \bibitem[Matsuda, Sato \& Takeda 1969]{MST69}
521: Matsuda, T., Sato, H., \& Takeda, H. 1969, 
522: Prog. Theor. Phys. 42, 219
523: 
524: \bibitem[Matsumoto et al. 1997]{Mat97}
525: Matsumoto, H. et al., 1997, \apj, 482, 133
526: 
527: \bibitem[Nakamoto, Umemura, \& Susa 1999]{NUS99}
528: Nakamoto, T., Umemura, M., \& Susa, H. 1999, submitted to MNRAS
529: 
530: \bibitem[Nakamura \& Umemura 1999]{NU99} 
531: Nakamura, F., \& Umemura, M. 1999, \apj, 515, 239
532:  
533: \bibitem[Nishi et al. 1998]{Nishi98}
534: Nishi, R., Susa, H., Uehara, H., Yamada, M. \& Omukai, K. 1998, 
535: Prog. Theor. Phys., 100, 881
536: 
537: \bibitem[Omukai \& Nishi 1998]{ON98}
538: Omukai, K., \& Nishi, R. 1998, \apj, 508, 141
539: 
540: \bibitem[Palla, Salpeter \& Stahler 1983]{PSS83}
541: Palla, F., Salpeter, E. E., \& Stahler, S. W. 1983 
542: \apj, 271, 632
543: 
544: \bibitem[Peebles 1993]{Pee93}
545: Peebles, P. J. E., 1993, in Principles of Phisical Cosmology,
546: (Princeton), p.484
547: 
548: \bibitem[Persic \& Salucci 1995]{PS95}
549: Persic, M., \& Salucci, P. 1995, \apjs, 99, 501
550: 
551: \bibitem[Pizzella et al. 1997]{Piz97}
552: Pizzella, A. et al. 1997, \aap, 323, 349
553: 
554: \bibitem[Rees \& Ostriker 1977]{RO77}
555: Rees, M. J., \& Ostriker, J. P. 1977, \mnras, 179, 541 
556: 
557: \bibitem[Rubin et al. 1985]{Rub85}
558: Rubin, V. C., Burstein, D., Ford, W. K. Jr., \& Thonnard, N. 1985, \apj,
559:  289, 81
560: 
561: \bibitem[Shier \& Fischer 1998]{SF98} 
562: Shier, L. M., \& Fischer, J, 1998, \apj, 497, 163
563: 
564: \bibitem[Silk 1977]{Silk77} 
565: Silk, J. 1977, \apj, 211, 638
566: 
567: \bibitem[Susa, Uehara \& Nishi 1996]{SUN96}
568: Susa, H., Uehara, H., \& Nishi, R. 1996, Prog. Theor. Phys., 96, 1073 
569: 
570: \bibitem[Susa \& Umemura 2000]{SU00}
571: Susa, H., \& Umemura, M. 2000, \apj, in press, astro-ph/0001169
572: 
573: \bibitem[Uehara et al. 1996]{Uehara96}
574: Uehara, H., Susa, H., Nishi, R., Yamada, M., \& Nakamura, T. 1996, 
575: \apj, 473, L95 
576: 
577: \bibitem[Umemura et al. 2000]{Ume00}
578: Umemura, M., Nakamoto, T., \& Susa, H. 1999, in preparation
579: \end{thebibliography}
580: \newpage
581: %%%%%%%%%%%%%%%%%%%%%%%
582: % (6) Figure Captions %
583: %%%%%%%%%%%%%%%%%%%%%%%
584: \onecolumn
585: \begin{figure}
586: \begin{center}
587: \subfigure[]{\psbox[width=80mm,vscale=1.0]{f1a.eps}}\goodgap
588: \subfigure[]{\psbox[width=80mm,vscale=1.0]{f1b.eps}}
589: \end{center}
590: \caption[f1a.eps,f1b.eps]{The bifurcation diagram. The prediction of galactic 
591: evolution is illustrated on the plane of 
592: the baryonic mass ($M_{\rm b}$) versus collapse redshift ($z_c$).  
593: Fig.\ref{fig1}a corresponds to
594: the UV evolution of model A (see text), and Fig.\ref{fig1}b 
595: for model B (see also text). 
596: In both of the figures, the half-tone shaded area (a) 
597: denotes the region in which pregalactic clouds evolve into early type
598: galaxies. 
599: The shaded area (b) leads to the formation of late-type galaxies.
600: The unshaded area (c) represents the region in which
601: the density perturbations cannot grow into bound objects, due to the thermal
602: pressure of ionized gas. 
603: Here we employ $M_{\rm SB}^{\rm max}$ as the bifurcation mass,
604: although the effect of dark matter may reduce the mass by factor
605: $(\Omega_{\rm b}/\Omega)^{1.2}$ at most. 
606: $M_{\rm shield}$ denotes the mass 
607: above which the cloud is initially self-shielded. }
608: \label{fig1}
609: \end{figure}
610: 
611: \begin{figure}
612: \begin{center}
613: \psbox[width=150mm,vscale=1.0]{f2.eps}
614: \end{center}
615: \caption[f2.eps]{Comparison of observed galaxies with the bifurcation theory.
616: Small open circles represent spiral galaxies in Persic \& Salucci (1995). 
617: Small solid triangles denote the E and E-S0 galaxies in Faber et al. (1989). 
618: Open stars are elliptical galaxies from  X-ray observations in Matsumoto
619: et al. (1997). The thick solid and dashed lines represent the
620:  bifurcation mass $M_{\rm SB}^{\rm max}$, respectively for model A and
621:  B, where both are identical at $z_{\rm c}<4 $.  Three thin solid lines
622:  marked as $1\sigma$, $2\sigma$, and $3\sigma$ are the prediction in the
623:  CDM cosmology, where $\sigma$ is the variance of CDM perturbations. The
624:  downarrow in the upper-left corner of the panel shows the maximal shift
625:  of the bifurcation mass from $M_{\rm SB}^{\rm max}$ to 
626: $M_{\rm SB}^{\rm min}$ when including dark matter.}
627: \label{fig2}
628: \end{figure}
629: \end{document}
630: 
631: 
632: 
633: 
634: 
635: 
636: