1: \documentclass[12pt,preprint]{aastex}

2: \newcommand{\Mbh}{$M_{\rm BH}$}

3: \newcommand{\ASCA}{{\it ASCA\/}}

4: \newcommand{\Chandra}{{\it Chandra\/}}

5: \newcommand{\ROSAT}{{\it ROSAT\/}}

6: \newcommand{\re}{$R_{\rm e}$}

7: \newcommand{\solarM}{$M_{\odot}$}

8: \newcommand{\solarLB}{$L_{\rm B,\odot}$}

9: \newcommand{\solarMtosolarLB}{$M_{\odot}/L_{\rm B,\odot}$}

10: \newcommand{\solarMtosolarLR}{$M_{\odot}/L_{\rm R,\odot}$}

11: \newcommand{\solarMtosolarLI}{$M_{\odot}/L_{\rm I,\odot}$}

12: \newcommand{\totM}{$M_{\rm tot}$}

13: \newcommand{\starMtoLB}{$M_{\rm star}/L_{\rm B}$}

14: \newcommand{\totMtoLB}{$M_{\rm tot}/L_{\rm B}$} 

15: \newcommand{\totMtoLR}{$M_{\rm tot}/L_{\rm R}$} 

16: 

17: 

18: \begin{document}

19: 

20: \shorttitle{Mass Distributions in NGC 1407}

21: \shortauthors{Zhang et al.}

22: 

23: \title {Probing the Mass Distributions in NGC 1407 and Its Associated Group with the X-ray Imaging Spectroscopic and Optical Photometric and Line-strength Indices Data}

24: 

25: \author{Zhongli Zhang\altaffilmark{1,2},

26:         Haiguang Xu\altaffilmark{1,2},

27:         Yu Wang\altaffilmark{1},

28:         Tao An\altaffilmark{3},

29:         Yueheng Xu\altaffilmark{4}

30:         and Xiang-Ping Wu\altaffilmark{2}}

31: 

32: \altaffiltext{1}{Department of Physics, Shanghai Jiao Tong

33: University, 800 Dongchuan Road, Shanghai 200240, PRC; 

34: e-mail: zebrafish@sjtu.edu.cn, hgxu@sjtu.edu.cn, yuwen\_wang@sjtu.edu.cn}

35: \altaffiltext{2}{National Astronomical Observatories, Chinese

36: Academy of Sciences, 20A Datun Road, Beijing 100012, PRC; 

37: e-mail: wxp@bao.ac.cn}

38: \altaffiltext{3}{Shanghai Astronomical Observatory, Chinese Academy of Sciences, 80 Nandan Road, Shanghai 200030, PRC; 

39: e-mail: antao@shao.ac.cn}

40: \altaffiltext{4}{Department of Physics and Astronomy, University of Leicester, University Road, Leicester LE1 7RH, UK;

41: e-mail: yx12@star.le.ac.uk}

42: 

43: 

44: %###############

45: \begin{abstract}

46: %###############

47: 

48: We present a study of the mass distributions in the bright E0 galaxy NGC 1407 and its associated

49: group by analyzing the high quality \Chandra~and \ROSAT~X-ray spectroscopic data. In order to 

50: probe the stellar mass distribution we calculated the B-band mass-to-light ratio profile by 

51: comparing the observed line-strength indices and multi-color photometric data with different 

52: stellar synthesis model predictions. Based on the recent survey results we also have modeled the 

53: contribution from other group members to the stellar mass. We find that the gas is single-phase 

54: with a temperature of $\simeq0.7$ keV within 1\re~(1\re~= 9.0 kpc), which is typical for elliptical 

55: galaxies. Outside 1\re~the gas temperature increases quickly outwards to $>1$ keV, indicating its 

56: group origin. With the high spatial resolution of \Chandra~we reveal that the X-ray surface 

57: brightness profile shows a central excess in the innermost region, and on both the total mass 

58: and dark matter profiles there is a flattened feature at about $^{<}_{\sim}1$\re, which 

59: coincides with the gas temperature transition from the galactic level to the group level. We 

60: speculate that this may be a mark of the boundary between the galaxy and group halos, as has 

61: been seen in some other cluster/group-dominating galaxies. The total mass and dark matter 

62: distributions within $0.85$\re~are cuspy and can be approximated by power-law profiles 

63: with indices of $\simeq2$, which are marginally consistent with the generalized NFW profiles with

64: $\zeta=2$. The mass in outer regions can be well fitted by a single NFW profile, and the derived 

65: concentration parameter c ($18.6\pm1.5$) is larger than the 68\% upper limit for a halo at $z=0$ 

66: with the given $M_{\rm vir}$. We find that the NGC 1407 group has a baryon-dominated core, while 

67: the mass in the $>1$\re~is dominated by dark matter. At the virial radius $r_{200}=572\pm118$ kpc, 

68: the inferred mass and mass-to-light ratio are 

69: $M_{200}=2.20\pm0.42\times 10^{13}$ \solarM~and

70: $M_{\rm vir}/L_{\rm B}=311\pm60$ \solarMtosolarLB,

71: respectively, showing that the NGC 1407 group is an extremely dark system even comparable to many

72: clusters of galaxies. Since the obtained total mass is lower than those given in the earlier

73: galaxy kinematic works, we speculate that NGC 1400 is not a virialized member in the group's 

74: gravitational potential well.

75: 

76: \end{abstract}

77: 

78: \keywords{galaxies: elliptical and lenticular, cD --- galaxies: individual (NGC 1407) --- X-rays: galaxies --- cosmology: dark matter}

79: 

80: 

81: 

82: %#####################

83: \section{Introduction}

84: %#####################

85: It has been well established that cold dark matter dominates the gravity on large scales 

86: in our universe. This is also true in clusters of galaxies for which observations via galaxy 

87: kinematics, X-ray imaging spectroscopy and lensing techniques have all given unassailable 

88: evidences that as much as about $80-90$\% of the gravitating mass is made of dark matter (see, e.g., 

89: Arnaud 2005 for a recent review). On galactic scales, kinematic studies of spiral galaxies 

90: via measurements of the rotation curves obtained with the radio HI and CO lines, as well as 

91: the optical H$_{\alpha}$ and [OIII] lines also have revealed the existence of a significant 

92: amount of unseen mass whose fraction rises steadily outwards (Sofue \& Rubin 2001). The 

93: mass-to-light ratios deduced from the observed constant rotation curves typically increase 

94: up to a few tens \solarMtosolarLB~within the last observed point, inferring a fraction 

95: of dark matter ranging from about 50\% in the Sa and Sb galaxies to about 90\% in the Sd 

96: and Sm galaxies.  

97: 

98: 

99: Within the frame of the hierarchical models of galaxy formation and the theories of galaxy 

100: mergers, such a remarkable amount of dark mass is also expected in elliptical galaxies, especially 

101: in those luminous ellipticals who are more massive than their spiral counterparts. However, 

102: in elliptical galaxies the stars move in random, usually non-circular orbits, and there is a 

103: lack of obvious rotation velocity tracers like bright HI regions, particularly at large projected 

104: radii ($> 1$\re, where \re~is the effective radius) where the dark matter is considered 

105: to be important. So technically it is hard to model the 3D radial gradients and anisotropies 

106: in the velocity tensor field. In addition, in a bright elliptical galaxy the diffuse gas emission 

107: is usually heavily contaminated by the emission from a large population of low-mass X-ray binaries 

108: (LMXBs). Instruments with high spatial resolution and large effective areas are needed to disentangle 

109: the two components from each other accurately in the X-ray study of the gravitational potential well. 

110: 

111: 

112: Despite of the difficulties, however, some early stellar photometric and kinematic works 

113: (e.g., Saglia et al. 1992; Bertin et al. 1994; Carollo et al. 1995) have indicated that 

114: in some ellipticals there may be as much dark matter as the luminous component within $\sim$\re. 

115: The derived mass-to-light ratios range from a few to a few tens of \solarMtosolarLB~with 

116: a tendency that the dark matter overwhelms the luminous matter at large radii, which supports

117: the possible homology of the galaxy formation processes between the ellipticals and spirals

118: (Bertola et al. 1993). As the absorption line profile measurements in elliptical galaxies improve

119: in recent years, the uncertainties in the rotation and velocity dispersion analyses caused by 

120: the degeneracy between anisotropy and mass distribution have been mitigated. Gerhard et al. (2001) 

121: found that the circular velocity curves (CVCs) of luminous ellipticals are flat to within 10\% 

122: in regions from $^{>}_{\sim}0.2$\re~out to 1-2\re, and they argued that about $10-40\%$ of the 

123: mass is dark within 1\re. In general, these kinematic and dynamic results are backed by the lensing 

124: measurements (Kochanek 1995; Griffiths et al. 1996; Keeton et al. 1998) and the X-ray imaging and 

125: spectroscopic observations (e.g., Forman et al. 1985; Mushotzky et al. 1994; Buote \& Canizares 

126: 1994, 1998; Buote et al. 2002; Loewenstein \& White 1999; Fukazawa et al. 2006; Humphrey et al. 2006b),

127: as well as by the most recent photometric and dynamic analyses of more than 2000 SDSS galaxies 

128: (Lintott et al. 2005). Moreover, the superb high spatial resolution of the \Chandra~X-ray 

129: Observatory also allowed Buote et al. (2002) to examine the position angle twist and ellipticity 

130: of the X-ray image of NGC 720 with high precision. The authors argued that there is a strong 

131: geometric evidence for dark matter, which is independent of the temperature profile and is hard 

132: to be explained by the modified gravity theories as an alternative to dark matter (e.g., Milgrom 1983). 

133:  

134: 

135: So far, however, unlike for the spiral galaxies the observational constraints on the dark 

136: matter content in ellipticals are diverse from case to case and, in some cases, unambiguous or

137: even controversial. For example, in the samples of Saglia et al. (1992), Bertin et al. (1994) 

138: and Gerhard et al. (2001) the kinematic and dynamic evidence for dark matter is found rather 

139: weak in some galaxies, which include the bright cD galaxy NGC 1399. A constant mass-to-light 

140: ratio cannot be ruled out with implication of no or less dark matter. In 2001 Baes \& Dejonghe

141: pointed out that after correcting for the dust absorption and scattering the need of dark matter 

142: in a few \re~may be reduced or eliminated in the kinematic study. Recently, Romanowsky et al. (2003) 

143: reported their analysis of the planetary nebula (PN) kinematics in the E1 galaxy NGC 3379, where 

144: the random stellar velocities are found to decline with radius. The calculated $M/L_{\rm B}$ is 

145: consistent with a constant, which infers a low dark matter fraction ($\leq 0.32$) at 5 \re, 

146: obviously conflicting with the standard galaxy formation theories. In order to respond to this 

147: challenge, Dekel et al. (2005) carried out merger simulations of disk galaxies and argued that 

148: the observed low random stellar velocities may be explained away in terms of the radial orbits 

149: of the halo stars that have been tidally ejected from the inner regions during merger events. 

150: No matter what the final conclusion may be, these examples clearly show the uncertainties in 

151: determining the dark matter content in elliptical galaxies. 

152: 

153: 

154: A correct estimation of the stellar mass is crucial for probing dark matter distribution 

155: in early-type galaxies. In previous works it is sometimes assumed that stellar mass-to-light 

156: ratio is spatially constant throughout the galaxy, and in few cases a certain ``typical value'' 

157: was adopted. However, with the well-calibrated I-band data for 25 E/S0 galaxies Cappellari et al. (2006) 

158: showed that the stellar mass-to-light ratio distributions in the early-type galaxies show 

159: a remarkable variation from source to source by a factor of about 5, with a root mean square 

160: deviation of about 35\% for the sample. This variation is likely to be beyond the uncertainties 

161: induced in the stellar population model, which primarily arise in the line-strength calibration 

162: and the model assumption on the IMF profile and so on. Moreover, in a galaxy there is possibly 

163: a measurable spatial variation of stellar mass-to-light ratio on the $\sim$\re~scales, as is 

164: implied by the measured optical line index gradients (Rampazzo et al. 2005) and optical B-R color 

165: gradients (e.g., Tamura \& Ohta 2003, 2004; Forte et al. 2005). A straightforward example is

166: the kinematic study of 21 round elliptical galaxies by Kronawitter et al. (2000), who found 

167: that the stellar mass-to-light ratio may vary with radius by a factor of up to $\simeq2$ 

168: in a few galaxies. Padmanabhan et al. (2004) also reported a $^{>}_{\sim}30\%$ variation 

169: of the stellar mass-to-light ratio in some cases based on their comprehensive study of SDSS 

170: data. 

171: 

172: 

173: As a step to get more insight into the dark matter distribution in early-type galaxies, in

174: this paper we present a multi-band study of the mass distributions in the bright elliptical 

175: galaxy NGC 1407 (E0, $z$=0.0059) and its associated group by incorporating both the spatially 

176: resolved X-ray spectroscopic data and optical multi-color photometric and line-strength data. 

177: In order to cover as wide a spatial range as possible, meanwhile, without the loss of the high 

178: spatial resolution in the central region, we drew the X-ray data from both the \Chandra~and 

179: \ROSAT~archives to probe the total mass and gas mass distributions, and drew the optical data 

180: from the literatures to calculate the stellar mass profile with a forward method. NGC 1407

181: is a non-cD, dominating galaxy that sits at the bottom of the group's gravitational potential 

182: well. According to the latest survey of Trentham et al. (2006), the group consists of 

183: about 250 member galaxies, of which about 85\% are dwarfs. The projected density of the globular 

184: clusters (GCs) falls off with the same slope as the halo light density of the central galaxy, 

185: indicating that the system may be dark matter-dominated (Perrett et al. 1996). By utilizing 

186: the velocities of 35 galaxies, Trentham et al. (2006) showed that the group has a total mass 

187: of $7\times10^{13}$\solarM~and a R-band mass-to-light ratio of \totMtoLR~=340\solarMtosolarLR. 

188: Note that NGC 1400 (non-barred SA0), the second brightest member that has an unusually large 

189: peculiar velocity of $1072~\rm{km~s^{-1}}$, was included in the calculation. The total mass of 

190: Trentham et al. (2006) is consistent with that of Gould (1993), however, it is larger than t

191: hat given in Quintana et al. (1994) no matter whether NGC 1400 is included or not. 

192: Based on the analysis of surface brightness fluctuations of Tonry et al. (2001), we adopted 

193: a distance to NGC 1407 of $26.8^{+3.4}_{-3.0}$ Mpc, which 

194: is slightly larger than that obtained with

195: $H_{0}=70$ km s$^{-1}$ Mpc$^{-1}$,

196: $\Omega_{m} = 0.3$ and $\Omega_{\Lambda} = 0.7$. 

197: We used the solar abundances of Grevesse \& Sauval (1998), where the iron abundance relative 

198: to hydrogen is $3.16\times10^{-5}$ in number. 

199: 

200: 

201: 

202: %########################################

203: \section{Observations and Data Reduction}

204: %########################################

205: NGC 1407 was observed with the ACIS instrument on board the \Chandra~X-ray observatory on August 26, 

206: 2000 for a total exposure of 50.3 ks. The CCD was operated at a temperature of $-110~^{\circ}{\rm C}$ 

207: with a frame time of 3.2 s. The center of the galaxy was positioned on the ACIS-S3 chip, with an 

208: offset of about $0.63^{\prime}$ from the nominal pointing for the S3 chip, so that most of the X-ray 

209: emissions of the galaxy was covered by the S3 chip. Although the ACIS I2-3, S1-2 and S4 chips were 

210: all in operation, we focused only on the data drawn from the S3 chip in this work. After removing the 

211: contamination of all the visually resolved points sources, we examined both the S1 chip and the 

212: boundary regions of the S3 chip where the emission of the diffuse gas is relatively weak. No background 

213: flare is found during the live operating time, which would otherwise significantly increase the count 

214: rate. We used the CIAO 2.3 software and included only the events with \ASCA~grades 0, 

215: 2, 3, 4, and 6, and removed bad pixels, bad columns and node boundaries. The resulted clean exposure 

216: time is 48.5 ks. In order to constrain on the imaging and spectral properties of the outer regions, 

217: we also have analyzed the \ROSAT~PSPC data of NGC 1407 that was acquired from an observation 

218: performed on August 16, 1995, which lasted for 21.3 ks. We followed the standard \ROSAT~PSPC data 

219: analysis procedures to process the data by using XSELECT v2.3 and FTOOLS v6.0.3. The obtained clean 

220: exposure is 17.7 ks.

221: 

222: 

223: 

224: %###########################################

225: \section{X-Ray Morphology and Point Sources}

226: %###########################################

227: %============================

228: \subsection{X-ray Morphology}

229: %============================

230: In Figure 1a we show the smoothed \Chandra~ACIS S3 image of NGC 1407 in 0.3--10 keV in 

231: logarithmic scale, which is drawn from a 

232: $5^{\prime}{\times}5^{\prime}$ ($38.5\times38.5$ kpc)

233: region that is centered at the X-ray peak (RA=03h40m11.77s Dec=-18d34m49s J2000). 

234: The image has been corrected for exposure, but not for background. We find that the position 

235: of the X-ray peak is consistent with the optical center of the galaxy 

236: (RA=03h40m11.9s Dec=-18d34m49s; 2MASS, Jarrett et al. 2003) 

237: within about $1.8^{\prime\prime}$. The strong diffuse X-ray emission covers the whole S3 

238: field of view, and is roughly symmetric within  1\re~

239: (1\re~$=1.17^{\prime}$ or 9.0 kpc; de Vaucouleurs et al. 1991).

240: In $1-2$ \re, the spatial distribution of diffuse X-ray emission is less symmetric and is 

241: relatively stronger in the east, south and west. We estimate that within $\simeq2$\re~about 

242: 3/4 of the diffuse X-ray has an origin from the hot interstellar medium (ISM), while the rest 

243: is from the unresolved point sources, most of which are LMXBs associated with the galaxy 

244: (Zhang \& Xu, 2004). Many point or point-like sources can be resolved visually in the \Chandra~

245: image, but only about half of them can be detected at the confidence level of $3\sigma$ (\S3.2). 

246: No remarkable irregularities or substructures such as X-ray filaments and cavities can be 

247: identified on the image. 

248: 

249: 

250: In Figure 1b, we show the \ROSAT~PSPC image whose field of view 

251: ($24^{\prime}{\times}24^{\prime}$, or $184.6\times184.6$ kpc) 

252: is much larger than that of the \Chandra~image. The image has been smoothed and exposure-corrected, 

253: too, but has not been corrected for background, either. Besides NGC 1407, the X-ray emissions 

254: associated with the group member NGC 1400 are also clearly detected. Being the second brightest 

255: group member in both optics and X-rays, NGC 1400 has a projected distance of about $11.7^{\prime}$ 

256: (90 kpc) southwest to NGC 1407 and has a much more compact appearance in X-rays. At the north-east 

257: of NGC 1400 there is a relatively weak, diffuse X-ray substructure aligned in the NGC 1400-NGC 1407 

258: direction (marked with a cross on the image), which is not identified on either optical or radio 

259: images. It may be related to a dwarf galaxy in the 

260: 1407 group (Trentham et al. 2006) or may be a background source. Considering the unusually high 

261: peculiar velocity of NGC 1400, it is still possible that the structure is formed due to the gas 

262: stripping by ram pressure. We also identified a weak X-ray source associate with the background SB0 

263: galaxy NGC 1402 ($z=0.014$; Strauss et al. 1992). On the \ROSAT~image there shows some point and 

264: extended sources that have a 0.2--2 keV count flux larger than $0.001~\rm{cts~s^{-1}}$. None of 

265: them can be identified with either a Galactic or extragalactic source based on the currently 

266: available literatures. Because the study of these sources is beyond the scope of this work, we 

267: excluded them in the following imaging and spectral analyses. X-ray emissions from other bright 

268: group members are not detected.

269: 

270: 

271: Both \Chandra~and \ROSAT~images reveal that NGC 1407 and its host group are relaxed and are 

272: rich in hot gas. A detailed study of the \ROSAT~X-ray surface brightness profiles 

273: indicates that the X-ray gas extends outwards to at least $12^{\prime}$ (10.3 \re; \S5.1). The 

274: gas emission is soft, as is shown in Figure 1c where the smoothed 2-dimensional hardness ratio (HR) 

275: distribution calculated with the \Chandra~data is plotted after removing all the resolved point sources. 

276: The hardness ratio is defined as HR=(H-S)/(H+S), where S and H are the exposure- and background-corrected 

277: counts in 0.3--2 keV and 2--10 keV, respectively. The emission is very soft (HR=$-0.94 \sim -0.50$) 

278: within 1\re, which corresponds to a thermalized plasma at about $0.6-1.0$ keV. Outside 2\re~the 

279: hardness ratio increases to higher values that is typical for group gas. At about $1.8^{\prime}$ 

280: (13.8 kpc) east of the galactic center there is a relatively cold (HR=-0.7) region. 

281: 

282: 

283: In Figure 1d we show the 1.43 GHz radio map of a $24^{\prime}\times24^{\prime}$ region around NGC 1407, 

284: which is obtained from the National Radio Astronomical Observatory (NRAO) archive. We identified NGC 1407 

285: as an extended source that has a flux density of $61.0\pm1.5$ mJy/b. The emission of the galaxy is 

286: asymmetric and extends farther towards west. A weak radio component, whose flux density is $9.6\pm2.2$ mJy/b 

287: ($\sim3\sigma$), is detected at about $6^{\prime}$ southwest of NGC 1407. Also there is a bridge-like feature 

288: linking between the weak component and NGC 1407. In the 1.43GHz image we failed to identify the second 

289: brightest group member NGC 1400. However, this galaxy can be identified in the 0.3 GHz and 8.5 GHz maps 

290: (not shown here) with flux densities of $64.0\pm20.0$ mJy/b and $8.9\pm0.4$ mJy/b, respectively.

291: The lack of apparent substructures in the radio band such as remarkable jets and lobes strengthens our 

292: conclusion that the X-ray gas in NGC 1407 is relaxed."

293: 

294: 

295: %===============================

296: \subsection{X-Ray Point Sources}

297: %===============================

298: We detected X-ray point sources on the \Chandra~S3 image in 0.3--10 keV by using the CIAO 

299: tool {\it celldetect\/} with a threshold of $3\sigma$, and then crosschecked the results in 

300: 0.7--7 keV and 2--10 keV separately by using both a wavelet arithmetic and visual inspection. 

301: After removing insignificant and fake detections, we totally detected 41 point sources within 

302: the central $\simeq2$\re, of which up to about 4 are probably unrelated background 

303: sources according to a Monte-Carlo simulation test based on the deep {\it Chandra} observations 

304: of blank fields (Mushotzky et al. 2000). We find that none of the resolved sources is identified 

305: with an optical counterpart. The spatial distribution of the point sources is nearly symmetric, 

306: which shows a clear concentration towards the galactic center. The source number density follows 

307: the B- and R-band optical light distributions approximately, indicating that most of them are 

308: intrinsically associated with the galaxy. In our previous work (Zhang \& Xu 2004) we studied 

309: the X-ray colors of these sources and their contribution to the total X-ray luminosity of the 

310: galaxy. We found that within 2\re~in 0.3--10 keV the resolved point sources account 

311: for about 18\% of the total emission; if the unresolved sources are taken into account, about 39\% 

312: of the total emission in 0.3--10 keV can be ascribed to LMXBs. The combined spectrum of all resolved 

313: off-center X-ray point sources can be well fitted with a single power-law model 

314: ($\Gamma=1.59\pm0.09$) 

315: when the absorption is fixed to the Galactic value ($5.42\times10^{20}$ cm$^{-2}$; 

316: Dickey \& Lockman, 1990).

317:  

318: 

319: The brightest point source (Src 1) is detected at the X-ray peak of NGC 1407 and has a total 

320: net count of $364.4\pm21.0$ cts, or an average count rate of 

321: $75.1\pm4.3\times10^{-4}~{\rm cts~s^{-1}}$

322: in 0.3--10 keV. Previous to the \Chandra~observation, there is no report about the detection 

323: of this source in X-rays. We have attempted to detect the possible temporal variabilities of 

324: this source on different time scales. In terms of the K-S test, the temporal variability of the

325: source is less significant, possibly because K-S tests are most sensitive around the median 

326: value of the independent variable. However, the visual inspection of the lightcurve suggests 

327: that it is variable on timescales of 2.5-3 hr during which its count rate changed significantly 

328: by about $30-60\%$ (Fig. 2), which is supported by the calculated variability parameter 

329: $S=(f_{\rm max}-f_{\rm min})/\sqrt{(\sigma_{f_{\rm max}}^{2}+\sigma_{f_{\rm min}}^{2})}=1.9$,

330: where $f_{\rm max}$ and $f_{\rm min}$ are the maximum and minimum fluxes, respectively, and

331: $\sigma_{f_{\rm max}}$ and $\sigma_{f_{\rm min}}$ are corresponding errors. We fitted the ACIS spectrum 

332: of the central source with an absorbed power-law model. To avoid the relatively poor calibration 

333: at low energies and instrumental background at high energies, we limited the fittings to 0.7--7 keV. 

334: The absorption was fixed to the Galactic value since allowing it to vary did not improve the fit. 

335: The fit is acceptable ($\chi^2_r$=10.2/11) and the best-fit photon index is $2.29^{+0.24}_{-0.23}$ 

336: (Table 1), which is steeper than that of the accumulative spectrum of the off-center sources. We 

337: also have attempted to fit the spectrum of Src 1 with an absorbed multi-color disk blackbody model. 

338: However, the fitting is rejected with large residuals in both the low energy and high energy ends 

339: ($\chi^2_r$=27.8/11). We estimate that the 0.3--10 keV luminosity of Src 1 is

340: $4.07\pm0.23\times10^{39}~\rm{erg~s^{-1}}$. 

341: This value is higher than that of the central sources in NGC 4472 and NGC 4649 (Soldatenkov et al. 2003) 

342: by one order of magnitude. Thus, we speculate that the central source may be an embedded low 

343: luminosity AGN because of its central location, rapid temporal variations, strong X-ray emissions 

344: and the appearance of a weak jet-like radio structure associated with it. If the source is an AGN, 

345: by using the tight relation between the black hole mass and the galaxy central velocity dispersion

346: $M_{\rm BH}=1.30(\pm0.36)\times10^{8}M_{\odot}(\sigma/200~\rm{km~s^{-1}})^{4.72(\pm0.36)}$ 

347: as was given in Merritt \& Ferrarese (2001), and the central velocity dispersion of $\sigma=272$ km s$^{-1}$ 

348: that is taken from McElroy (1995), we estimate the mass of the black hole as 

349: $5.55^{+2.37}_{-1.96}\times10^{8}$ \solarM. 

350: Quite similar results can be obtained if the \Mbh-$\sigma$ relation derived alternatively by 

351: Wyithe (2006) for a local sample of 31 galaxies and the central velocity dispersion of Bernardi et al.  

352: (2002; 279 km s$^{-1}$) are adopted instead.

353: 

354: 

355: 

356: %##########################

357: \section{Spectral Analyses}

358: %##########################

359: %======================

360: \subsection{Background}

361: %======================

362: In the \Chandra~spectral analysis of the ACIS S3 data the usual way to obtain a background 

363: spectrum, which primarily consists of the cosmic, instrumental and non X-ray particle components, 

364: is to utilize the observation dataset directly by extracting the background spectrum from a 

365: local, uncontaminated region on the S3 chip that is located far away enough from the source. 

366: Alternatively we may draw the background spectrum from the \Chandra~blank field datasets in 

367: the same detector region where the source spectrum is extracted. To avoid 1) the field-to-field 

368: variations in the cosmic background, which is typically of the order of a few tens percent 

369: on arcmin scales and even more remarkable on larger scales, and 2) the time-dependent variations 

370: in the particle and instrumental backgrounds that may be significant in some observations, the 

371: use of former method is more preferred in practice whenever it is applicable. 

372: 

373: 

374: NGC 1407 is dominating a nearby group that is rich in X-ray gas. The diffuse X-ray emission

375: of the group spans over a wide region that covers the whole \Chandra~field of view. As we 

376: will show in \S5.1, the diffuse soft emission of the gas extends outwards to about 10.3 \re~($12^{\prime}$). 

377: At the edge of ACIS S3 chip, the emission from the hot gas is estimated to account for about 

378: $10-20\%$ of the total flux in 0.7--7 keV. Thus we are precluded from extracting a clean 

379: background spectrum directly in the CCD boundary regions. For the spectral fittings of the 

380: inner regions this will not be a problem; the X-ray background therein is overwhelmed by the 

381: emission of the galaxy in the selected bandpass, so even the \Chandra~blank field spectra 

382: can be applied as a good approximation of the real background if its high energy end, where the 

383: instrumental effects dominate, is well normalized to the observation. However, for the outer 

384: regions the use of the blank field spectra may result in severe misestimates of the model 

385: parameters due to the spatial and temporal variations of the different background components 

386: that may have changed their relative strengths. In fact, when we examined the spectrum 

387: extracted from the source-free region on the BI chip S1, we found that there is an extra 

388: hard component as compared with the blank field spectrum. Such a hard excess is also seen 

389: in the spectrum extracted in the boundary region of the S3 chip in 3--10 keV, where the gas

390: emission is relatively weak. The extra hard component found on the S1 and S3 chips are quite 

391: similar in spectral energy distribution, which can both be described by a Power-law model with 

392: $\Gamma=1.0$. The nature of this extra hard component is unclear. Possibly it is a feature 

393: of the local cosmic background, or it is induced by particle events. Therefore, the use 

394: of the blank field spectra should be rejected in our analysis.

395: 

396: 

397: In order to construct the \Chandra~S3 background spectrum, we extracted the spectrum of a 

398: $7.5^{\prime} \times 0.4^{\prime}$ region at the boundary of the S3 chip and then subtracted 

399: the contaminating soft emission component contributed by the hot diffuse gas from it. The 

400: selected region is located at about $5^{\prime}$ away from the galactic center, which is well 

401: beyond the outermost annulus used to extract the source spectra. As will be shown in \S4.2  , 

402: the contribution of LMXBs is negligible. We modeled the gas emission with an 

403: absorbed APEC component, whose temperature ($kT$=1.30 keV) and abundance ($Z$=0.21 solar) are obtained 

404: by studying the \ROSAT~PSPC spectrum extracted from the same region. Since in the \ROSAT~fittings the 

405: normalization of the APEC component is not well constrained, we deduced it from the modeling of the PSPC 

406: surface brightness profile (\S5.1). We find that in the selected background region the obtained ACIS S3 

407: background accounts for $82^{+5}_{-2}\%$ of the total counts in 0.7--7 keV and has a spectral 

408: energy distribution consistent with that obtained for the source-free region on the ACIS S1 

409: chip. The calculated background level also agrees with that derived in the analysis of the 

410: X-ray surface brightness profile (\S5.1) and that used in Humphrey et al. (2006b). In 0.2--2 keV 

411: where the particle and instrument components are minor, the deduced background count rate 

412: is close the average X-ray background count rate from the ROSAT All-Sky Survey diffuse background 

413: maps, which is available at http://heasarc.gsfc.nasa.gov. 

414: 

415: 

416: For the \ROSAT~PSPC spectral analysis, we extracted the background spectrum from a region about 

417: $13^{\prime}$ away from the galactic center, where the 0.2--2 keV surface brightness (\S5.1) is 

418: less than $3\sigma$ above the mean local PSPC background level 

419: ($7.26\times10^{-6}$ cts cm$^{-2}$ s$^{-1}$ arcmin$^{-2}$).

420:  

421: 

422: %==============================================

423: \subsection{Projected and Deprojected Analysis}

424: %==============================================

425: We performed both the projected and deprojected spectral analyses of the \Chandra~and \ROSAT~

426: spectra by using a model that consists of an APEC and power-law component, both subjected to a 

427: common absorption. In the deprojected analysis, we used the XSPEC model ``projct'' to evaluate 

428: the influence of the outer spherical shells onto the inner ones. In the \Chandra~spectral 

429: fittings, we extracted the ACIS S3 spectra from 4 annular regions that are centered at the X-ray 

430: peak. We limited the energy range to 0.7--7 keV to minimize the effects of both the instrumental 

431: background at higher energies and the calibration uncertainties at lower energies. By applying 

432: the latest CALDB we also have corrected for charge transfer inefficiency (CTI) and continuous 

433: degradation in the ACIS quantum efficiency, which is especially severe at low energies. To 

434: compensate for the degradation of the ACIS energy resolution we included an additional 5\% 

435: systematic error.  We first restricted the fittings to 3.5--7 keV, where the gas emission 

436: is typically less than $5\%$ and thus can be ignored. We found that an absorbed power-law model 

437: can give an acceptable fit to the spectra, and the obtained absorptions and photon indices are

438: consistent with the Galactic value and photon index for the resolved off-center point sources 

439: ($\Gamma=1.59$; \S3.2), respectively. Thus, to achieve a better statistics in the fittings of

440: the 0.7--7 keV spectra, we fixed the photon indices to 1.59 and adopted the normalizations of 

441: the power-law components as have been obtained in 3.5--7 keV. Again, we fixed the absorption 

442: to the Galactic value because if we allow it to vary, the fitting did not improve. The best-fits 

443: are shown in Table 1. We see that the deprojected analysis gives acceptable fits, while the 

444: projected fits are poor for the inner three annuli. The residual distributions indicate a 

445: second thermal component needed for these annuli, which is most probably caused by the 

446: projection effect. 

447: 

448:  

449: We extracted the \ROSAT~PSPC spectra in 0.2--2 keV from 8 annular regions, with the 4 inner annuli 

450: chosen as the same as for the \Chandra~analysis and the outer 4 annuli spanning over $2.1-10.3$ \re. 

451: We find that the soft emission of the gas always dominates in the spectra. For the inner 4 regions, 

452: we estimate that the hard emission component, which is mostly from LMXBs, account for up to $\simeq20\%$ 

453: of the 0.2--2 keV flux. Technically it is difficult to disentangle such a weak, hard component from 

454: the dominating soft emission component precisely in the limited bandpass of PSPC. Therefore when we 

455: fitted the \ROSAT~spectra of the inner 4 regions we always fixed the ratio of hard component to soft 

456: component referring to the \Chandra~results. For the outer 4 annuli, there is evidence that the 

457: relative contribution of the hard component is even lower. Forbes et al. (2006) reported that in NGC 1407 

458: the globular cluster (GC) density decreases exponentially outside 1.4 \re, with only about 

459: $\sim20\%$ of the total $\simeq1300$ GCs distributed in $r>2.1$\re. Since in the non-cD early-type 

460: galaxies, typically a few tens percent of the LMXBs are found in GCs (Xu et al. 2005 and references 

461: therein), the low GC density in the outer regions suggests a significant lack of LMXBs in the same 

462: regions. Therefore we fitted the \ROSAT~spectra of the outer 4 regions only with an absorbed APEC 

463: component. Since the abundances of the outermost three annuli are poorly determined, we tied them 

464: together during both the projected and deprojected fittings. For the same reason we tied the temperatures 

465: of the fifth and the sixth annuli together for the deprojected fitting. The fittings are marginally acceptable or poor for the 

466: projected, and are acceptable for the deprojected analyses (Table 1). 

467: 

468: 

469: %=================================================

470: \subsection{3D Temperature and Abundance Profiles}

471: %=================================================

472: In Figure 3, we illustrate the three-dimentional gas temperature and abundance distributions

473: based on our deprojected spectral analysis. Within 2.1 \re~where both the \Chandra~

474: and \ROSAT~data are available, the gas temperature and abundances obtained with the \Chandra~

475: and \ROSAT~agree very well with each other within the errors, except that in $0.53-2.1$\re~

476: the \ROSAT~temperature is lower. In the analyses that follows, we adopt the \Chandra~best-fits 

477: for the inner 4 annuli and the \ROSAT~best-fits for outer ones. We noticed that within 1\re, 

478: the gas temperature is below 1 keV but still higher than 0.6 keV, which is typical 

479: for bright early-type galaxies. The gas temperature increases steadily outwards, reaches a maximum 

480: value of 

481: $1.37\pm0.35$ keV in $2.1-5.1$\re, 

482: and then drops to about 1.1 keV in the outermost annulus. The high temperature in $^{>}_{\sim}1$\re~

483: suggests the group origin of the gas. We have attempted to model the temperature profile with 

484: an analytical form, and found that

485: \begin{equation}\label{eq:temp_model} 

486: kT=\left\{\begin{array}{ll}

487: c_0 +\frac{c_1(\frac{r}{r_{\rm c}})^{\eta}}{1 +(\frac{r}{r_{\rm c}})^{\eta}} & \mbox{$r<3.4$\re} \\

488: \\

489: c_0 +\frac{c_1(\frac{34-r}{r_{\rm c}})^{\eta}}{1 +(\frac{34-r}{r_{\rm c}})^{\eta}} & \mbox{$r>3.4$\re} 

490: \end{array}\right.

491: \end{equation}

492: (Pratt \& Arnaud 2003) can best describe the observed temperature profile. The parameters are

493: $c_0=0.66$ keV, 

494: $c_1 =0.71$ keV, 

495: $r_c = 0.94$ \re~and 

496: $\eta=3.78$ 

497: for $r<3.4$\re, and 

498: $c_0=0.74$ keV, 

499: $c_1 =0.70$ keV, 

500: $r_c = 25.5$\re~and 

501: $\eta=11.22$ 

502: for the outer regions.

503: The abundance at the center of the galaxy is about 0.56 solar. It increases to about 1.40 solar 

504: at $\sim0.85$\re, and then decreases rapidly to 0.21 solar in $^{>}_{\sim}3.4$\re. We 

505: approximate the abundance distribution in the analytical form as 

506: \begin{equation}\label{eq:abun_model} 

507: Z = c_0 +c_1 \times {\rm exp}(\frac{r-r_{0}}{\sigma})^{-2},

508: \end{equation}

509: where $r_{0}=0.7$\re. Within 0.7 \re, we obtain 

510: $c_0=0.28$ solar, 

511: $c_1 =1.06$ solar and 

512: $\sigma =0.58$. 

513: In outer regions, the parameters are

514: $c_0=0.21$ solar, 

515: $c_1=1.13$ solar and 

516: $\sigma=1.12$. 

517: In general, our results agree with those of Matsushita (2001), Fukazawa et al. (2006) 

518: and Humphrey et al. (2006a,b) very well, although different annular-binnings were used in these works.

519: 

520: 

521: 

522: %###########################

523: \section{Mass Distributions}

524: %###########################

525: %==================================================================

526: \subsection{X-ray Surface Brightness Profiles and Gas Distribution}

527: %==================================================================

528: In Figure 4 we show the X-ray surface brightness profiles $S(r)$ of our target obtained with 

529: \Chandra~ACIS in 0.7--7 keV and \ROSAT~PSPC in 0.2--2 keV, which have been corrected for 

530: exposure but not for background. In order to improve the fitting statistics and to model the 

531: gas distribution over as wide a region as possible, we fitted the two profiles jointly. By 

532: assuming hydrodynamic equilibrium, spherical symmetry, and that the gas is ideal and thermalized, 

533: we integrated the gas emission along the line of sight by using

534: \begin{equation}

535: S(r) = S_{0} \int^{\infty}_{r} \Lambda (T,Z) n_{g}^{2}(R) \frac{RdR}{\sqrt{R^{2}-r^{2}}}+S_{\rm bkg},

536: \end{equation}

537: where $R$ is the 3-dimensional radius, $\Lambda(T,Z)$ is the cooling function calculated by 

538: taking into account both the temperature and metallicity gradients as are given by the best-fit 

539: deprojected spectral model in \S4.3, $n_{g}(R)$ is the gas density, and $S_{\rm bkg}$ is the 

540: background. For $n_{g}(R)$ we have attempted to apply either the $\beta$ model, which takes 

541: the form of

542: \begin{equation}

543: n_{g}(R) = n_{g,0} \left[ 1 + (R/R_{c})^{2} \right] ^{-3\beta/2},

544: \end{equation} 

545: or the two-$\beta$ model 

546: \begin{equation}

547: n_{g}(R) = \left\{

548:            n_{g,1}^{2} \left[ 1 + (R/R_{c1})^{2} \right] ^{-3\beta_{1}}

549:          + n_{g,2}^{2} \left[ 1 + (R/R_{c2})^{2} \right] ^{-3\beta_{2}}

550:            \right\} ^{1/2},

551: \end{equation}

552: where $R_{c}$ is the core radius, $\beta$ is the slope, and suffixes 1 and 2 are used to represent

553: the two emission components. For inner regions, we have removed the contributions from both resolved 

554: and unresolved point sources based on our \Chandra~spectral analysis (\S4.2). For the $>2.1$\re~ 

555: regions where the \Chandra~spectral information is not available, we estimate that the hard X-rays 

556: from the point sources accounts for up to about 5\% of the total flux in the \ROSAT~PSPC spectra, 

557: so we simply ignored this component. We also eliminated the faint X-ray substructure located between 

558: NGC 1407 and NGC 1400 (\S3.1). 

559: 

560: 

561: We find that the fit with the $\beta$ distribution should be rejected on the 90\% confidence 

562: level ($\chi^2_r$=167.9/60), which shows significant residuals in the central $0.2$\re~for the \Chandra~

563: Surface brightness profile that infers the existence of a central excess (Figure 4a). The two-$\beta$ model, 

564: on the other hand, gives acceptable fit to the data ($\chi^2_r$=68.9/56; Fig. 4) with $R_{c1}=0.12\pm0.01$\re~

565: and $\beta_{1}=0.70\pm0.01$ for one component, and $R_{c2}=0.83\pm0.01$\re~and 

566: $\beta_{2}=0.45\pm0.01$ for another. Using the best-fit gas distribution we find that the gas density 

567: at the galactic center and at 1\re~are $\sim0.1~\rm{cm^{-3}}$ and $\sim0.002$ cm$^{-3}$, respectively. 

568: The gas density decreases to $\simeq10^{-4}$ cm$^{-3}$ at about $10.3$\re~where the count 

569: rate is about $3\sigma$ above the mean background level. The best-fit \Chandra~background level 

570: is $1.0\times10^{-5}$ cts s$^{-1}$ cm$^{-2}$ arcmin$^{-2}$ and is in excellent agreement with 

571: what we have adopted in the spectral analysis (\S4). We also notice that even if the \ROSAT~data 

572: of the inner regions are excluded from the fittings, the best-fit parameters do not change 

573: significantly at the 90\% confidence level. This is not surprising because central excess 

574: emission is essentially restricted in a region whose size is similar to the \ROSAT~PSPC PSF.

575: 

576: 

577: %========================

578: \subsection{Stellar Mass}

579: %========================

580: \noindent{\it NGC 1407}\\

581: \indent We first employed the PEGASA-HR code (Fioc \& Rocca-Volmerange 1997; Le Borgne et al. 2004) 

582: to deduce the stellar mass-to-light ratio \starMtoLB~in NGC 1407. By assuming the solar abundance 

583: ratios at the inferred age and metallicity, and utilizing an empirical stellar library that 

584: includes both the thermally pulsating asymptotic giant branch stars (AGB) and post-AGB phases, 

585: the code has been designed to degenerate the effects of age and metallicity by creating high 

586: resolution synthetic spectral energy distributions (R=10 000) that can be compared with the 

587: observed line-strength indices and photometric data. We quoted the B- and R-band photometric 

588: data of NGC 1407 from de Carvalho et al. (1991), both of which exhibit similar profiles to those 

589: provided earlier by Lauer (1985) and Franx et al. (1989), and drew the K-band photometric data from 

590: Jarrett et al. (2003). To calibrate the data, we normalized the B-, R- and K-band magnitudes 

591: within 1\re~to 

592: $\mu_B=21.79$ mag arcsec$^{-2}$, 

593: $\mu_R=20.24$ mag arcsec$^{-2}$ and 

594: $\mu_K=19.09$ mag arcsec$^{-2}$, 

595: respectively (Prugniel \& Simien 1996), and then corrected the data for 

596: Galactic extinctions (Schlegel et al. 1998). Like in many other elliptical galaxies (e.g., Burkert 1993), 

597: the B- and R- band surface brightness profile $S(r)$ in 0.1--0.7\re~and the overall K-band 

598: $S(r)$ can be described with the $1/4$ law 

599: \begin{equation}

600: I(R) = 8.0 S_{\rm e} 10^{-3.331 [(R/R_{\rm e})^{1/4}-1]}, 

601: \end{equation}    

602: where $S_{\rm e}=6.34\times10^{-11}~\rm{erg~s^{-1}~cm^{-2}~arcmin^{-2}}$ (B-band),  

603: $1.13\times10^{-10}~\rm{erg~s^{-1}~cm^{-2}~arcmin^{-2}}$ (R-band) and

604: $3.12\times10^{-10}~\rm{erg~s^{-1}~cm^{-2}~arcmin^{-2}}$ (K-band).

605: In the innermost region, however, both the B- and R-band $S(r)$ shows 

606: a clear excess beyond the $1/4$ law, which can be characterized by a ``Nuker'' model (Faber et al. 1997) 

607: \begin{equation}

608: I(r) = 2^{(b-c)/a} I(r_{\rm br})(r_{\rm br}/r)^{c}[1+(r/r_{\rm br})^{a}]^{(c-b)/a},

609: \end{equation}

610: where $a$ parameterizes the sharpness of the break, 

611: $b$ is the asymptotic outer slope, 

612: $c$ is the asymptotic logarithmic slope inside $r_{\rm br}$, 

613: and $r_{\rm br}$ is the break radius. We find that in the B-band 

614: $a=2.20$, $b=1.67$, $c=0.05$ and $r_{\rm br}=0.05$\re,

615: and in the R-band

616: $a=2.20$, $b=1.72$, $c=0.05$ and $r_{\rm br}=0.05$\re.

617: By modeling the observed surface brightness profile we calculated that the B-band luminosity 

618: within 1\re~is $3.5\times10^{10}$ \solarLB. 

619: We adopted the measured line indices in four luminosity weighted annuli ($0-1/16$\re, $1/16-1/8$\re, 

620: $1/8-1/4$\re~ and $1/4-1/2$\re), which include the emission- and velocity-corrected H$\beta$, 

621: Mgb, Fe5015, Fe5270 and Fe5335 indices (Table 2). Using the method described in \S4 of Rampazzo et al. (2005), 

622: the indices have been transformed into the Lick-IDS system, which is originally designed to 

623: describe the spectral absorption features by including a set of atomic and molecular bands, 

624: each consisting of a central bandpass and two pseudocontinuum bands flanked to the red and 

625: blue sides (more details about the Lick-IDS system is described in Trager et al. (1998)).  

626: The H$\beta$ line index was directly corrected by comparing with the H$\alpha$ lines other 

627: than by comparing with the [O III] lines, since the H$\beta$/[O III] correlation has a relatively 

628: larger dispersion that varies between 0.33 and 1.25 (Trager et al. 2000; Denicol\'{o} et al. 2005).

629:  

630: 

631: In the calculations we assumed that 1) the stellar population is coeval, i.e., the line-strength 

632: indices and color gradients are mostly due to the metallicity gradient (Michard 2005), 2) the star 

633: formation rate can be approximated by an exponentially decreasing form 

634: SFR(t)$\propto e^{-t/p}$, 

635: where $p$ is the decay time in Myr, and 3) the initial mass function (IMF) takes the form of 

636: either the Salpeter profile (Salpeter 1955) or the Kroupa profile (Kroupa 2001) in the mass range 

637: $0.1-120$\solarM. After considering the effects of dust, galactic wind, as well as the nebular 

638: emission, we compared the model-predicted line-strength indices, B-R and B-K colors with the 

639: observed data (Table 2). We found that magnesium is always underestimated by the model, which confirms the 

640: results of Humphrey \& Buote (2006a), Howell (2005) and Thomas et al. (2005) that NGC 1407 is an 

641: $\alpha$-enriched system. In order to reconcile the assumption of the solar abundance ratios 

642: in the PEGASE code with the $\alpha$ overabundance, we alternatively employed a combinative 

643: index MgFe as a tracer of total stellar metallicity, which is defined either as  

644: ${\rm [MgFe]_{T}}=\sqrt{{\rm Mgb}(0.72\times{\rm Fe}5270+0.28\times{\rm Fe}5335)}$ (Thomas et al. 2003)

645: or as

646: ${\rm [MgFe]_{C}}=(0.45Mgb+Fe5015)/2$ (Cappellari et al. 2006).

647: A recent study with excellent dynamical models and high quality spectra from the SAURON panoramic 

648: integral-field spectrograph by Cappellari et al. (2006) shows that the \starMtoLB~calculated 

649: with the so-defined MgFe index is weakly affected by non-solar $\alpha$/Fe. This is also shown

650: in Table 3 where the deduced age and metallicity of the stellar population of NGC 1407 in all 

651: annuli are insensitive to the choice of IMF and MgFe definations. Within the errors the stellar 

652: age and metallicity deduced with different combinations of IMF and MgFe agree well with each 

653: other. In all cases the derived stellar age is close to 10 Gyr, which is consistent with some 

654: previous works ($7.4\pm1.8$ Gyr, Thomas et al. 2005; $9.5\pm2.2$ Gyr, Howell 2005), and the 

655: metallicity always reaches its maximum value in $1/8-1/4$\re~and then decreases to a lower value 

656: in $1/4-1/2$\re. The \starMtoLB~deduced with the Slapeter IMF, on the other hand, is apparently 

657: larger than those deduced with the Kroupa IMF (Table 3) at the 90\% confidence level (see \S6.1

658: for a detailed discussion about the errors). We notice that the model predictions with the 

659: Salpeter IMF fit the observed data significantly worse than those with the Kroupa IMF. This 

660: may be caused by the fact that the Salpeter IMF is too rich in low-mass stars (e.g., 

661: Weiner et al. 2001). If a diet Salpeter IMF (Bell et al. 2003) is adopted instead, the 

662: \starMtoLB~values may be reduced by $\sim30\%$, which is in excellent agreement with the 

663: values deduced with the Kroupa IMF. 

664: 

665: 

666: For the time being the line index measurements are not available for the regions outside 0.5\re.

667: Considering that both the 2MASS H-K and J-K colors keep approximately invariant within the 

668: error in 0.5-2\re~(Jarrett et al. 2003), which infers that the stellar mass-to-light ratio is 

669: roughly constant in the same region, we fixed the stellar mass-to-light ratio in $r>0.5$\re~

670: to 4.45 \solarMtosolarLB~as is derived in 0.25-0.5\re. 

671: 

672: 

673: \noindent{\it Other group members}\\

674: \indent More than 250 galaxies have been identified in the NGC 1407 group, of which about 85\% 

675: are dwarfs. By referring to the data in Trentham et al. (2006) we find that the 2-dimensional galaxy 

676: number density roughly follows a power-law profile with an index of $\sim-2/3$. Utilizing the

677: mean B-band stellar mass-to-light ratio as have been obtained for NGC 1407, we modeled the 

678: stellar mass distribution in $<10.3$\re. We find that, excluding NGC 1407, all other member 

679: galaxies contribute about 25\% of the total stellar mass, or $\simeq$1.7\% of the total mass 

680: within 10.3 \re.

681: 

682: 

683: %==================================

684: \subsection{Total Gravitating Mass}

685: %==================================

686: Assuming that the gas is ideal and the group is in the hydrostatic equilibrium state, 

687: we calculate the gravitating mass distribution by using 

688: \begin{equation}

689: M_{\rm tot}(R) = -\frac{kTR}{G\mu m_{p}}\left(\frac{d\rm ln \it n_{g}}{d\rm ln \it R}

690: +\frac{d\rm ln \it T}{d\rm ln \it R}\right),

691:  \end{equation}

692: where $M(R)$ is the total mass within the radius $R$, $n_{g}$ is the gas density, G 

693: is the gravitational constant, $m_{p}$ is the proton mass and $\mu$ is the mean molecular weight. 

694: In the calculation we adopt the best-fit gas density profile obtained in the joint

695: \Chandra~and \ROSAT~fittings of the X-ray surface brightness profiles (\S5.1, Fig. 4) and the 

696: best-fit gas temperature profile (\S4.3) obtained in the spectral modelings (\S4.2). The resulting 

697: gravitating mass distribution is shown in Figure 5a, along with the 90\% errors determined by 

698: performing Monte-Carlo simulations that account for the ranges of temperature and gas density 

699: allowed by the data. The derived total mass of NGC 1407 is 

700: $2.75^{+0.47}_{-0.56}\times10^{11}$ \solarM~at 1\re~

701: and 

702: $6.6^{+1.2}_{-4.1}\times10^{12}$ \solarM~at 10.3 \re, the gas halo boundary. 

703: We noticed that at $R \approx 0.85$ \re~there is a clear change in the slope of $M(R)$, exhibiting 

704: a shoulder-like structure that has been observed in galaxy clusters  (e.g., Abell 1795, 

705: Xu et al. 1998 and Ettori et al. 2002) and other groups (e.g., the NGC 1600 group, Sivakoff et al. 2004). 

706: In Figure 5b we show the mass-to-light ratio \totMtoLB~profile. We find that at $^{<}_{\sim}0.85$\re~the 

707: lower and upper limits of the ratio are $\simeq5$ and $\simeq10$\solarMtosolarLB, respectively. 

708: At the boundary of optical envelope ($\simeq5.1$\re) 

709: \totMtoLB~is about 50\solarMtosolarLB, indicating the existence of a large amount of dark 

710: matter. At the gas halo boundary \totMtoLB~increases to over 90\solarMtosolarLB. 

711: 

712: 

713: Due to the appearance of the baryonic component that dominates in the inner region, the total 

714: gravitating mass profiles of early-type galaxies are usually inconsistent with the dark halo 

715: profiles given in Navarro et al. (1997; the NFW profile) and Moore et al. (1999; the generalized 

716: NFW profile), which can be expressed as

717: \begin{eqnarray}

718: \rho(R)&  = & \frac{\rho_c(z)\delta_c}{(R/R_s)^{\zeta}(1+R/R_s)^{3-\zeta}}, 

719: \label{eqn.generalized_nfw}

720: \end{eqnarray}

721: where 

722: $\zeta<2$ ($\zeta=1$ for the NFW profile), 

723: $\rho_c(z) = 3H(z)^2/8\pi G$ is the critical density of the universe at the redshift $z$, $R_s$ is 

724: the scale radius, and $\delta_c$ is a characteristic dimensionless density; when expressed in 

725: terms of the concentration parameter $c$, $\delta_c$ takes the form

726: \begin{equation}

727: \delta_c = \frac{200}{3}\frac{c^3}{\ln(1+c) - c/(1+c)}. \label{eqn.conc}

728: \end{equation}

729: In NGC 1407 this phenomenon is also observed. The total mass profile of NGC 1407 in $>0.85$\re, 

730: however, can be well fitted with the NFW model. The obtained best-fit parameters are 

731: $r_s=3.4\pm0.4$\re~and $c=18.6\pm1.5$. 

732: The inferred virial radius $r_{200}$, where the average mass density is 200 times the current 

733: critical density of the universe, is $r_{200}=572\pm118$ kpc. At $r_{200}$ the inferred virial 

734: mass and mass-to-light ratio are  

735: $M_{200}=2.20\pm0.42\times 10^{13}$ \solarM~

736: and

737: $M_{\rm vir}/L_{\rm B}=311\pm60$ \solarMtosolarLB, respectively.

738: 

739: 

740: %=======================

741: \subsection{Dark Matter}

742: %=======================

743: By subtracting the contributions of the black hole (\S3.2), stars (\S5.2) and gas (\S5.1) from 

744: the total mass, we calculated the dark matter distribution in NGC 1407 and its associated group.

745: We plot the dark mass profile in Figure 5c in which a slope change at $\sim0.85$\re~can be 

746: clearly seen. Neither the NFW profile nor the generalized NFW profiles can give an accepted fit 

747: to this abnormal profile, similar to what we have found for the total mass profile. We also have 

748: tentatively tested the model consisting of two mass components, where each component takes the 

749: form of the NFW, the generalized NFW or the beta profile. None of these combinations can give 

750: an acceptable fit to the global dark matter distribution. For regions outside 0.85 \re, 

751: where the dark component begins to dominate the gravity, we find that the generalized NFW profile 

752: with $\zeta=0$ or $\zeta=1$ (=NFW) can shape the dark matter profile well, with 

753: $r_s=5.6$\re~for $\zeta=0$) or 9.3\re~for $\zeta=1$.

754: In Figure 5d we illustrate the dark matter fraction $f_{\rm dark}$ as a function of radius, 

755: along with the 90\% errors. We find that in the central region 

756: ($^{<}_{\sim}0.85$\re~or 7.7 kpc)

757: the lower and upper limits of \totMtoLB~are about 5 and 10 \solarMtosolarLB, respectively,

758: inferring that the dark matter fraction is less than about 50\%. In the outer regions, however, 

759: the dark matter fraction increases fast outwards and begins to dominate the total gravitating 

760: mass at about 2\re, where the dark matter fraction is $80\pm5$\%.

761: 

762: 

763: A reliable description of the dark matter profile in early-type galaxies depends on to what extent 

764: errors and uncertainties on the stellar and gas mass are minimized. For NGC 1407, although the 

765: resulting stellar population mass-to-light ratio determined from absorption-line strengths appears 

766: to be a reasonable approximation to the dynamical mass-to-light ratio (see, e.g., Cappellari et al. 2006), 

767: and there is no apparent radio, optical or X-ray evidence against the assumption of the hydrostatic 

768: equilibrium, we would like to emphasize that the results still needs to be inspected further in the 

769: future by comparing with those derived from other independent methods

770: 

771: 

772: %%%%%%%%%%%%%%%%%%%% 

773: \section{Discussion}

774: %%%%%%%%%%%%%%%%%%%%

775: %===================================================================

776: \subsection{Errors and Uncertainties on Stellar Mass-to-Light Ratio}

777: %===================================================================

778: We find that within 0.5\re~the calculated B-band mass-to-light ratio is nearly consistent with 

779: a constant distribution of 5.3\solarMtosolarLB~ (Thomas ${\rm [MgFe]_{T}}$) or 4.8\solarMtosolarLB~

780: (Cappellari ${\rm [MgFe]_{C}}$) at the 90\% confidence level (Table 3), if the Kroupa IMF is 

781: assumed. We also have crosschecked the results by adopting the diet-Salpeter IMF (Bell et al. 2003) 

782: and the Chabrier (2003) IMF. The results are essentially the same as those obtained with the Kroupa 

783: IMF, with a deviation of less than 10\%. If we convert the obtained B-band stellar mass-to-light 

784: ratios into their I-band counterparts, our results agree very well with the Cappellari et al.'s mean 

785: value for NGC 1407 (2.44\solarMtosolarLI). The calculated stellar mass-to-light ratios by using the 

786: Salpeter IMF are systematically larger by about a factor of 1.7. This should be ascribed to the fact 

787: that the Salpeter IMF overestimates the number of low-mass stars, as has been proposed by, e.g., 

788: Weiner et al. (2001) and Cappellari et al. (2006). No matter which IMF is used, the derived age 

789: is in excellent agreement with those in earlier works (Thomas et al. 2005; Howell 2005; 

790: Humphrey et al. 2006a), and is also consistent with the mean age of the early-type galaxies 

791: at $z=0$ that is obtained with the Spitzer data (Bregman et al. 2006).

792: 

793: 

794: As we found in \S5.2, the stellar population in NGC 1407 is $\alpha$-enriched, so we introduced

795: a combinative index MgFe as a reasonable tracer of the total metallicity of the stellar population.

796: To crosscheck our results on the stellar age and metallicity with an alternative approach, we 

797: also have interpolated the H$\beta$, Mgb, Fe5270, Fe5335 and Fe5015 index values given in the 

798: model table in Thomas et al. (2003), where the $\alpha$-products and iron are treated separately 

799: so that the impact of the $\alpha/{\rm Fe}$ bias is minimized. The derived ages are in good 

800: agreement with our PEGASE results, although the derived metallicities are marginally consistent 

801: with, or slightly higher than the PEGASE values due to the bluer red giant branches in the 

802: stellar evolutionary tracks that are used, as well as the higher adopted $\alpha/{\rm Fe}$ (0.3) 

803: comparing to the value (0) in PEGASE code. Within the frame of the simple stellar population 

804: model (Maraston 2005), these values infer quite a similar \starMtoLB~distribution to what we 

805: have obtained with the PEGASE code. 

806: 

807: 

808: In addition to the possible systematic errors and uncertainties that are discussed above, there 

809: are a number of added error and uncertainty sources in the calculations of stellar mass-to-light 

810: ratios. In particular, by comparing the results obtained with the Kroupa IMF, the diet-Salpeter 

811: IMF and the Chabrier IMF, we estimate that the model uncertainties on the IMF profile may introduce 

812: a typical error of about 10\%. The size of this error is similar to that caused by the systematic 

813: errors or model uncertainties of the built-in stellar spectrum library of the Pegase code, e.g., 

814: the uncertainty in the stellar effective temperatures. Additional errors of about 5\% also come 

815: from the modeling of the star formation history and the evolution of galactic winds. 

816: Combining these errors with those occurred in the measurements of photometric data and line indices 

817: ($5-10\%$), the resulting stellar mass-to-light ratios have a typical error of about $15-20\%$. 

818: More detailed correlative descriptions about the errors in the single stellar population models 

819: can be found in Trager et al. (2000) and Le Borgne et al. (2004), where consistent conclusions were 

820: given.

821: 

822: 

823: %=====================================================

824: \subsection{Total Mass-to-Light Ratio and Dark Matter}

825: %=====================================================

826: Like in other early-type galaxies (e.g., Lintott et al. 2005) and spiral galaxies (e.g., Kormendy and Bender 1999), 

827: we find that NGC 1407 has a baryon-dominated core. At 1\re, the mass-to-light ratio and dark 

828: matter fraction are estimated to be \totMtoLB~=7.7\solarMtosolarLB~and $f_{\rm dark}=43^{+14}_{-12}\%$, 

829: respectively, which is typical for elliptical galaxies (e.g., Cappellari et al. 2006; 

830: Fukazawa et al. 2006; Padmanabhan et al. 2004). In $>1$\re, \totMtoLB~and $f_{\rm dark}$ 

831: increases outwards quickly, which reaches 

832: \totMtoLB~=$22.2^{+7.8}_{-5.9}$\solarMtosolarLB~and 

833: $f_{\rm dark}=75^{+5}_{-5}\%$ at 2\re,

834: $46.2^{+20.3}_{-23.2}$\solarMtosolarLB~and $90^{+2.8}_{-2.0}\%$ at 5\re, 

835: and

836: $91.0^{+17.4}_{-56.3}$\solarMtosolarLB~and $95^{+1.2}_{-1.3}\%$ at 10\re.

837: This clearly implies the existence of a large amount of dark matter that dominating the gravitating 

838: mass. The measured \totMtoLB~of NGC 1407 is close to that found in NGC 1600 by Sivakoff et al. 

839: (2004; \totMtoLB~=$10\pm3$ and $24\pm6$\solarMtosolarLB~at about 1 and 4\re, respectively), who 

840: used the same X-ray technique as ours. Our \totMtoLB~is also close to, or significantly larger 

841: than those for NGC 3923 (\totMtoLB~is between $17-32$\solarMtosolarLB~at about 10\re) and  

842: NGC 720 (6.0, 12.3 and 18.7\solarMtosolarLB~at 1, 2 and 3\re, respectively), as were found by 

843: Buote and Canizares (1998) and Buote et al. (2002) who applied a geometric test technique to 

844: the \ROSAT~and \Chandra~images. Apparently, our \totMtoLB~of NGC 1407 is significantly larger 

845: than the controversial results of NGC 821 ($13-17$\solarMtosolarLB), NGC 3379($5-8$\solarMtosolarLB), 

846: NGC 4494 ($5-7$\solarMtosolarLB) and NGC 4697 ($\sim11$\solarMtosolarLB), which was measured 

847: in $4-6$\re~by Romanowsky et al. (2003) with the planetary nebula spectrograph.   

848: 

849: 

850: At about 10\re~the inferred total mass are $6.43^{+1.23}_{-3.98}\times10^{12}$ \solarM, 

851: which are 1-6 times larger than the typical values for groups of galaxies (Humphrey et al. 2006b). 

852: At the virial radius $R_{200}$, the extrapolated gravitating mass and mass-to-light ratio are 

853: $2.20\pm0.42\times10^{13}$ \solarM~and $311\pm60$ \solarMtosolarLB, 

854: respectively. We find that these values are marginally consistent with those of Quintana et al. (1994) 

855: and are slightly larger than those of Humphrey et al. (2006b). Our results are lower by a factor 

856: of $2-3$ than those of Trentham et al. (2006) and Gould (1993) in which NGC 1400 was included 

857: in the kinematic analysis. This casts some doubts on the identity of NGC 1400 as a virialized 

858: member in the group. Since NGC 1400 is likely to be at the distance of the NGC 1407 group, as 

859: was suggested by Perrett et al. (1996) who compared the shapes of GC luminosity functions of 

860: NGC 1400 and NGC 1407, the observed unusually large peculiar velocity may imply that NGC 1400 

861: is intruding into and merging with the group. By comparing the mass-to-light ratio of the NGC 1407 

862: group with those listed in, e.g., Sanderson \& Ponman (2003), we find that, similar to the NGC 1600 

863: group, the NGC 1407 group is an extremely dark system even comparable to many clusters of galaxies. 

864: 

865: 

866: We find that the total mass distribution in the NGC 1407 group shows a slop change, or a 

867: flattened feature at about $^{<}_{\sim}0.85$\re~(7.7 kpc). Outside this radius the dark 

868: matter fraction increases rapidly outwards. This is in consistence with the result in the 

869: sample work of Humphrey et al. (2006b). This flattened feature may be caused by the change 

870: of the relative contribution of the gas mass as compared with the stellar mass. However, 

871: we notice that a similar feature is also seen on the mass profile of dark matter, to derive 

872: which the contributions of both bright and dwarf members have been carefully taken into account. 

873: To examine if the appearance of the flattened feature is model-dependent, we have repeated 

874: our calculations in \S5.1, \S5.3 and \S5.4 by starting from the deprojected X-ray surface 

875: brightness profile (e.g., Chatzikos et al. 2006), to derive which the background given in 

876: \S4.1 is used. We find that within the errors the results agree nicely with what we presented 

877: above, indicating that the flattened, or the shoulder-like feature is intrinsic. 

878: We speculate that the slope changes on both the \totM~and \totMtoLB~profiles are likely 

879: to be a mark of the boundary between the galaxy and group halos, since in $1-2$\re~we also 

880: observe a transition of the gas temperature from the galactic level ($\simeq0.7$ keV) to the 

881: group level ($>1$ keV). Similar phenomenon and its analogs have been observed in clusters 

882: of galaxies (Ikebe 1995); Xu et al. (1998); Ettori et al. 2002; also Makishima et al. 2001 

883: and references therein), as well as in other groups (NGC 4472 and NGC 4649, Brighenti \& Mathews 1997; 

884: NGC 507, Paolillo et al. 2003; NGC 1600, Sivakoff et al. 2004; IC 1459 and NGC 1399, Fukazawa et al 2006; 

885: NGC 4125 and NGC 4261, Humphrey et al. 2006b). We find that in more than half of these cases 

886: the slope change on mass profile appears at $7\sim10$ kpc, meanwhile in roughly half of the 

887: cases an outward temperature increase is observed. We also notice that the slope change 

888: is always seen in the cluster/group-dominating galaxies, or galaxies dominating a sub-cluster. 

889: In the isolated or nearly isolated elliptical galaxies, such as NGC 1404 (Paolillo et al. 2002),

890: NGC 4555 (O'Sullivan \& Ponman 2004), NGC 4636 (Brighenti \& Mathews 1997) and NGC 6482 (Humphrey et al. 2006b), 

891: such a feature is not clearly detected. Because NGC 6482 is believed to dominate a fossil group,

892: the observed slope change on mass profile may finally vanish when the halo collapse enters the

893: final stage.

894: 

895: 

896: The mass distribution of NGC 1407 and its group in $>1$\re, where the mass is dominated 

897: by dark matter, can be well fitted by a single NFW profile. The best-fit parameter are 

898: $r_s=1.3$\re~($\zeta=0$) or 3.5 \re~($\zeta=1$). The derived concentration 

899: parameter c ($18.6\pm1.5$) is within the range allowed for groups, but is beyond the 68\% 

900: scatter for a halo at $z=0$ at the given $M_{\rm vir}$ (Gnedin and Ostriker 2001; Jing 2000). 

901: Despite of the relatively large errors we find that the inferred total mass and dark matter 

902: distributions within $0.85$\re~are cuspy and can be approximated by power-law profiles 

903: with indices of $\sim2$, which are marginally consistent with the generalized NFW profiles

904: with $\zeta=2$.

905: 

906: %========================================

907: \subsection{Stellar Metallicity Gradient}

908: %========================================

909: Like what was found in some other giant elliptical galaxies (Arimoto et al. 1997 and references 

910: therein), the calculated stellar metallicity in NGC 1407 is not spatially uniform. This is 

911: mostly determined by the observed spatial variations of the metal line indices, and is almost 

912: independent of the model assumptions on the IMF profile and so on. The inward metallicity 

913: enhancement has been predicted by the monolithic collapse models (e.g., Eggen et al. 1962). 

914: Within the frame of such models, more metal-rich stars were born in the inward-dissipating 

915: gas that had been enriched on their way towards the galactic center, meanwhile the evolving 

916: stellar winds tend to expel more metals out of the outer regions where the gravitational 

917: potential is shallow. Numerical simulations show that the stellar metallicity gradients are 

918: also adjusted by mergers. Although the mergers tend to smear out any existing gradients, 

919: they also may supply a large amount of gas and trigger star formation in the inner regions. 

920: So the resulting metallicity gradients are not necessary to be as steep as is predicted by 

921: the monolithic collapse models (e.g., Sanchez-Blazquez et al. 2006 and references therein), 

922: as is observed in NGC 1407. 

923: If the stellar metallicity gradient at $\sim0.5$\re~ is indeed adjusted by gas inflows, by 

924: calculating the time that the gas therein needed to radiatively cool down to reach the current 

925: temperature ($\simeq0.7$ keV), which is about 4 Gyr, we tentatively estimate that such a gas 

926: inflow event should occurred at least 4 Gyr ago. From then on the gas therein has sufficient 

927: time to reestablish an equilibrium state.

928: 

929: 

930: 

931: %###################################

932: \section{Summary}\label{sec:summary}

933: %###################################

934: 

935: By incorporating the \Chandra~and \ROSAT~X-ray spectroscopic data with the optical line-strength

936: indices and multi-color photometric data, we studied the mass distributions in the bright E0 galaxy

937: NGC 1407 and its associated group. We find that the gas is single-phase with a temperature of

938: $\simeq0.7$ keV within 1\re~(1\re~= 9.0 kpc), which quickly increases to $>1$ keV in $1-2$\re.

939: We reveal that the X-ray surface brightness profile shows a central excess in the innermost region,

940: which coincides with a flattened feature at about $^{<}_{\sim}1$\re~(7.7 kpc) on the total mass

941: and dark matter profiles. This feature is speculated to be a mark of the boundary between the galaxy

942: and group halos, as has been seen in some other cluster/group-dominating galaxies. We find that the

943: total mass and dark matter distributions within $0.85$\re~(7.7 kpc) are cuspy and can be approximated

944: by power-law profiles with indices of $\sim2$, which are marginally consistent with the generalized

945: NFW profiles. The mass in outer regions can be well fitted by a single NFW profile, and the derived

946: concentration parameter c ($18.6\pm1.5$) is larger than the 68\% upper limit for a halo at $z=0$ at

947: the given $M_{\rm vir}$. We conclude that although the NGC 1407 group has a baryon-dominated core,

948: it is an extremely dark system even comparable to many clusters of galaxies. At the virial radius

949: $r_{200}=572\pm118$ kpc, the calculated mass and mass-to-light ratio are 

950: $M_{200}=2.20\pm0.42 \times 10^{13}$ \solarM~and $M_{\rm vir}/L_{\rm B}=311\pm60$ \solarMtosolarLB,

951: respectively. We also argue that the unusually large peculiar velocity of NGC 1400 may reflect the

952: fact that the galaxy may have not be virialized in the group's gravitational potential well.

953: 

954: 

955: %###############

956: \acknowledgments

957: %###############

958: 

959: We thank Roberto Rampazzo and Annibalii for kindly providing us with the optical line-strength indices 

960: that are used in this paper. We thank Damien Le Borgne, Claudia Maraston and Scott C. Trager for 

961: their helpful suggestions and comments. This work was supported by the National Science Foundation 

962: of China (Grant No. 10273009, 10233040 and 10503008), Shanghai Key Projects in Basic Research No. 04JC14079, 

963: and the Ministry of Science and Technology of China, under Grant No. NKBRSF G19990754.

964: 

965: 

966: 

967: %##########

968: %REFERENCES

969: %##########

970: \begin{references}

971: \reference{}Arimoto, N., Matsushita, K., Ishimaru, Y., Ohashi, T., \&  Renzini, A. 1997, ApJ, 477, 128

972: \reference{}Arnaud, M. 2005, astro-ph/0508159

973: \reference{}Baes, M., \& Dejonghe, H. 2001, ApJ, 563, L19 

974: \reference{}Bell, E. F., McIntosh, D. H., Katz, N., \& Weinberg, M. D. 2003, ApJS, 149, 289

975: \reference{}Bernardi, M., Alonso, M. V., da Costa, L. N., Willmer, C. N. A., Wegner, G., Pellegrini, P. S., Rit\'{e}, C., \& Maia, M. A. G. 2002, AJ, 123, 2990

976: \reference{}Bertin, G., Bertola, F., Buson, L. M., Danzinger, I. J., Dejonghe, H., Sadler, E. M., Saglia, R. P., de Zeeuw, P. T., \& Zeilinger, W. W. 1994, A\&A, 292, 381

977: \reference{}Bertola, F., Pizzella, A., Persic, M., \& Salucci, P. 1993, ApJ, 416, L45

978: \reference{}Bregman, J. N., Temi, P., \& Bregman, J. D. 2006, astro-ph/0604399

979: \reference{}Brighenti, F., \& Mathews, W. G. 1997, ApJ, 486, L83

980: \reference{}Buote, D. A., \& Canizares, C. R. 1994, ApJ, 427, 86

981: \reference{}Buote, D. A., \& Canizares, C. R. 1998, MNRAS, 298, 811

982: \reference{}Buote, D. A., Jeltema, T. E., Canizares, C. R., \& Garmire, G. P. 2002, ApJ, 577, 183 

983: \reference{}Burkert, A. 1993, A\&A, 278, 23

984: \reference{}Cappellari, M., Bacon, R., Bureau, M., Damen, M. C., Davies, R. L., de Zeeuw, P. T., Emsellem, E., Falc\'{o}n-Barroso, J., Krajnovi\'{c}, D., Kuntschner, H., \& 5 coauthors, 2006, MNRAS, 366, 1126

985: \reference{}Carollo, C. M., de Zeeuw, P. T., van der Marel, R. P., Danziger, I. J., \& Qian, E. E. 1995, ApJ, 441, L25 

986: \reference{}Chabrier, G. 2003, PASP, 115, 763

987: \reference{}Chatzikos, M., Sarazin, C. L., \& Kempner, J. C. 2006, ApJ, 643, 751

988: \reference{}de Carvalho, R. R., da Costa, L. N., \& Djorgovski, S. 1991, ApJS, 76, 1067

989: \reference{}Dekel, A., Stoehr, F., Mamon, G. A., Cox, T. J., Novak, G. S., \& Primack, J. R. 2005, Nature, 437, 707

990: \reference{}Denicol\'{o}, G., Terlevich, R., Terlevich, E., Forbes, D. A., Terlevich, A., \& Carrasco, L. 2005, MNRAS, 356, 1440

991: \reference{}de Vaucouleurs, G., de Vaucouleurs, A., Corwin, Jr., H. G., Buta, R. J., Paturel, G., \& Fouque, P. 1991, Third Reference Catalogue of Bright Galaxies (RC3 Catalog)

992: \reference{}Dickey, J. M., \& Lockman, F. J. 1990, ARA\&A, 28, 215

993: \reference{}Eggen, O. J., Lynden-Bell, D., \& Sandage, A. R. 1962, ApJ, 136, 748

994: \reference{}Ettori, S., Fabian, A. C., Allen, S. W., \& Johnstone, R. M. 2002, MNRAS, 331, 635

995: \reference{}Faber, S. M., Tremaine, S., Ajhar, E. A., Byun, Y., Dressler, A., Gebhardt, K., Grillmair, C., Kormendy, J., Lauer, T. R., \& Richstone, D. 1997, AJ, 114, 1771

996: \reference{}Fioc, M., \& Rocca-Volmerange, B. 1997, A\&A, 326, 950

997: \reference{}Forbes, D. A., S\'{a}nchez-Bl\'{a}zquez, P., Phan, A. T. T., Brodie, J. P., Strader, J., \& Spitler, L. 2006, MNRAS, 366, 123 \reference{}Forman, W., Jones, C., \& Tucker, W. 1985, ApJ, 293, 102

998: \reference{}Forte, J. C., Faifer, F., \& Geisler, D. 2005, MNRAS, 357, 56

999: \reference{}Franx, M., Illingworth, G., \& Heckman, T. 1989, AJ, 98, 538

1000: \reference{}Fukazawa, Y., Botoya-Nonesa, J. G., Pu, J., Ohto, A., \& Kawano, N. 2006, ApJ, 636, 698

1001: \reference{}Gerhard, O., Kronawitter, A., Saglia, R. P., \& Bender, R. 2001, AJ, 121, 1936

1002: \reference{}Gnedin, O. Y., \& Ostriker, J. P. 2001, ApJ, 561, 61

1003: \reference{}Gould, A. 1993, ApJ, 403, 37

1004: \reference{}Grevesse, N., \& Sauval, A. J. 1998, SSRv, 85, 161

1005: \reference{}Griffiths, R. E., Casertano, S., Im, M., \& Ratnatunga, K. U. 1996, MNRAS, 282, 1159

1006: \reference{}Howell, J. H. 2005, AJ, 130, 2065

1007: \reference{}Humphrey, P. J., \& Buote, D. A. 2006a, ApJ, 639, 136

1008: \reference{}Humphrey, P. J., Buote, D. A., Gastaldello, F., Zappacosta, L., Bullock, J. S., Brighenti, F., \& Mathews, W. G. 2006b, astro-ph/0601301

1009: \reference{}Ikebe, Y. 1995, PhDT, 15

1010: \reference{}Jarrett, T. H., Chester, T., Cutri, R., Schneider, S. E., \& Huchra, J. P. 2003, AJ, 125, 525

1011: \reference{}Jing, Y. P. 2000, ApJ, 535, 30

1012: \reference{}Keeton, C. R., Kochanek, C. S., \& Falco, E. E. 1998, ApJ, 509, 561 

1013: \reference{}Kochanek, C. S. 1995, ApJ, 453, 545

1014: \reference{}Kormendy, J., \& Bender, R. 1999, ApJ, 522, 772

1015: \reference{}Kronawitter, A., Saglia, R. P., Gerhard, O., \& Bender, R. 2000, A\&AS, 144, 53

1016: \reference{}Kroupa P. 2001, MNRAS, 322, 231

1017: \reference{}Lauer, T. R. 1985, ApJS, 57, 473

1018: \reference{}Le Borgne, D., Rocca-Volmerange, B., Prugniel, P., Lan\c con, A., Fioc, M. \& Soubiran, C. 2004, A\&A, 425, 881

1019: \reference{}Lintott, C., Ferreras, L., \& Lahav, O. 2005, astro-ph/0512175 

1020: \reference{}Loewenstein, M., \& White, R. E., III 1999, ApJ, 518, 50 

1021: \reference{}Makishima, K., Ezawa, H., Fukuzawa, Y., Honda, H., Ikebe, Y., Kamae, T., Kikuchi, K., Matsushita, K., Nakazawa, K., Ohashi, T., \& 3 coauthors, 2001, PASJ, 53, 401	

1022: \reference{}Maraston, C. 2005, MNRAS, 362, 799

1023: \reference{}Matsushita, K. 2001, ApJ, 547, 693

1024: \reference{}McElroy, D. B. 1995, ApJS, 100, 105

1025: \reference{}Merritt, D., \& Ferrarese, L. 2001, ApJ, 547, 140 

1026: \reference{}Michard R. 2005, A\&A, 441, 451

1027: \reference{}Milgrom, M. 1983, ApJ, 270, 365

1028: \reference{}Moore, B., Quinn, T., Governato, F., Stadel, J., \& Lake, G. 1999, MNRAS, 310, 1147

1029: \reference{}Mushotzky, R. F., Loewenstein, M., Awaki, H., Makishima, K., Matsushita, K., \& Matsumoto, H. 1994, ApJ, 436, L79

1030: \reference{}Mushotzky, R. F., Cowie, L. L., Barger, A. J., \& Arnaud, K. A. 2000, Nature, 404, 459

1031: \reference{}Navarro, J. F., Frenk, C. S., \& White, Simon D. M. 1997, ApJ, 490, 493

1032: \reference{}O'Sullivan, E., \& Ponman, T. J. 2004, MNRAS, 354, 935

1033: \reference{}Padmanabhan, N., Seljak, U., Strauss, M. A., Blanton, M. R., Kauffmann, G., Schlegel, D. J., Tremonti, C., Bahcall, N. A., Bernardi, M., Brinkmann, J., \& 2 coauthors 2004, NewA, 9, 329

1034: \reference{}Paolillo, M., Fabbiano, G., Peres, G., \& Kim, D.-W. 2002, ApJ, 565, 883 

1035: \reference{}Paolillo, M.; Fabbiano, G., Peres, G., \& Kim, D.-W. 2003, ApJ, 586, 850

1036: \reference{}Perrett, K. M., Hanes, D. A., Butterworth, S. T., Kavelaars, J. J., Geisler, D., \& Harris, W. E. 1996, AAS, 189, 7116

1037: \reference{}Pratt, G. W., \& Arnaud, M. 2003, A\&A, 408, 1

1038: \reference{}Prugniel, Ph., \& Simien, F. 1996, A\&A, 309, 749

1039: \reference{}Quintana, H., Fouque, P., \& Way, M. J. 1994, A\&A, 283, 722 

1040: \reference{}Rampazzo, R., Annibali, F., Bressan, A., Longhetti, M., Padoan, F., \& Zeilinger, W. W. 2005, A\&A, 433, 497

1041: \reference{}Romanowsky, A. J., Douglas, N. G., Arnaboldi, M., Kuijken, K., Merrifield, M. R., Napolitano, N. R., Capaccioli, M., \& Freeman, K. C. 2003, Science, 301, 1696

1042: \reference{}Saglia, R. P., Bertin, G., \& Stiavelli, M. 1992, ApJ, 384, 433

1043: \reference{}Salpeter E. E. 1955, ApJ, 121, 161

1044: \reference{}Sanchez-Blazquez, P., Gorgas, J., \& Cardiel N. 2006, astro-ph/0604571 

1045: \reference{}Sanderson, A. J. R., \& Ponman, T. J. 2003, MNRAS, 345, 1241

1046: \reference{}Schlegel, D. J., Finkbeiner, D. P., \& Davis, M. 1998, ApJ, 500, 525

1047: \reference{}Sivakoff, G. R., Sarazin, C. L., \& Carlin, J. L. 2004, ApJ, 617, 262

1048: \reference{}Sofue, Y., \& Rubin, V. 2001, ARA\&A, 39, 137

1049: \reference{}Soldatenkov, D. A., Vikhlinin, A. A., \& Pavlinsky, M. N. 2003, AstL, 29, 298 

1050: \reference{}Strauss, M. A., Huchra, J. P., Davis, M., Yahil, A., Fisher, K. B., \& Tonry, J. 1992, ApJS, 83, 29

1051: \reference{}Tamura, N., \& Ohta, K. 2003, AJ, 126, 596

1052: \reference{}Tamura, N., \& Ohta, K. 2004, MNRAS, 355, 617

1053: \reference{}Thomas, D., Maraston, C. \& Bender, B. 2003, MNRAS, 339, 897

1054: \reference{}Thomas, D., Maraston, C., Bender, R., \& de Oliveira, C. M. 2005, ApJ, 621, 673

1055: \reference{}Tonry, J. L., Dressler, A., Blakeslee, J. P., Ajhar, E. A., Fletcher, A. B., Luppino, G. A., Metzger, M. R., \& Moore, C. B. 2001, ApJ, 546, 681

1056: \reference{}Trager, S. C., Faber, S. M., Worthey, G., \& Gonz\'{a}lez, J. J. 2000, AJ, 119, 1645

1057: \reference{}Trager, S. C., Worthey, G., Faber, S. M., Burstein, D., \& Gonzalez, J. J. 1998, ApJS, 116, 1

1058: \reference{}Trentham, N., Tully, R. B., \& Mahdavi, A. 2006, MNRAS, 369, 1375 

1059: \reference{}Weiner, B. J., Sellwood, J. A., \& Williams, T. B. 2001, ApJ, 546, 931

1060: \reference{}Wyithe, J. Stuart B. 2006, MNRAS, 365, 1082

1061: \reference{}Xu, H., Makishima, K., Fukazawa, Y., Ikebe, Y., Kikuchi, K., Ohashi, T., \& Tamura, T. 1998, ApJ, 500, 738

1062: \reference{}Xu, Y., Xu, H., Zhang, Z., Kundu, A., Wang, Y., \& Wu, X. 2005, ApJ, 631, 809 

1063: \reference{}Zhang, Z., \& Xu, H. 2004, ChJAA, 4, 221

1064: 

1065: 

1066: 

1067: 

1068: \end{references}

1069: 

1070: 

1071: 

1072: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

1073: \clearpage

1074: 

1075: %=======

1076: %Table 1

1077: %=======

1078: \begin{deluxetable}{lcccccccr}

1079: \tabletypesize{\scriptsize}

1080: \tablecaption{\Chandra~and \ROSAT~Spectral Analyses}

1081: \tablewidth{0pt}

1082: \tablehead{

1083: & & 

1084: \multicolumn{2}{c}{Non-thermal Component} & &

1085: \multicolumn{3}{c}{Thermal Component} &\\

1086: \cline{3-4} 

1087: \cline{6-8}

1088: \colhead{Source} &

1089: \colhead{Model\tablenotemark{a} \,} &

1090: \colhead{$\Gamma$ or $kT_{\rm in}$} & 

1091: \colhead{$F_{\rm X}$\tablenotemark{b} \,} & &

1092: \colhead{$kT$} & 

1093: \colhead{$Z$} &

1094: \colhead{$F_{\rm X}$\tablenotemark{b} \,} &\\

1095: & &

1096: \colhead{(keV)} & & &

1097: \colhead{(keV)} & 

1098: \colhead{(solar)} & &

1099: \colhead{$\chi^2$/d.o.f}

1100: }

1101: 

1102: \startdata

1103: \multicolumn{9}{c}{(1) Central Source}\\

1104: \hline

1105:  Src 1\tablenotemark{c} \, &PL   & $2.29^{+0.24}_{-0.23}$ & $0.47^{+0.05}_{-0.06}$ & &\nodata &\nodata &\nodata & 10.2/11\\

1106:                          &DBB  & $0.49^{+0.15}_{-0.10}$ & $0.30^{+0.15}_{-0.14}$ & &\nodata &\nodata &\nodata & 27.8/11\\

1107: 

1108: \hline

1109: \multicolumn{9}{c}{(2) \Chandra~Projected Analysis (0.7--7 keV)\tablenotemark{d} \,}\\

1110: \hline

1111: 

1112: 1 &PL+APEC   &1.59/fixed &$1.26\pm0.07$ & &$0.68\pm0.02$ &$0.52^{+0.11}_{-0.15}$ &$3.31\pm0.92$ &71.0/48\\

1113: 2 &PL+APEC   &1.59/fixed &$0.70^{+0.22}_{-0.09}$ & &$0.73\pm0.02$ &$0.55^{+0.18}_{-0.14}$ &$3.04\pm0.86$ &82.5/48\\

1114: 3 &PL+APEC   &1.59/fixed &$1.31^{+0.17}_{-0.15}$ & &$0.91\pm0.03$ &$0.73^{+0.43}_{-0.18}$ &$2.05\pm0.69$ &62.2/48\\      

1115: 4 &PL+APEC   &1.59/fixed &$1.97^{+0.47}_{-0.31}$ & &$1.21^{+0.09}_{-0.12}$ &$0.38^{+0.27}_{-0.14}$ &$2.24\pm0.69$ &52.7/48\\ 

1116: 

1117: \hline

1118: \multicolumn{9}{c}{(3) \Chandra~Deprojected Analysis (0.7--7 keV)\tablenotemark{d} \,}\\

1119: \hline

1120: 1 &PL+APEC   &1.59/fixed &\nodata & &$0.66\pm0.02$ &$0.60^{+0.41}_{-0.20}$ &\nodata &617/634\\

1121: 2 &PL+APEC   &1.59/fixed &\nodata & &$0.68\pm0.02$ &$0.63^{+0.46}_{-0.20}$ &\nodata &\nodata\\

1122: 3 &PL+APEC   &1.59/fixed &\nodata & &$0.85^{+0.04}_{-0.02}$ &$1.34\pm0.57$ &\nodata &\nodata\\  

1123: 4 &PL+APEC   &1.59/fixed &\nodata & &$1.25\pm0.09$ &$0.52^{+0.31}_{-0.25}$ &\nodata &\nodata\\ 

1124: 

1125: \hline

1126: \multicolumn{9}{c}{(2) \ROSAT~Projected Analysis (0.2--2 keV)\tablenotemark{e} \,}\\

1127: \hline

1128: 

1129: 1 &PL+APEC   &1.59/fixed &$0.76\pm0.52$ & &$0.66^{+0.09}_{-0.12}$ &$0.73^{+0.13}_{-0.11}$ &$1.55\pm0.24$ &82.9/69\\

1130: 2 &PL+APEC   &1.59/fixed &$0.42\pm0.32$ & &$0.67^{+0.06}_{-0.07}$ &$0.98^{+0.13}_{-0.11}$ &$3.53\pm0.25$ &63.0/69\\

1131: 3 &PL+APEC   &1.59/fixed &$0.78\pm0.08$ & &$0.64^{+0.09}_{-0.10}$ &$1.44\pm0.52$ &$1.69\pm0.55$ &83.0/69\\      

1132: 4 &PL+APEC   &1.59/fixed &$1.24\pm0.37$ & &$1.03^{+0.30}_{-0.20}$ &$0.69^{+0.46}_{-0.18}$ &$2.23\pm0.27$ &70.4/69\\ 

1133: 5 &APEC      &\nodata &\nodata                          & &$1.37^{+0.55}_{-0.32}$ &$0.80^{+0.67}_{-0.35}$ &$2.32\pm0.31$ &51.2/69\\

1134: 6 &APEC      &\nodata &\nodata                          & &$1.50\pm0.48$ &$0.21\pm0.07$ &$2.69\pm0.46$ &185.0/209\\

1135: 7 &APEC      &\nodata &\nodata                          & &$1.09^{+0.42}_{-0.19}$ &0.21/tied              &$4.66\pm0.68$ &\nodata\\      

1136: 8 &APEC      &\nodata &\nodata                          & &$1.09\pm0.35$ &0.21/tied              &$2.63\pm0.75$ &\nodata\\ 

1137: 

1138: 

1139: \hline

1140: \multicolumn{9}{c}{(3) \ROSAT~Deprojected Analysis (0.2--2 keV)\tablenotemark{e} \,}\\

1141: \hline

1142: 1 &PL+APEC   &1.59/fixed &\nodata & &$0.73^{+0.27}_{-0.19}$ &$0.59\pm0.17$ &\nodata &531/576\\

1143: 2 &PL+APEC   &1.59/fixed &\nodata & &$0.69\pm0.08$ &$0.78\pm0.08$ &\nodata &\nodata\\

1144: 3 &PL+APEC   &1.59/fixed &\nodata & &$0.63\pm0.10$ &$1.33^{+0.14}_{-0.18}$ &\nodata &\nodata\\  

1145: 4 &PL+APEC   &1.59/fixed &\nodata & &$1.12^{+0.44}_{-0.32}$ &$0.60^{+0.59}_{-0.22}$ &\nodata &\nodata\\ 

1146: 5 &APEC      &\nodata    &\nodata & &$1.37\pm0.35$ &$0.49^{+0.64}_{-0.25}$ &\nodata &\nodata\\

1147: 6 &APEC      &\nodata    &\nodata & & 1.37/tied             &$0.21^{+0.06}_{-0.07}$ &\nodata &\nodata\\

1148: 7 &APEC      &\nodata    &\nodata & &$1.22\pm0.35$ & 0.21/tied             &\nodata &\nodata\\  

1149: 8 &APEC      &\nodata    &\nodata & &$1.08\pm0.30$ & 0.21/tied             &\nodata &\nodata\\ 

1150: 

1151: 

1152: 

1153: \enddata

1154: 

1155: \tablenotetext{a}{Models used in the spectral fittings are 

1156: PL = the power-law model,

1157: DBB = the multiple color disk model (diskbb) and

1158: APEC = the thermal plasma model. 

1159: In all cases the column density are fixed to the Galactic value $5.42\times10^{20}$ cm$^{-2}$ 

1160: (Dickey \& Lockman, 1990), since allowing it to vary does not improve the fit.}

1161: 

1162: \tablenotetext{b}{0.7--7 keV fluxes for the \Chandra~observation or 0.2--2 keV fluxes for \ROSAT~observation

1163: in $10^{-13}$ ergs cm$^{-2}$ s$^{-1}$ (90\% errors), which have been corrected for absorption.}

1164: \tablenotetext{c}{The central source.}

1165: \tablenotetext{d}{Annuli used to extract the \Chandra~ACIS S3 spectra: 

1166: $0-0.25$\re,

1167: $0.25-0.53$\re,

1168: $0.53-0.91$\re~and

1169: $0.91-2.1$\re.}

1170: \tablenotetext{e}{Annuli used to extract the \ROSAT~PSPC spectra: 

1171: $0-0.25$\re,

1172: $0.25-0.53$\re,

1173: $0.53-0.91$\re,

1174: $0.91-2.1$\re,

1175: $2.1-3.4$\re,

1176: $3.4-5.1$\re,

1177: $5.1-7.7$\re~and

1178: $7.7-10.3$\re.

1179: }

1180: 

1181: \end{deluxetable}

1182: 

1183: %========

1184: %Table 2

1185: %========

1186: \begin{deluxetable}{cccccccc}

1187: \tabletypesize{\scriptsize}

1188: \tablecaption{Line-strength Indices\tablenotemark{a} \,and Multi-color Photometric Data\tablenotemark{b} \,}

1189: \tablewidth{0pt}

1190: \tablehead{

1191: \colhead{Radius} &

1192: \colhead{H$\beta$} &

1193: \colhead{Mgb} & 

1194: \colhead{Fe5015} & 

1195: \colhead{Fe5270} &

1196: \colhead{Fe5335} & 

1197: \colhead{B-R} & 

1198: \colhead{B-K}\\

1199: \colhead{(\re)} &

1200: \colhead{(\r{A})} &

1201: \colhead{(\r{A})} & 

1202: \colhead{(\r{A})} & 

1203: \colhead{(\r{A})} &

1204: \colhead{(\r{A})} & 

1205: \colhead{(mag)} & 

1206: \colhead{(mag)}

1207: }

1208: 

1209: \startdata

1210: 0-1/16   & $1.48\pm0.10$ & $5.29\pm0.08$ & $6.37\pm0.21$ & $3.25\pm0.11$ & $2.75\pm0.14$ & $1.63\pm0.14$ & $4.59\pm0.14$\\

1211: 1/16-1/8 & $1.67\pm0.10$ & $5.28\pm0.09$ & $6.11\pm0.21$ & $3.08\pm0.12$ & $2.44\pm0.14$ & $1.61\pm0.14$ & $4.44\pm0.14$\\

1212: 1/8-1/4  & $2.05\pm0.10$ & $5.24\pm0.09$ & $6.11\pm0.21$ & $3.19\pm0.12$ & $2.72\pm0.14$ & $1.61\pm0.14$ & $4.34\pm0.14$\\

1213: 1/4-1/2  & $1.81\pm0.10$ & $5.05\pm0.09$ & $5.68\pm0.21$ & $2.30\pm0.12$ & $1.88\pm0.14$ & $1.60\pm0.14$ & $4.30\pm0.14$\\

1214: \enddata

1215: \tablenotetext{a}{The line-strength indices transformed into the Lick-IDS system (Rampazzo et al. 2005).}

1216: \tablenotetext{b}{B- and R-band photometric data are obtained from de Carvalho et al. (1991), while K-band photometric data are from Jarrett et al. (2003).}

1217: 

1218: \end{deluxetable}

1219: 

1220: 

1221: 

1222: 

1223: %========

1224: %Table 3

1225: %========

1226: 

1227: \begin{deluxetable}{cccccccc}

1228: \tabletypesize{\scriptsize}

1229: \tablecaption{Stellar Ages, Metallicities and Mass-to-Light Ratios}

1230: \tablewidth{0pt}

1231: \tablehead{

1232: \colhead{Radius} &

1233: \colhead{\starMtoLB} &

1234: \colhead{Z\tablenotemark{b} \,} & 

1235: \colhead{age} & &

1236: \colhead{\starMtoLB} &

1237: \colhead{Z\tablenotemark{b} \,} & 

1238: \colhead{age} \\

1239: \colhead{(\re\tablenotemark{a} \,)} &

1240: \colhead{(\solarMtosolarLB)} &

1241: \colhead{($Z_{\odot}$)} & 

1242: \colhead{(Gyr)} & &

1243: \colhead{(\solarMtosolarLB)} &

1244: \colhead{($Z_{\odot}$)} & 

1245: \colhead{(Gyr)} 

1246: }

1247: 

1248: \startdata

1249: \multicolumn{8}{c}{PEGASE code}\\

1250: \hline

1251: & 

1252: \multicolumn{3}{c}{Salpeter IMF+${\rm [MgFe]_{T}}$} & & 

1253: \multicolumn{3}{c}{Kroupa IMF+${\rm [MgFe]_{T}}$} \\

1254: \cline{2-4} 

1255: \cline{6-8}

1256: 0-1/16   &$9.8^{+1.1}_{-1.3}$ & $1.7\pm0.4$ &$9.7^{+0.8}_{-0.9}$  && $5.8^{+0.8}_{-0.7}$ & $1.5^{+0.4}_{-0.5}$ & $10.4^{+1.1}_{-0.8}$\\

1257: 1/16-1/8 &$9.2^{+1.6}_{-1.4}$ & $1.3^{+0.7}_{-0.3}$ &\nodata                  && $5.5^{+1.1}_{-0.7}$ & $1.1^{+1.0}_{-0.6}$ & \nodata \\

1258: 1/8-1/4  &$9.4^{+1.2}_{-1.1}$ & $2.6\pm0.2$ &\nodata                  && $5.6^{+1.0}_{-0.7}$ & $2.5^{+0.2}_{-0.3}$ & \nodata \\

1259: 1/4-1/2  &$7.4\pm1.2$ & $0.5^{+0.3}_{-0.2}$ &\nodata                  && $4.3^{+0.9}_{-0.5}$ & $0.3^{+0.2}_{-0.1}$ & \nodata \\

1260: \hline

1261: & 

1262: \multicolumn{3}{c}{Salpeter IMF+${\rm [MgFe]_{C}}$} & & 

1263: \multicolumn{3}{c}{Kroupa IMF+${\rm [MgFe]_{C}}$} \\

1264: \cline{2-4} 

1265: \cline{6-8}

1266: 

1267: 0-1/16   &$8.9^{+0.9}_{-1.0}$ & $1.8\pm0.3$ &$8.5^{+1.0}_{-0.7}$  && $5.0^{+0.7}_{-0.6}$ & $1.6\pm0.3$ & $8.8^{+1.0}_{-1.3}$\\

1268: 1/16-1/8 &$8.7^{+1.1}_{-1.3}$ & $1.6^{+0.6}_{-0.4}$ &\nodata                  && $5.0\pm0.7$ & $1.5\pm0.5$ & \nodata \\

1269: 1/8-1/4  &$8.3^{+1.2}_{-0.6}$ & $2.6^{+0.3}_{-0.5}$ &\nodata                  && $4.8^{+0.9}_{-0.4}$ & $2.4^{+0.2}_{-0.3}$ & \nodata \\

1270: 1/4-1/2  &$7.8^{+1.3}_{-1.1}$ & $1.0\pm0.3$ &\nodata                  && $4.6^{+0.9}_{-0.8}$ & $0.9^{+0.5}_{-0.3}$ & \nodata \\

1271: \hline

1272: \multicolumn{8}{c}{Thomas and Maraston models\tablenotemark{c} \,} \\

1273: \hline

1274: 0-1/16   &$10.0^{+1.2}_{-1.1}$ & $2.5\pm0.2$ & $9.2\pm0.7$ && $6.3^{+0.8}_{-0.6}$ & $2.5\pm0.2$ & $9.2\pm0.7$\\

1275: 1/16-1/8 &$9.8^{+1.2}_{-1.1}$ & $2.3^{+0.2}_{-0.3}$ &\nodata                  && $6.2^{+0.8}_{-0.7}$ & $2.3^{+0.2}_{-0.3}$ & \nodata \\

1276: 1/8-1/4  &$9.8^{+1.2}_{-1.1}$ & $2.3^{+0.2}_{-0.3}$ &\nodata                  && $6.2^{+0.8}_{-0.7}$ & $2.3^{+0.2}_{-0.3}$ & \nodata \\

1277: 1/4-1/2  &$8.8^{+1.2}_{-1.1}$ & $1.7\pm0.2$ &\nodata                  && $5.6\pm0.7$ & $1.7\pm0.2$ & \nodata \\

1278: 

1279: \enddata

1280: 

1281: \tablenotetext{a}{Effective radius (1\re~=$1.17^{\prime}$).}

1282: \tablenotetext{b}{Calculated stellar total abundance.}

1283: \tablenotetext{c}{Stellar ages and abundances are calculated by using the Thomas model (Thomas et al. 2003). \starMtoLB~is calculated based on the Maraston model (Maraston 2005).}

1284: \end{deluxetable}

1285: 

1286: 

1287: 

1288: \clearpage

1289: 

1290: %========

1291: %figure1

1292: %========

1293: \begin{figure}

1294: \epsscale{1.0}

1295: \begin{center}

1296: \includegraphics[width=12.0cm,angle=0]{f1.ps}

1297: \end{center}

1298: \figcaption{

1299: (a): \Chandra~ACIS S3 image in 0.3--10 keV in logarithmic scale with the intensity contours 

1300: overlapped. The image has been corrected for exposure but not for background, and has been 

1301: smoothed by using a minimum signal-to-noise ratio of $3\sigma$ and a maximum signal-to-noise 

1302: ratio of $5\sigma$ per beam. The circle on the image represents the region of 1\re. 

1303: All detected point sources ($3\sigma$) are removed.

1304: (b): \ROSAT~PSPC image in 0.2--2 keV in logarithmic scale with the intensity contours 

1305: overlapped. The image has been corrected for exposure but not for background, and has been 

1306: smoothed in the same way as in Figure 1a. The \Chandra~observation field is marked with a box. 

1307: The cross indicates the weak, extented structure aligned in the NGC 1407-NGC 1400 direction.

1308: (c): Hardness ratio distribution that has been smoothed in the same way as in Figure 1a. The 

1309: hardness ratio is defined as the ratio of exposure- and background-corrected counts in 0.3--2 keV 

1310: to those in 2-10 keV, which ranges from -0.94 to 0.54 in this image.

1311: (d): 1.43 GHz radio map obtained from the National Radio Astronomical Observatory (NRAO) archive, 

1312: with peak flux density of $50.6~{\rm mJy/b}$ and contour levels of $(-1,1,2,3,4,8,16,32)~{\rm mJy/b}$.

1313: \label{fig1}}

1314: \end{figure}

1315: 

1316: %========

1317: %figure2

1318: %========

1319: \begin{figure}

1320: \epsscale{1.0}

1321: \begin{center}

1322: \includegraphics[width=6.0cm,angle=270]{f2.ps}

1323: \end{center}

1324: \figcaption{Background-corrected lightcurve of the central point source. 

1325: \label{fig2}}

1326: \end{figure}

1327: 

1328: %===========

1329: %figure3

1330: %===========

1331: \begin{figure}

1332: \epsscale{1.0}

1333: \begin{center}

1334: \includegraphics[width=8.0cm,angle=270]{f3.ps}

1335: \end{center}

1336: \figcaption{Deprojected gas temperature distribution (90\% errors) and abundance distribution 

1337: (68\% errors) obtained with \Chandra~ACIS S3 (solid crosses) and \ROSAT~PSPC (dotted crosses). 

1338: Solid lines are the best-fit models. 

1339: \label{fig3}}

1340: \end{figure}

1341: 

1342: %===========

1343: %figure4

1344: %===========

1345: \begin{figure}

1346: \epsscale{1.0}

1347: \begin{center}

1348: \includegraphics[width=12.0cm,angle=270]{f4.ps}

1349: \end{center}

1350: \figcaption{

1351: (a): The \Chandra~0.7--7 keV surface brightness profile (solid crosses) of the gas emission, which

1352: has been corrected for exposure but not for background. The best-fit beta model and two-beta model are shown as 

1353: a dashed line and a solid line, respectively.  

1354: (b): The \ROSAT~0.2--2 keV surface brightness profile (solid crosses) of the gas emission, which

1355: has been corrected for exposure but not for background. The best-fit two-beta model is shown 

1356: as a solid line.  

1357: (c): Calculated gas density distribution. 

1358: \label{fig4}}

1359: \end{figure}

1360: 

1361: 

1362: 

1363: %===========

1364: %figure5

1365: %===========

1366: 

1367: \begin{figure}

1368: \epsscale{1.0}

1369: \begin{center}

1370: \includegraphics[width=8.0cm,angle=270]{f5a.ps}

1371: \includegraphics[width=8.0cm,angle=270]{f5b.ps}

1372: \end{center}

1373: \figcaption{

1374: (a): Calculated total gravitating mass profile (solid line) and the 90\% errors (dotted lines), 

1375: the gas mass profile (dash-dotted line) and the stellar mass profile (dashed line).

1376: (b): Radial mass-to-light ratio distribution and the 90\% errors (dotted lines). 

1377: (c): Radial dark matter distribution and the best-fit generalized NFW models for the $>0.85$\re~regions. 

1378: The dashed line is for $\zeta=0$ and the dash-dotted line is for $\zeta=1$. Note that the models give a poor

1379: fit to the mass profile of the innermost region.

1380: (d): Darkmatter fraction (solid line) and the 90\% errors (dotted lines).

1381: \label{fig5}}

1382: \end{figure}

1383: 

1384: 

1385: \end{document}

1386: